A. Properties

A few properties of the liquid oils produced by selected flash pyrolysis and the PERC and LBL processes are listed in Table 8.13. It as apparent that there are some basic differences between the two classes of oils. The oils from the flash pyrolysis processes are quite similar, as are the oils from the PERC and LBL processes. But there are major differences in their elemental analyses and

energy values. The carbon analyses and energy values are much lower and the oxygen analyses are much higher for the pyrolytic oils than for the PERC and LBL oils. This might be expected, since a synthesis gas atmosphere is used to carry out the PERC and LBL processes. The lower yield of oil from the LBL process is apparent when compared to the oil yield from the PERC process.

This process, originally called the Pearson-BrightStar Process, was developed by BrightStar Technology, Inc. It consists of the conversion of biomass feedstocks, particularly sawdust and wood chips, by steam gasification in indirectly heated, tubular reactors to afford synthesis gas suitable for methanol production (Smith, Stokes, and Wilkes, 1993) or a medium-energy gas suitable for use in gas turbines (Menville, 1996). A 0.9- to 4.5-t/day pilot plant was operated in Mississippi with sawdust and wood chip feedstocks, but sewage sludge, other biomass feedstocks, and lignites have been tested. The process gasifies partially dried wood at 10 to 15% moisture content with injected steam at a steam-to — carbon ratio of about 2 at low pressure and high temperature to maximize synthesis gas and minimize methane formation. The process is believed by BrightStar to be the first of its kind to utilize externally heated tubular reactors through which the feedstock and steam are passed. BrightStar Synfuels Com­pany, a joint venture of BrightStar and Syn-Fuels Corp., completed construction of a commercial demonstration module in 1996 in Louisiana. This plant re­quires about 22 dry t/day of wood residue feedstock and has a net energy output of 13.2 GJ/h exclusive of the energy required for reformer firing.

Some biomass species are able to reduce C02 via the initial sugars produced in photosynthesis to higher energy hydrocarbons, most of which have terpene structures. Because the energy values of terpene hydrocarbons can be as high or higher in some cases than conventional motor fuel components, and because some of the terpenes have been used as motor fuels and chemical feedstocks, a somewhat more detailed discussion of the biochemical pathways of hydrocar­bon production in biomass is worthwhile. The terpenes are isoprene adducts having the generic formula (C5H8)„, where n is 2 or more. A large number of terpene derivatives in various states of oxidation and unsaturation (terpenoids) may also be formed. Perhaps the best-known example of natural hydrocarbon production is high-molecular-weight polyisoprene rubber having very high stereospecihcity from the cambium of the hevea rubber tree (Hevea braziliensis), a member of the Euphorbiaceae family that grows in Brazil and in other tropical climates. In contrast, the Brazilian tree, Copaifera multijuga, a member of the Caesalpiniaceae family, produces relatively pure liquid sesquiterpene hydrocar­bons (C15H24), not in the cambium but in the heartwood from pores that run vertically throughout the tree trunk (Calvin, 1983). In some members of the Euphorbiaceae family such as E. lathyris, the biosynthetic pathway leads mainly to the acyclic dihydrotriterpene squalene, Сз0Н50, which then undergoes inter­nal cyclization to form C30 terpenoid alcohols and sterols. In still other biomass species, lower molecular weight acyclic and alicyclic isoprene adducts are formed as monoterpenes (C10Hi6) and diterpenes (C20H48). Certain aquatic, unicellular biomass such as the green microalgae Botryococcus braunii are reported to accumulate terpene-type hydrocarbon liquids within the cells, sometimes in large amounts depending on the growth conditions.

The major steps in the mechanisms of terpene and polyisoprene formation in plants and trees are known, and this knowledge should help improve the natural production of terpene hydrocarbons (Fig. 3.4). Mevalonic acid (1), a key intermediate derived from plant sugars via acetylcoenzyme A, is succes-

-co2

-h2o

■ C10Monoterpenes

CH,

V.

+ 111.

-OPP

FIGURE 3.4 Biochemical pathways to terpenes.

sively transformed into 5-diphosphomevalonic acid (II) and the five-carbon intermediate isopentenylpyrophosphate (III) via phosphorylation, dehydra­tion, and decarboxylation. Isomerization of a portion of III then occurs to form dimethylallylpyrophosphate (IV). The isomers III and IV combine by head-to-tail condensation to form another allylic pyrophosphate containing 10 carbon atoms (V), which can be converted to monoterpenes. Inorganic phosphate is released in the process. Continuation of this process leads to all other terpenes. The chain can successively build up by five-carbon units to yield sesquiterpenes and diterpenes containing 15 and 20 carbon atoms via VI and VII by additional head-to-tail condensations. The same structural unit

is inserted in each step. A single enzyme, farnesyl pyrophosphate cyclase, is reported to be involved in the cyclization of the farnesyl units in C. multijuga to yield many sesquiterpenes (Calvin, 1983). Alternatively, tail-to-tail conden­sation of two C15 farnesyl units can yield the 30-carbon compound squalene followed by a large number of cyclizations and rearrangements to yield an array of natural triterpenoids, as already mentioned. Similar condensation of two C20 units yields phytoene, a precursor of carotenoids. This information is expected to help in the development of genetic engineering methods to control the hydrocarbon structures and yields.

The energy potential of municipal biosolids is small. At an average higher heating value of 19 MJ/dry kg (Table 3.3), the energy content of all the primary and treated biosolids produced in 1995 in the United States can be estimated to be 0.163 and 0.113 EJ/year, both of which are much less than the energy potential of MSW. This relationship is a permanent one because of the nature of MSW and biosolids generation. Nevertheless, several combined treatment — energy recovery processes alluded to earlier for extracting energy from biosolids are in commercial use in many wastewater treatment facilities. The recovered energy is generally utilized on site as a captive fuel for the facility.

Landfilling of MSW is the preferred disposal method. But the shortage of suitable sites and the regulations and controls now applicable to the construc­tion and operation of new landfills, the operation of existing landfills, and the closure methods and subsequent monitoring requirements have led to renewed interest in incineration. MSW disposal by open-air burning and incineration in small — and large-scale facilities without energy recovery has been practiced for many years. Some small — and many large-scale MSW incineration systems now incorporate energy recovery systems for steam and thermal energy. Some produce electric power as discussed below. MSW incineration plants with energy recovery span a large throughput range—about 50 to 4000 t/day of MSW

In the United States, three main technologies are used for waste-to-energy facilities: mass burn for MSW, modular mass burn, and RDF (Berenyi, 1995). Many of the modern plants in operation are based on European combustion hardware and utilize a waste-heat boiler or a waterwall system to produce steam. In mass burn technology, which is used in the majority of facilities, MSW is combusted as received or with minimum processing to reduce the size of the pieces and clumps present in the mixture and to separate some of the material. At most locations, large appliances, car batteries, and hazardous materials are removed at the tipping floor. Most mass burn plants use waterwall incineration technologies; some use refractory-lined furnaces, rotary combus­tors, and a few other configurations. Modular mass burn facilities often use one or more small-scale combustion units to process smaller quantities of MSW than the waterwall systems. Steam is commonly generated from the hot flue gases in many modular plants using a two-chamber furnace design. Final combustion occurs in the second chamber. For plants based on RDF technolo­gies, MSW is first shredded and then separated into the combustible fraction or RDF, and selected recyclables, such as the ferrous fraction, aluminum, and glass. RDF is usually burned in semisuspension or suspension-fired furnaces or cofired with other fuels. It is also cofired in minor amounts with coal to produce steam in some of the larger power plants or with dewatered municipal biosolids in some plants. Separation of the RDF and recyclables is accomplished with various combinations of magnets, eddy-current devices, air classifiers, trommel screens, rotary drums, flotation devices, and pulping devices (see Chapter 6). Some processes involve the production of powdered RDF or pellet­ized material for use as fuels.

Numerous industrial solid wastes are disposed of in incinerators that have energy recovery capability. Most of these systems are smaller than MSW incin­erators. The compositions of specific industrial wastes are more uniform than those of MSW, but the range of waste categories is so broad that special hardware and furnaces must sometimes be used. Rotary kilns, multihearth furnaces, and fluidized-bed incinerators have been employed for industrial waste incineration systems.

This process is a two-stage, entrained-flow, slagging process for conversion of lignite and subbituminous coal. The preheated water slurry of coal is fed to the first stage where it is mixed with oxygen, the feed rate of which is controlled to maintain the reactor temperature in a specific range, depending on the properties of the coal. The second stage reduces the temperature of the raw product gas to about 1000°C. The coal is almost completely converted to carbon monoxide, carbon dioxide, and hydrogen.

IGT’s U-GAS Process

Tampella Corporation is commercializing the U-GAS process, which was devel­oped by the Institute of Gas Technology. Tampella has constructed a 10-MW, integrated U-GAS-combined cycle power plant in Finland that uses coal, peat, and wood wastes as feedstocks. U-GAS incorporates a single-stage, fluid-bed gasifier in which coal reacts with steam and air at 950 to 1090°C at pressures from atmospheric to 3.55 MPa to yield a low-energy gas. Oxygen can be substituted for air, in which case a medium-energy gas is produced.

It was not too many years ago that humans’ basic survival depended in whole or in part on the availability of biomass as a source of foodstuffs for human and animal consumption, of building materials, and of energy for heating and cooking. Not much has changed in this regard in the Third World countries since preindustrial times. But industrial societies have modified and added to this list of necessities, particularly to the energy category. Biomass is now a minor source of energy and fuels in industrialized countries. It has been replaced by coal, petroleum crude oil, and natural gas, which have become the raw materials of choice for the manufacture and production of a host of derived products and energy as heat, steam, and electric power, as well as solid, liquid, and gaseous fuels. The fossil fuel era has indeed had a large impact on civilization and industrial development. But since the reserves of fossil fuels are depleted as they are consumed, and environmental issues, mainly those concerned with air quality problems, are perceived by many scientists to be directly related to fossil fuel consumption, biomass is expected to exhibit increasing usage as an energy resource and feedstock for the produc­tion of organic fuels and commodity chemicals. Biomass is one of the few renewable, indigenous, widely dispersed, natural resources that can be utilized to reduce both the amount of fossil fuels burned and several greenhouse gases emitted by or formed during fossil fuel combustion processes. Carbon dioxide, for example, is one of the primary products of fossil fuel combustion and is a greenhouse gas that is widely believed to be associated with global warming. It is removed from the atmosphere via carbon fixation by photosynthesis of the fixed carbon in biomass.

Precipitation as rain, or in the form of snow, sleet, or hail, depending on atmospheric temperature and other conditions, is governed by movement of air and is generally abundant wherever air currents are predominately upward. The greatest precipitation should therefore occur near the equator. The average annual precipitation in the continental United States is shown in Fig. 4.2 (Visher, 1954); Table 4.3 (U. S. Dept, of Commerce, 1995) is a summary of the average monthly and annual precipitation at different locations in the United States. The average annual rainfall is about 79 cm.

The moisture needs of aquatic biomass are presumably met in full because growth occurs in liquid water, but the growth of terrestrial biomass is often water-limited. The annual requirements for good growth have been found for many biomass species to be in the range 50 to 76 cm (Roller et al, 1975). Some crops, such as wheat, exhibit good growth with much less water, but they are in the minority. Without irrigation, water is supplied during the growing season by the water in the soil at the beginning of the season and by

rainfall. Figure 4.3 illustrates the normal precipitation recorded in the continen­tal United States during the growing season, April to September (Visher, 1954). This type of information and the established requirements for the growth of terrestrial biomass can be used to divide the United States into precipitation regions as shown in Fig. 4.4 (Visher, 1954). The regions that are more produc­tive for biomass generally correlate with the precipitation regions, as shown in Figs. 4.5 and 4.6 (Visher, 1954). It should be realized, though, that rainfall alone is not quantitatively related to productivity of terrestrial biomass because of the differences in soil characteristics, water evaporation rates, and infiltra­tion. Also, as suggested in Fig. 4.5, certain areas that have low precipitation can be made productive through irrigation. Some areas of the country that vary widely in precipitation as a function of time, such as many western states, will produce moderate biomass yields and often sufficient yields of cash crops without irrigation to justify commercial production.

The transpiration of water to the atmosphere through biomass stomata is proportional to the vapor pressure difference between the atmosphere and the saturated vapor pressure inside the leaves. Transpiration is obviously affected by atmospheric temperature and humidity. The internal water is essential for biomass growth. The efficiency of utilizing this water (water-use efficiency, WUE) has been defined as the ratio of biomass accumulation to the water consumed, expressed as transpiration or total water input to the system. Analy­sis of the transpiration phenomenon and the possibilities for manipulation

“U. S. Dept, of Commerce (1995).

of WUE have led some researchers to conclude that biomass production is inextricably linked to biomass transpiration. Agronomic methods that mini­mize surface runoff and soil evaporation, and biochemical alterations that reduce transpiration in C3 plants, have the potential to increase WUE. But for water-limited regions, the fact remains that without additional water, the research results indicate that these areas cannot be expected to become regions of high biomass yields (Sinclair, Tanner, and Bennett, 1984). Irrigation and full exploitation of humid climates are of highest priority in attempting to increase biomass yields in these areas.

A. Temperature

Most biomass species grow well in the United States at temperatures between 15.6 and 32.3°C (60 and 95°F). Typical examples are corn, kenaf, and napier grass. Tropical grasses and certain warm-season biomass have optimum growth

temperatures in the range 35 to 40°C (95 to 104°F), but the minimum growth temperature is still near 15°C (Ludlow and Wilson, 1970). Cool-weather bio­mass such as wheat may show favorable growth below 15°C, and certain marine biomass such as the giant brown kelp only survive in water at temperatures below 20 to 22°C (North, 1971). The average number of days per year in the continental United States where the temperature is less than 6.1°C (43°F) and essentially no biomass growth occurs are shown in Fig. 4.7 (Visher, 1954). Table 4.4 is a summary of average monthly and annual temperature fluctuations with time and location in the United States (U. S. Dept, of Commerce, 1995). The growing season is clearly longer in the southern portion of the country. In some areas such as Hawaii, the Gulf states, southern California, and the southeastern Atlantic states, the temperature is usually conducive to biomass growth most of the year.

The effect of temperature fluctuations on net C02 uptake is illustrated by the curves in Fig. 4.8 (El-Sharkawy and Hesketh, 1964). As the temperature increases, net photosynthesis increases for cotton and sorghum to a maximum value and then rapidly declines. Ideally, the biomass species grown in an area should have a maximum rate of net photosynthesis as close as possible to the average temperature during the growing season in that area.

It is normally not necessary to reduce the water content of high-moisture — content or wet biomass feedstocks for microbial conversion processes. This contrasts with thermal conversion processes such as combustion. Dry biomass burns at higher temperatures and thermal efficiencies than wet biomass. For example, the flame temperatures of green wood containing 50 wt % moisture and dry wood in conventional combustors that supply boiler heat are about 980°C and 1260 to 1370°C, respectively (cf. FBT, Inc., 1994). Flame tempera­ture is directly related to the amount of heat necessary to evaporate the moisture contained in the wood—the lower the moisture content, the lower the amount of energy needed to remove the water and the higher the boiler efficiency. Although flame temperature is not the actual bed temperature in advanced — design, fluid-bed combustors, the effects on temperature and efficiency are the same. The maximum amounts of acceptable moisture in wood fuels for conventional furnace systems are illustrated in Table 6.1, and the typical moisture contents and heating values of several biomass fuels for combustion in commercial fluid-bed, grate, and suspension firing units are presented in Table 6.2. With the exception of suspension firing units, for which the moisture content of the fuel is usually in the 20-wt % range, the maximum moisture content range is 55 to 65 wt %. Indeed, combustion of biomass containing 65 wt % moisture in conventional grate-type systems can result in lowering of the adiabatic flame temperature to the point where self-sustained combustion does not occur.

Many of the large-scale biomass combustion systems for producing heat, hot water, or steam accept biomass fuels containing relatively large amounts of moisture and are operated without much apparent concern for the effects of moisture content of the fuel on the combustion process itself. One of the

largest biomass-fueled power plants equipped with traveling grates operates very well with wood chips containing an average of 50 wt % moisture, although a few initial handling and storage problems caused by high-moisture fuel supplies had to be solved (Tewksbury, 1987). Another power plant equipped with traveling grates operates very well on fuel containing about the same amount of moisture and consisting of a mixture of about 80% hogged mill wastes and 20% wood chips (Ganotis, 1988). Some air drying of stringy bark fuels is needed in the spring to eliminate fuel handling problems. Still another power plant equipped with a fluidized, bubbling-bed combustor operates well with a mixture of about 40% whole tree chips, 20% sawmill residues, and 40% agricultural residue from almond orchards. The fluid-bed combustors are designed to operate with fuels having a variable moisture content up to about 50 wt % (Normoyle and Gershengoren, 1989).

Thus, biomass fuels containing up to about 50 to 55 wt % moisture do not require pre-drying for acceptable performance in combustion systems that are designed for such fuels. However, a moisture content of 15 wt % is generally recognized as optimum for efficient thermochemical gasification of biomass (Miles, 1984), although several thermochemical gasification processes satisfac­torily convert feedstocks containing up to about 30 to 35 wt % moisture. Also, in certain types of thermochemical gasification processes such as steam gasification, water is a reactant and the contained water in the feedstock can be beneficial (Chapter 8).

To illustrate more quantitatively the effect of moisture content on the performance of a thermochemical process, consider the direct combustion of sugarcane bagasse in a conventional boiler to raise steam and the effects on boiler efficiency of bagasse moisture content relative to the other sources of efficiency losses. The results of a complex series of calculations to examine boiler losses and efficiencies are shown in Table 6.3 (Institute for Energy Studies, 1977). A typical amount of excess air used in bagasse-fired boilers was chosen as 30%, the moisture content of the bagasse ranged from 0 to 60 wt % in 10-wt % increments, and the stack gas temperatures ranged from 177°C (450 K) to 260°C (533 K) in approximately 28°C increments. Boiler efficiency is 100 minus the sum of the boiler losses in percentage units. The boiler losses, other than those caused by the moisture content of the bagasse, include those due to dry gases of combustion, which refers to nonuseful heat losses; those due to moisture in air; those due to the moisture formed on bagasse combustion; and other losses. Plots of bagasse moisture content against boiler efficiency and the losses due to bagasse moisture content alone at stack gas temperatures of 450 К are shown in Fig. 6.1. The analysis shows that when the moisture content of the bagasse is more than about 35 wt %, it has a greater impact on boiler efficiency losses than the other moisture sources. Note that the incremental improvement in boiler efficiency for drying the bagasse from about 35 to 0.0 wt % moisture increases boiler efficiencies by only about 7%. It was concluded by the analysts who performed the study that sugar mills can increase boiler efficiency about 5% by drying the bagasse from its typical level of 48 wt % moisture to about 35 wt % moisture. Bagasse moisture reductions by solar drying and drying with stack gases were suggested as low-cost approaches to increasing boiler efficiencies. It was also concluded that the higher temperature generated using dryer bagasse can increase heat transfer efficiencies in addition to reducing stack gas losses. This efficiency increase results from both higher temperature differentials and lower furnace gas velocities. Historically, the drying of bagasse to improve boiler efficiencies was proposed in the early 1900s. The same principles apply generally to improving the efficiencies of biomass combustion processes.

Predrying of biomass has sometimes been justified in the past only for large — scale operations, or where low-cost energy is available as waste heat. It is important to realize, however, that the absence of any capability to predry feedstock for thermochemical conversion has sometimes caused severe operat­ing problems, particularly for gasification processes. In one of the early fluid — bed gasification plants fueled with wood chips and sawdust to produce low — energy gas as an on-site boiler fuel, it was very difficult to control combustion. The industrial gas burners installed in the plant did not function satisfactorily

TABLE 6.3 Boiler Efficiency and Losses Due to Moisture on Bagasse Combustion” Sources and amount of boiler losses Bagasse moisture (wt %) Stack gas temperature (°С [°F]) Combustion dry gases (%) Moisture in air (%) Moisture in bagasse (%) Moisture of combustion (%) Other losses (%) Boiler efficiency (%) 0 177 [350] 5.86 0.15 0.00 8.07 5.00 80.92 204 [400] 6.95 0.18 0.00 8.23 5.00 79.64 232 [450] 8.04 0.21 0.00 8.39 5.00 78.36 260 [5001 9.12 0.24 0.00 8.55 5.00 77.09 10 177 [350] 5.86 0.15 1.64 8.19 5.00 79.16 204 [400] 6.94 0.18 1.67 8.35 5.00 77.86 232 [450] 8.03 0.21 1.70 8.51 5.00 76.55 260 [500] 9.11 0.24 1.73 8.67 5.00 75.25 20 177 [350] 5.88 0.15 3.51 8.30 5.00 77.16 204 [400] 6.96 0.18 3.58 8.46 5.00 75.82 232 [450] 8.05 0.21 3.65 8.62 5.00 74.47 260 [500] 9.14 0.24 3.72 8.78 5.00 73.12 30 177 [350] 5.86 0.15 6.55 8.42 5.00 74.02 204 [400] 6.94 0.18 6.68 8.59 5.00 72.61 232 [450] 8.03 0.21 6.81 8.76 5.00 71.19 260 [500] 9.11 0.24 6.94 8.92 5.00 69.79

“Adapted from Institute for Energy Studies (1977). Various assumptions are used for this analysis. The stack gases exhibit perfect gas behavior with constant specific heat equal to 0.558 kj/kg. Superheated steam has a specific heat of 1.093 kj/kg and is constant with temperature. The fuel inlet air temperature is 26.7°C (80°F) and the relative humidity of the air is 60% at 101.3 kPa. The combustion efficiency is 98%. The unburned bagasse fraction is 0.03. The ash sediment fraction is 0.015 in bone-dry bagasse. The psychrometry of the stack gases is not significantly different than that of air because the partial pressure of C02 in the stack gases is small. The calculated value is 15.86 kPa (2.3 psia) and is approximately constant for constant excess air, independent of the bagasse moisture content. “Other losses” include losses due to unburned combustibles (about 3%), losses due to radiation (about 0.5%), and unaccounted — for losses (about 1.5%).

EFFICIENCY (%) LOSSES (%)

90 25

і

with the product gas (Bircher, 1982). These problems were attributed to large variations in the quality of the gas caused by accepting wood feedstock at any moisture content up to 50 wt %, which in turn resulted in large swings in gas heating values from about 3 to 8 MJ/m3 (about 80 to 200 Btu/ft3). Drying of the feedstock has been found to be extremely important in wood gasification because it is only through the availability of a uniform feedstock that consistent gas quality can be assured (Miller, 1987).

General

Whatever the actual mechanisms of biomass pyrolysis, many reactions take place and many products are formed. The older biomass pyrolysis processes were carried out in the batch mode over long periods of time for char produc­tion. As the technology developed, other processes were designed to operate in the batch or continuous modes over shorter residence times at moderately higher temperatures. Depending on the pyrolysis temperature, the char fraction contains inorganic materials ashed to varying degrees, any unconverted organic solids, and carbonaceous residues produced on thermal decomposition of the organic components. The liquid fraction is a complex mixture of water and organic chemicals having lower average molecular weights than the feedstock components. For highly cellulosic biomass feedstocks, the liquid fraction usu­ally contains acids, alcohols, aldehydes, ketones, esters, heterocyclic deriva­tives, and phenolic compounds. The tars contain native resins, intermediate carbohydrates, phenols, aromatics, aldehydes, their condensation products, and other derivatives. The pyrolysis gas is a low — to medium-energy gas having a heating value of about 3.9 to 15.7 MJ/m3 (n) (100 to 400 Btu/SCF). It contains carbon dioxide, carbon monoxide, methane, hydrogen, ethane, ethylene, minor amounts of higher gaseous organics, and water vapor.

It is apparent that if one wishes to obtain pure chemicals by biomass pyrolysis, further processing to separate the reaction mixture is necessary. As will be shown later, this did not hinder commercial use of biomass pyrolysis for the manufacture of specific chemicals. The slow, destructive distillation of biomass was commercial technology for the production of several commodity chemicals long before fossil fuels became the preferred feedstocks. Hardwood pyrolysis once served as an important commercial source of methanol, acetic acid, ketones, and other chemicals.