Residential biomass fuels are usually chunks and pieces of wood and logs that are burned in small woodstoves and fireplaces, in contrast to medium — and large-scale municipal, industrial, commercial, and power-generating facilities that burn a large variety of virgin and waste biomass fuels such as MSW, RDF, sawdust, bagasse, rice hulls, wood chips, and industry-specific wastes. The residential wood fuels range from about 15 to 50% moisture content depending on the extent of air drying. The moisture content of seasoned firewood is typically about 20%. Most of the older woodstoves consist of conventional downdraft, updraft, or crossdraft fireboxes with fixed steel grates. Airflow is adjusted manually and the flue connections are sized for maximum loadings. Electric air blowers are sometimes used in the larger wood-burning appliances. The ash collects in a compartment below the grate. A common characteristic of the older woodstoves is that they permit long fuel residence times in the firebox to maximize fuel usage. Over the years, hundreds of woodstove mod­els of different designs have been marketed, many of which are claimed to have superior performance. Some of the modifications did effect small in­creases in thermal efficiency, but many were stricdy aesthetic changes.

With few exceptions, it was not until the 1980s when major advances were made in woodstoves in response to government mandates to reduce pollution.

Catalytic woodstoves with secondary combustion chambers are good exam­ples of the application of modem technology to improve operating efficien­cies. These appliances are at the high end of the efficiency scale as compared to noncatalytic woodstoves; fireplaces are at the low end. A comparison of the operating efficiencies of woodstoves is shown in Table 7.3 (Long and Weaver, 1985). The woodstove efficiency in this comparison corresponds to the usable heat over the energy content of the fuel input unadjusted for moisture content. The efficiencies range from a low of 13% for conventional fireplaces to 75% for airtight, catalytic woodstoves with a secondary combustion chamber.

As discussed in Chapter 6, pellet-burning stoves for residential use, or pellet stoves as they are generally called, and pellet fuels made from wood, wood wastes, straws, RDF, waste paper, and other waste biomass have been commer­cially available for several years. Residential pellet standards have been pro­posed by the Pellet Fuels Institute in the United States (Table 6.5), where the annual market has averaged about 35,000 pellet stoves over a 10-year period (Pickering, 1995, 1996). Pellet stoves are marketed in both free-standing and fireplace-insert models. These stoves are equipped with hoppers which hold about 20 kg or more of pellets that are auger-fed, usually from the top, into the combustion chamber. The advanced design units employ forced air flow past the pellets, and passage of the hot combustion products through a heat

TABLE 7.3 Comparison of Woodstove Efficiencies”

Woodstove type Efficiency (%)

75 An airtight stove with a catalytic afterburner and a

smoke chamber such as a double-drum stove or a design that provides for improved burning in a secondary combustion chamber.

63 Any of the openable stoves with tight fitting doors

that would not fall into the airtight catalytic category.

50 Non-airtight stoves with flue connections smaller

than 20.3 cm (8 in.).

38 Non-airtight stove with a 20.3-cm flue connection.

25 Equipped with heatalator or similar device to

improve heating effect.

13 Without heatalator.

“Long and Weaver (1985). The efficiency does not include a correction for moisture content. When adjustments are made for moisture, the efficiency of the average airtight woodstove or fireplace for wood fuel with 20% moisture, for example, would be 50% rather than 63%, or 10% rather than 13%.

exchanger to heat circulating room air, and then through a simple flue to the outside. The advanced stoves are relatively complex and require several mo­tors, fans, and electronic modules to control fuel and combustion air.

Small furnaces and boilers for wood fuels with and without backup oil or natural gas have been designed for burning logs, wood chips, sawdust, and pelletized wood wastes for central space heating in northern climates (с/. Brandon, 1981; Sahrman, 1983). The loading systems are gravity-fed hoppers or screw-fed hoppers (stokers) for chips, pellets, and sawdust. Logs and split logs are loaded manually. When thermal energy storage is employed, provisions are generally made for hot water storage and circulation through radiators when heat is required. One example of the controls for a conventional furnace and boiler is those used for one type of manually loaded round wood system (Brandon, 1981). Regulation of the heat output is achieved by varying the amount of combustion air available. Demand for heat from the building thermo­stat activates a control motor that opens a primary air vent to the furnace; when no heat is required, the vent closes and the fire dies down. If the wood fire is not controlled by the primary air vents, the heat is dumped to the building by switching fans or water circulators. This ensures safe operation by preventing overheating of the furnace or the boiler. Automated feeding systems for particulate fuels are often controlled so that both combustion air and fuel feeding are adjusted with the demand for heat. Controls for advanced space-heating units usually take into account the fact that wood combustion cannot be controlled by instantaneous on-off devices, as an oil or gas burner can. Combustion can be sustained in a low-level, standby mode, or a fast-start device such as an auxiliary oil burner can be employed to renew combustion when heat is needed. There are many variations in designs and controls for these systems. In comparative tests of 10 commercially available units for residential use, the overall efficiencies in terms of wood fuel input over useful output during the heating season ranged from about 40 to 50% for 9 of the 10 units evaluated (Brandon, 1981). This is quite high for small systems.

This process can be operated on all types of coal without pretreatment. Dried, pulverized coal and oxygen are converted in a horizontal, entrained-flow gas­ifier at about 1820°C and near-atmospheric pressure. The raw gas is quenched with water to solidify entrained molten ash, scrubbed to remove entrained solids, and purified to remove hydrogen sulfide and a controlled quantity of carbon dioxide. The resulting product is used as synthesis gas.

Shell Oil Co. Process

This process is a dry feed, entrained-flow, high temperature-high pressure, slagging gasifier that converts a wide variety of coals from lignite to bituminous to a medium-energy gas for combined cycle power generation. The unit oper­ates with pressurized, predried coal, oxygen, and steam at 1500°C and attains carbon conversions above 99%.

Texaco, Inc. Process

This process is a single-stage, pressurized, entrained-flow slagging process that uses a water slurry of ground coal which is pumped along with oxygen to the gasifier. The operating temperature in the gasifier is 1200 to 1500°C. Careful control of the oxygen feed to maintain a reducing atmosphere results in a synthesis gas that is predominantly carbon monoxide and hydrogen.

I. INTRODUCTION

It is well known that developed or industrialized nations consume more energy per capita than developing or Third World countries, and that there is a correlation between a country’s living standards and energy consumption. In general, the higher the per-capita energy consumption, the higher the living standard. However, the rapid worldwide increase in the consumption of fossil fuels in the twentieth century to meet energy demand, mostly by industrialized nations, suggests that the time is not too distant before depletion begins to adversely affect petroleum and natural gas reserves. This is expected to result in increased usage of alternative biomass energy resources.

The potentially damaging environmental effect of continued fossil fuel usage is another factor that will affect biomass energy usage. It has not been estab­lished with certainty that on a global basis, there is a specific relationship between fossil fuel consumption and environmental quality. There is also considerable disagreement as to whether increased fossil fuel consumption is the primary cause of global climate change. But niost energy and environmental specialists agree that there is a strong relationship between localized and

l

regional air quality in terms of pollutant concentrations and fossil fuel con­sumption. The greater the consumption of fossil fuels, especially by motor vehicles and power plants, the greater the levels of air pollution in a given region.

These issues are briefly examined in this chapter to provide a starting point and a foundation for development of the primary subject of this book—energy and fuels from virgin and waste biomass. Special emphasis is given to the United States because it utilizes about one quarter of the energy consumed in the world.

The intensity of the incident solar radiation at the earth’s surface is one of the key factors in photosynthesis, as shown in Chapter 3. Except in a few rare cases, nat­ural biomass growth will not occur without solar energy. Isolation varies with ge­ographic location, time of day, and season of the year, and as is well known, it is high in the tropics and near the equator. The approximate changes of insolation with latitude are illustrated in Table 4.1 (Brinkworth, 1973), and Fig. 4.1 shows how the mean annual insolation varies at the earth’s surface with geographic location (Crutchfield, 1974). A more quantitative summary of average total daily insolation values over the continental United States is shown in Table 4.2 (U. S. Dept, of Commerce, 1970). At a given latitude, the incident radiation is not constant and often exhibits large changes over relatively short distances. Although several environmental factors influence biomass productivity, there is usually a relatively good correlation between the annual yields of dry biomass per unit area and the average insolation value (see Table 3.1). All other factors being equal, it is generally true that the higher the insolation, the higher the annual yield of a particular energy crop provided it is adapted to the local environment. C4 biomass species often exhibit higher productivities in terms

of growth rates and annual yields because of their capability to utilize incident solar radiation at higher efficiencies for photosynthesis.

A. Fundamentals

Dewatering refers to the removal of all or part of the contained moisture from biomass as a liquid. Drying is a similar process, except that the moisture is removed as vapor. In some cases where a waste or virgin biomass feedstock is thermally processed directly for energy recovery, it may be necessary to partially dry the raw feed before conversion. Otherwise, more energy might be consumed by the conversion process than would be produced in the form of energy or fuel. Open-air solar drying is usually the lowest cost drying method, if it can be used. Raw materials that are not sufficiently stable to be dried by solar methods can be dried more rapidly using industrial dryers such as spray dryers, drum dryers, and convection ovens if cost permits. For large — scale drying applications, forced-air furnaces and drying systems designed to use hot stack gases are sometimes just as efficient.

Since a large portion of a feedstock’s equivalent energy content can be expended for drying, there is a balance between the cost of moisture removal, the incremental improvement in efficiency on conversion, and the advantages of handling drier feedstock. The key biomass property that should obviously be examined, in addition to conversion process requirements, is the moisture content of the fresh biomass, the methods available for its partial or total removal, and the effects, if any, on the properties of the remaining biomass. The moisture content of biomass is as variable as the multitude of biomass species available as potential feedstocks.

Aquatic biomass is one category of feedstock that can be classified as high — water content biomass. Freshwater, marine, and microalgal biomass, such as water hyacinth, giant brown kelp, and Chlorella, respectively, contain large amounts of intracellular moisture. The water content is usually over 95 wt % of the fresh biomass (see Appendix A, Part A.8, for the definitions of wet and dry weight percentages). Several types of municipal, industrial, and farm animal wastes are also produced in association with water and are potential feedstocks. One example is untreated municipal biosolids. Because of the nature of the collection systems (i. e., dilution with water to facilitate localized disposal and transport in municipal lines to wastewater treatment plants), raw municipal biosolids contain over 95 wt % water. They are stabilized at the wastewater treatment plant by subjecting them to various primary and secondary treat­ments which reduce volatile solids, BOD, and COD, and are dewatered for disposal. Examples of high-water-content industrial wastes are pulp and paper mill sludges and alcoholic beverage industry sludges. They are often subjected to biological stabilization and dewatering processes similar to those used for municipal biosolids. The major farm animal wastes that are potential feedstocks for conversion processes are cattle, hog, and poultry manures. The moisture contents of most of these wastes range up to 80 wt %.

Terrestrial biomass, as freshly harvested green biomass, generally contains 40 to 60 wt % moisture. These potential feedstocks include most herbaceous species, softwoods, and hardwoods. Agricultural crop residues that have been exposed to open-air solar drying contain less moisture, often in the 15-wt % range or less. Straws are good examples. Other examples consist of by-products such as corn cobs, rice hulls, and nut shells from the processing of agricultural and orchard crops. Many of these residuals have even lower moisture levels. As-received municipal solid waste (MSW) usually contains 10 to 30 wt % moisture depending upon the season of the year and geographic location where it is collected. All of these terrestrial biomass species and residuals are suitable feedstocks for one or more different thermochemical conversion processes. Microbial conversion is also feasible in many cases after suitable pretreatment. Unlike many high-water-content biomass species, most terrestrial virgin bio­mass and residuals are sufficiently stable to undergo solar or thermal drying.

Woody biomass is the largest source of standing biomass and is the preferred feedstock for large-scale, integrated, biomass production-conversion systems. So it is worthwhile to consider how moisture accumulates in these materials internally during growth and from humid air. Water is contained in all tissues of a tree, both dead and alive (Mirov, 1949). Young leaves and roots contain up to 90 wt % moisture; tree trunks contain as much as 50 wt %. In the transpiration process, water is absorbed by the roots from the soil, pushed into the sapwood, and then pulled up to the leaves above ground where it is given off to the atmosphere by evaporation through the stomata, the same openings that admit C02 (Chapter 4). Various osmotic and diffusional forces drive the water from the soil to the leaves through semipermeable membranes and capillary passages. About one-half of the solar energy falling on the leaves supplies the energy to facilitate transpiration, which is necessary for photosyn­thesis to occur. The stoichiometric photosynthesis of 100 kg of cellulose requires about 55.6 kg of water, but during the process, the tree transpires about 100,000 kg of water to the atmosphere.

Wood also absorbs moisture from humid air and is the equivalent of an elastic gel that exhibits limited swelling as water vapor is taken up from the air. Two different mechanisms are operative: adsorption and absorption (Nikitin et al, 1962). In adsorption, moisture is transferred from air to the wood surfaces and results from the attraction between polar water molecules and the negatively charged surfaces of the wood. The negative charges involve functional groups on the surface that can carry full or partial negative charges or organic molecules that can exist as dipoles with the negative ends clustered on the surface. The amount of moisture adsorbed on wood surfaces is relatively small; it ranges up to about 5 to 6 wt % of the wood at 20°C and 100% relative humidity. In absorption, water molecules are drawn into the permeable pores of the wood by spongelike processes due to diffusional and osmotic forces followed by capillary condensation. A large number of fine capillaries in the wood fibers facilitate this process. The amount of moisture absorbed within the woody structures depends upon the pore diameters and distribution of the capillaries. In spruce wood pulps, for example, the amount of water vapor absorbed at 20°C and 100% relative humidity is about 25 wt %. The maximum total amount of water taken up from air at ambient conditions by adsorption and absorption is about 30 wt % of the wood, but can reach 200 wt % or more if the wood is soaked in liquid water (Nikitin et al, 1962). Exposure to precipitation is the third mechanism that raises the moisture content of green wood to an average of about 50 wt %. It is evident from this explanation of the various mechanisms of water uptake by wood that transpiration of soil moisture is essential for tree growth to occur, that the temperature and humid­ity of the surrounding air affects the moisture content of the wood if sufficient time is allowed to reach equilibrium moisture levels, and that wood is hygro­scopic. The same principles are applicable to most terrestrial biomass because of the vascular structure typical of many species.

The pyrolysis of biomass feedstocks may be endothermic or exothermic, de­pending on the temperature of the reactants. For most biomass containing highly oxygenated hemicellulosics and cellulosics as the major components, pyrolysis is endothermic at temperatures below about 400 to 450°C and exo­thermic at higher temperatures. Once the necessary temperature has been reached in a properly designed system, little or no external heat is needed to sustain the process. The principal exothermic reactions that occur in the gas and solid phases during biomass pyrolysis are shown in Table 8.2. These reactions include the reduction of the carbon oxides to methane and methanol, the water gas shift reaction, and carbonization of celluloses. Substantial quanti­ties of hydrogen are required for reduction of carbon oxides to methane and methanol, but hydrogen is not required for the water gas shift, which produces hydrogen, and the char formation reactions listed in Table 8.2. The pyrolysis temperature should be high enough to generate the requisite hydrogen for reduction of the carbon oxides. The water formed on biomass pyrolysis and the vaporization of the physically contained moisture in the fresh feed can participate in the water gas shift reaction. Interestingly, the exothermicity of cellulose carbonization is quite high per monomeric unit (C6Hi0O5). Since the char yields on conventional biomass pyrolysis range up to about 35 wt %, and the fixed carbon contents of the chars are high, char formation would be expected to be a dominant driving force for biomass pyrolysis at the lower temperatures at which autogenous pyrolysis begins, but which generate less hydrogen. At these temperatures, pyrolysis is generally reaction — rate controlled, and at higher temperatures, the process becomes mass-transfer controlled. Energy to initiate biomass pyrolysis can be supplied by an external heat source or a portion of the pyrolysis products such as char and low — energy gas.

In the mid 1970s, Andco Incorporated developed and commercialized a slag­ging process called the Andco-Torrax System for converting MSW to low — energy gas and an inert glassy aggregate (Davidson and Lucas, 1978; Mark, 1980). Plants ranging in size from 2.4 to 7.5 t/h were installed in Europe, Japan, and the United States. Refuse is charged into the top of the gasifier without prior preparation except to shear or crush bulky items to about one meter or less in the longest dimension. As the refuse descends within the gasifier, it is dried and then pyrolyzed by the hot, oxygen-deficient gases produced in the hearth area. Char from the pyrolysis process and the noncom­bustible materials continue to descend. Primary combustion air at temperatures of about 1000°C is admitted to the gasifier immediately above the hearth to oxidize the char at temperatures sufficient to slag the inerts. The slag is continuously drained from the gasifier. The low-energy gas, the heating value of which is about 4.7 to 5.9 MJ/m3 (n), from the top of the gasifier is burned in a secondary combustion chamber at slagging temperatures. The slag is collected from this unit also. The heat from the secondary combustion chamber is used for hot water, steam, and power production. It is believed that this process can be used for waste disposal and energy recovery with combined feeds of MSW and tires, sludge, or waste plastics, and for the manufacture of cement. The first Andco-Torrax plant in the United States was built at Disney World in Orlando, Florida. This plant was used for waste paper from restaurants in the theme park. The performance of the gasifier in this plant was felt to be unsatisfactory because “arching” of the feedstock frequently occurred in the upper zone of the gasifier and resulted in feed stoppage. This problem could probably have been eliminated without design changes by densifying the very light waste paper feed.

It is important to examine the potential amounts of energy and biofuels that might be produced from biomass carbon resources and to compare these amounts with fossil fuel demands. This would make it possible to estimate the percentage of energy demand that might be satisfied by particular bio­mass types.

A. Virgin Biomass

Consider first the incident solar radiation, or insolation, that strikes the earth’s surface. At an average daily insolation worldwide of about 220 W/m2 (1676 Btu/ ft2), the annual insolation on about 0.01% of the earth’s surface is approximately equal to all the primary energy consumed by humans each year. For the United States alone, the insolation on about 0.1 to 0.2% of its total surface is equivalent to its total annual energy consumption.

The most widespread and practical process for capture of this energy as organic fuels is the growth of virgin biomass. As already discussed, extremely large quantities of carbon are fixed each year in the form of terrestrial and aquatic biomass. Using the figures in Table 2.2, the energy content of standing biomass carbon; that is, the renewable, above-ground biomass reservoir that in theory could be harvested and used as an energy resource, is about 100 times the world’s annual energy consumption. At a nominal biomass heating value of 18.6 GJ/dry t (16 X 106 Btu/dry ton) and assuming that the world’s total annual coal, oil, and natural gas consumption is about 315 EJ (1993), the solar energy trapped in 16.9 Gt of dry biomass, or about 7.6 Gt of biomass carbon, would be equivalent to the world’s consumption of these fossil fuels. Since it is estimated that about 77 Gt of carbon, or 171 Gt of dry virgin biomass equivalent, most of which is wild and not controlled by humans, is fixed on the earth each year, it is certainly in order to consider biomass as a raw material for direct use as fuel or for conversion to large supplies of substitute fossil fuels. Under controlled conditions, dedicated biomass species might be grown specifically as energy crops or for multiple uses including energy. Relatively rapid replacement of the biomass utilized can take place through regrowth.

A more realistic assessment of biomass as an energy resource can be made by calculating the average surface areas needed to produce sufficient biomass at different annual yields to meet certain percentages of fuel demand for a particular country, and then to compare these areas with those that might be made available. Such an assessment for the United States could, for example, address the potential of biomass for conversion to SNG as shown in Table 2.5. For this analysis, the annual U. S. demand for natural gas is projected to reach 26.5 EJ (25.1 quad) by 2010 at an annual growth rate in consumption of 1.2% (U. S. Dept, of Energy, 1994). It is assumed that biomass, whether it be trees, plants, grasses, algae, or water plants, has a heating value of 18.6 GJ/dry t, is grown under controlled conditions in “methane plantations” at yields of 20 and 50 dry t/ha-year, and is converted in integrated biomass planting, harvest­ing, and conversion systems to SNG at an overall thermal efficiency of 50%. These conditions of biomass production and conversion either are within

the range of present technology and agricultural practice, or are believed to be attainable in the near future. The average total plantation areas were then calculated to meet 1.42, 10, 50, and 100% of the projected U. S. demand in 2010 for natural gas. A percentage of 1.42 is equal to a daily pro­duction of 26.9 X 106 m3 at normal conditions (1 X 109 SCF) of dry SNG. The range of areas required at the low yield level is between 2,023,000 and 142,500,000 ha, or 0.2 and 14.9% of the 50-state U. S. area. At the high yield level, the areas are between 809,000 and 57,000,000 ha, or 0.08 and 6.0% of the 50-state U. S. area. To put this analysis in the proper perspective, the results are shown in graphical form in Fig. 2.2 at the two yield levels together with the percentage area of the United States needed at any selected gas demand supplied by SNG from biomass. Relatively large areas are required, but not so much as to make the use of land or freshwater biomass for energy applications impractical. When compared with the area distribution pattern of the United States (Table 2.6) (USDA Forest Service, 1989), it is seen that selected areas or combinations of areas might be utilized for biomass energy. Areas that are not used for productive purposes might be suitable, or possibly biomass for both energy and foodstuffs or energy and forest products applications can be grown simultaneously or sequentially in ways that would benefit both. Also, relatively small portions of the bordering oceans might supply the needed biomass growth areas, in which case, marine plants would be grown and har­vested.

AREA REQUIRED (million ha) 50-STATE AREA (%)

“Adapted from USDA Forest Service (1989). The data for forest, rangeland, and other land are for 1982. The data for inland water are for 1990. The data for other water are for 1970. Forest areas are at least 10% stocked by trees of any size, or formerly having such tree cover and not currently developed for nonforest use. Transition land is forest land that carries grasses or forage plants used for grazing as the predominant vegetation. Climax vegetation on rangelands is predominantly grasses, grass-like plants, forbs, and shrubs suitable for grazing and browsing. Other land areas include crop and pasture land and farmsteads, strip mines, permanent ice and snow, and land that does not fit any other land cover.

This approach to the preliminary assessment of the potential of biomass energy presumes that suitable conversion processes are available for conversion of biomass to SNG. Other processes could be used to manufacture other synfuels such as synthesis gas, alcohols, esters, and hydrocarbons. The direct route, alluded to in Fig. 2.1 as natural production of hydrocarbons, can possibly bypass the harvesting-conversion routes. As already mentioned, some biomass species produce hydrocarbons as metabolic products. Natural rubber, glycer­ides, and terpenes from selected biomass species, for example, as well as other reduced compounds could be extracted and refined to yield conventional or substitute fossil fuels.

A second source of renewable carbon is the deposits and reservoirs of essentially non-energy carbon forms—ambient C02 and the lithospheric car­bonates. The availability of such raw materials cannot be questioned, although low-cost separation and energy-efficient recovery of very small concentrations of C02 from the atmosphere present technological challenges. Another basic problem resides in the fact that all of the energy must be supplied by a sec­ond raw material, such as hydrogen. Hydrogen would have to be made avail­able in large quantities from a nonfossil source or the purpose of the synfuel system to produce renewable fuels would be defeated. Conceptually, there is no difficulty in developing such hydrogen sources. Hydrogen can be produced by water electrolysis and thermochemical and photolytic splitting of water. Electrical power and thermal energy can be supplied by nonfossil-powered nu­clear reactors, and by means of hydroelectric and wind systems, ocean thermal gradients, wave action, and solar-actuated devices. Hydrogen can also be man­ufactured from biomass and by direct action of solar energy on certain cat­alytic surfaces.

As already pointed out, about 16.9 Gt of dry biomass, or about 7.7 Gt of biomass carbon, would have approximately the same energy content as the total global consumption of coal, oil, and natural gas (in 1993). This amount of carbon corresponds to less than 1.0% of the total standing biomass carbon of the earth. Under present conditions of controlled and natural production of fixed carbon supplies, the utilization of some of this carbon for energy applications seems to be a logical end use of a renewable raw material. Forest biomass is especially interesting for these applications because of its abundance. The expansion of controlled production of virgin biomass in dedicated energy crop systems should also be considered because this would result in new additions to natural biomass carbon supplies. For example, the biomass carbon supplies in marine ecosystems might conceivably be increased under controlled conditions over the current low levels by means of marine biomass energy plantations in areas of the ocean that are dedicated to this objective. Unused croplands and federal lands might also be used for the production of herbaceous or woody biomass energy crops.

The productivity of some salt marshes is similar to that of seaweeds. Spartina altemiflora has been grown at net annual yields of about 33 dry t/ha-year, including underground material, on optimum sites (Westlake, 1963). Other emergent communities in brackish water, including mangrove swamps, appear to have annual organic productivities of up to 35 t/ha-year (Westlake, 1963), but insufficient information is available to judge their value in biomass energy systems. Freshwater swamps are believed to be highly productive and offer opportunities for energy production. Both the reed Arundo donax, and bulrush Scirpus lacustris appear to produce 57 to 59 t/ha-year yields (Westlake, 1963); if these can be sustained, they should be suitable candidates for biomass energy usage. Cattail (Typha spp.) is a wetland biomass that has been proposed as an energy resource (Pratt and Andrews, 1980). It grows naturally in monocul­tures, is highly productive, has few insect pests, and can be grown on marginal lands. Managed stands are reported to yield 25 to 30 dry t/ha-year of cattail in the northern climates of the United States (Minnesota).

A strong aquatic biomass candidate for energy applications is the water hyacinth (Eichhomia crassipes) (Klass, 1974). This biomass species is highly productive, as might be expected because it grows in warm climates and has submerged roots and aerial leaves like reedswamp plants. It has been estimated that water hyacinth could be produced at rates up to about 150 t/ha-year if the plants were grown in a good climate, the young plants always predominated, and the water surface was always completely covered (Westlake, 1963). Some evidence has been obtained to support this growth rate (McGarry, 1971; Yount and Grossman, 1970). If such yields can be maintained on a steady-state basis, water hyacinth could possibly turn out to be a prime aquatic biomass candidate as a nonfossil carbon source for synfuels manufacture as well as other potential applications such as the manufacture of paper. Water hyacinth currently has no competitive uses and is considered to be an undesirable species on inland waterways. Many attempts have been made to rid navigable streams in Florida of water hyacinth without success; the plant is a very hardy, disease-resistant species (Del Fosse, 1977).

MSW is a complex mixture of inorganic and organic materials (Table 5.1). Efficient separation and economic recovery of RDF and the components that can be recycled is the ultimate challenge to engineers who specialize in design­ing resource recovery equipment for the large-scale processing of solid wastes generated by urban communities. Unfortunately, the number of MSW plants designed to recover recyclables and RDF make up only about 20 to 25% of the total MSW-to-energy facilities in the United States. This is probably caused by the success of mass burn technologies (Chapter 7) and the fluctuating markets for recyclables. Nevertheless, the processing schemes and hardware employed to separate MSW are innovative and justify some elaboration. Liter­ally hundreds of hardware designs and machines have been developed to separate and recover most of the components in MSW. Some resource recovery facilities have even installed equipment for recovering coinage, which is just a small fraction of the total mass of MSW.

One of the first comprehensive resource recovery plants in the world was built in Dade County, Florida (Todd, 1984; Berenyi and Gould, 1988). A brief description of this facility when it was in full-scale operation to re­cover recyclables and RDF is informative. The plant was designed to pro­cess 2720 t/day (3000 ton/day) of MSW, but it frequently processed over 3630 t/day (4000 ton/day), and could process 4540 t/day (5000 ton/day) if only household garbage were received. It was designed to accept, in addition to household garbage, a wide variety of solid wastes including trash, garden clippings, trees, tires, plastics, pathological wastes, white goods (i. e., stoves, refrigerators, air conditioners, etc.), and industrial, commercial, and demolition wastes. RDF and shredded tires, approximately 1000/day, were burned for on­site power generation in a 77-MW power plant, and glass, aluminum, ferrous metals, and other materials including the ash and flyash were recovered and sold. The plant achieved a 97% volumetric reduction compared to as-received MSW. Only 6 wt % of the total incoming MSW remained as unsalable residue; this was disposed of in a landfill. The plant also conformed to all effluent, leachate, emissions, noise, and odor requirements. Impressive results such as these depended on the availability and reliability of efficient separation methods.

A simplified description of the first comprehensive materials recovery facility of its type in the United States illustrates how one plant was designed to accomplish some of these separations (Waste Management, Inc., 1977). The plant, called Recovery 1, was built in New Orleans, Louisiana, to process 590 t/day of MSW. The waste was delivered and unloaded at one of two receiving pit conveyors, and transported by conveyors to the first separation unit, a 13.7-m long by 3-m diameter rotating trommel that contained circular holes 12 cm in diameter. Plastic and paper bags tumbling in the trommel were broken open by lifters. The smaller, heavier objects such as heavy metal and glass bottles that fell through the holes were transported directly to a magnetic ferrous recovery station and an air classifier. The larger and lighter materials such as paper, textiles, and aluminum containers that passed through the trommel were conveyed to a 746-kW primary shredder. The shredded material was then conveyed to the ferrous recovery station and the air classifier. In the air classifier, a high-speed air current blows the light materials out of the top of the classifier. This fraction, RDF, consists of shredded paper, plastic, wood, yard wastes, and food wastes. The heavy fraction is essentially glass, aluminum, other nonferrous metals, and some organic material. It was routed to the recovery building for further processing. A secondary, 746-kW shredder system handled oversized, bulky wastes without passage through the trommel. The output was also conveyed to the air classifier, where RDF was obtained as the overhead, and the heavy fraction was conveyed to the recovery building. Each shredder system was sized to process 590 t of MSW in about 12 h to ensure operating reliability.

Three modules were located in the recovery building. The first module consisted of a vibrating screen to separate the shredded material by particle size, a drum magnet to separate residual ferrous material, an eddy current separator to remove the nonmagnetic aluminum and other nonferrous metals, and a small hammermill to further shred the aluminum fraction to increase its bulk density. The output from the first module consisted of the ferrous fraction, the aluminum fraction, and a fraction that contained primarily glass and some nonferrous metals. The glass fraction containing some residual nonferrous metal was conveyed to the second recovery module, which con­sisted of a crusher, another vibrating screen, a rod mill, and a two-deck, fine — mesh vibrating screen. The glass fraction was crushed and screened in the second module. The smaller fraction was treated with a pulsed water stream that separated the light fraction, which was discarded. The heavier glass frac­tions were pumped as slurries to the bottom deck of the fine-mesh second screen to separate the larger particles for crushing in the rod mill. Recycling of the milled material back to the top deck of the fine-mesh screen yielded a glass cullet fraction for further treatment in the third module, and a nonferrous metal fraction which was removed from the second screen. The third module contained a hydrocyclone, a froth flotation tank, and a glass dryer. The glass cullet fraction from the second module was mixed with clean water in a prefloat tank to remove any remaining organic particles, separated from the slurry through centrifugal separation and froth flotation, dried, and conveyed to the loadout building for shipment. RDF was recovered from the air classifier, and the ferrous, aluminum, and glass fractions were recovered from the “bottoms” of the classifier.

This is a simplified description of how MSW is separated into recyclables and fuel. There are many refinements of these operations.