Processes categorized as fast pyrolysis systems are continuously operated at temperatures generally in the range of 400 to 650°C and residence times of a few seconds to a fraction of a second. Manipulation of these parameters permits the bulk product yields to be changed from those of conventional pyrolysis systems within a wide range, but the products are still chars, liquids, and gases plus water. Fast pyrolysis is characterized by high heating rates and rapid quenching of the liquid products to terminate additional conversion of the products downstream of the pyrolysis reactor. The selectivity for specific chemi­cals is usually low, as in the case of conventional pyrolysis. Very rapid heating of biomass results in the fragmentation of the polymeric components in biomass to afford 60 to 70 wt % primary vapor products composed of oxygenated monomers and polymer fragments (Diebold et ah, 1987). Rapid, efficient quenching of the product streams and short residence times tend to “freeze the product compositions so that they correspond more closely with the chemicals formed initially on biomass pyrolysis. More details are presented in Section III on fast pyrolysis.

A. Fixed Carbon

The carbonaceous residues from biomass pyrolysis are in the charcoal fraction. These residues are called “fixed carbon” by most energy specialists. The gener­ally accepted definition of fixed carbon was originally promulgated by coal chemists. It is the amount of combustible material remaining in a sample of coal, coke, or bituminous material after removal of moisture, volatile matter, and ash, and is expressed as a percentage of the original material. American Society for Testing and Materials (ASTM) procedures have been developed for determination of each of these parameters: moisture (ASTM D 3173; 104- 110°C for 1 h), volatile matter (ASTM D 3175; 950°C for 7 min), and ash (ASTM D 3174; gradual heating to redness and finishing ignition at 750°C). Fixed carbon is the difference between 100 and the sum of these determinations (ASTM D 3172) and is essentially the elemental carbon in the original coal sample plus the carbonaceous residue formed on heating the coal sample at 950°C for 7 min. As the temperature rises above 300°C, coals emit volatile matter that consists of gases, oils, and tars. The residues contain elemental carbon, some of the higher molecular weight polynuclear aromatic hydrocar­bons formed in the process, and a few other high-molecular-weight compounds that are also formed in the process. Peat, which is derived from biomass, is categorized by some specialists as “young coal” and does contain some elemen­tal carbon. In contrast, all carbon in biomass is fixed carbon in the same sense that organic nitrogen is fixed nitrogen, but elemental carbon is not present in biomass. The photosynthetic fixation of C02 results in the formation of fixed carbon. But since there is no elemental carbon as such in biomass, its fixed carbon content can also be considered to be zero. In other words, the terminol­ogy “fixed carbon” in biomass is a misnomer.

If biomass is subjected to the ASTM D 3172 procedure for determination of fixed carbon, chemical transformation of a portion of the organic carbon in biomass into carbonaceous material occurs as described here. All of the fixed carbon determined by the ASTM procedure is therefore generated by the analytical method. Furthermore, the amount of fixed carbon generated depends on the heating rate used to reach biomass pyrolysis temperatures and the time the sample is subjected to these temperatures. Nevertheless, such analyses are valuable for the development of thermal conversion processes for biomass feedstocks. But application of the ASTM procedures to biomass might more properly be called a method for determination of pyrolytic carbon or coking yields. In the petroleum industry, the Conradson carbon (ASTM D 189, differ­ential heating with a gas burner for total of 30 min to final temperature of cherry-red crucible) and the Ramsbottom carbon (ASTM D 524, 549°C for 20 min) procedures are used to determine the coking tendency on pyrolysis of petroleum products. Use of these procedures with biomass would be ex­pected to give somewhat different results for fixed carbon than ASTM D 3172.

Table 8.6 is a tabulation of the fixed carbon, volatile matter, and ash analyses of selected biomass species, biomass derivatives, and coals as determined by the ASTM D 3172 procedure. The data for the wood species, wood barks, and herbaceous biomass species show that significant quantities of pyrolytic carbon are produced by this method. The pyrolytic chars listed, which already contain substantial amounts of elemental carbon because of the nature of the pyrolysis process, contain more fixed carbon than the coal samples listed in this table. In contrast, MSW and the papers in MSW, which are high in celluloses, contain considerably less fixed carbon suggesting that the lignins in biomass contribute more to production of fixed carbon than the other components. This is expected because of the nature of the chemical structures of the lignins, and the fact that papers are low in lignins. The times and temperatures used for the ASTM D 3172 procedure coupled with the data in Table 8.6 suggest that for maximum yields of noncarbonaceous products to be obtained on biomass pyrolysis, short reaction times should be used at relatively low pyrolysis temperatures. These conditions would be expected to yield smaller amounts of charcoal, tars, and gases, and larger amounts of liquid products. As will be discussed later, opti­mum conditions have been developed for separately maximizing charcoal and liquid product yields in biomass pyrolysis.

Because of the amounts of sample, labor, and time required to perform the ASTM D 3172 procedure, it is recommended that thermogravimetry (TG) and differential thermogravimetry (DTG) be used for moisture and proximate analysis of biomass and the rapid estimation of their thermal conversion charac­teristics. Application of these techniques shows that the proximate analyses of standard coal samples agree closely or match the values obtained with the ASTM procedure (с/. Kumar and Pratt, 1996). Thermogravimetric procedures are used to determine the thermal stabilities and properties of inorganic and organic materials and can be carried out with small samples in laboratory equipment. The results are reasonably accurate and reproducible. TG and DTG employ sensitive thermobalances and automated data processors to measure weight loss and the rate of weight loss of the samples as a function of tempera­ture, respectively. A TG curve (thermogram) records the weight of the sample with time at a preset temperature or a programmed heating rate in an inert or reactive gaseous atmosphere. Some laboratory instrumentation has also been designed to operate at elevated pressures (cf. Johnson, 1979). Differentiated TG data with time or temperature provides the rate of weight loss. A TG curve is used for moisture and proximate analyses, and a DTG curve can be used to

“Adapted from Graboski and Bain (1979) and Jenkins and Ebeling (1985). The data on barks and pyrolytic chars are from Howlett and Gamache (1977). The data on MSW are from Mass and Ghosh (1973). The data on coals are from Bituminous Coal Research (1974). The data on herbaceous biomass are from Jenkins and Ebeling (1985). The data on woods are from Howlett and Gamache (1977) and Jenkins and Ebeling (1985).

examine combustion, pyrolysis, and gasification characteristics. The kinetics of conversion and the conditions for process optimization can also be estimated using TG and DTG. Analysis of selected biomass components by these tech­niques indicates that pyrolysis is initiated at 150-350° C for hemicelluloses, 275-350°C for celluloses, 250-500°C for lignins, 500-620°C for latex, and 550- 900°C for high-molecular-weight resins and oils (cf. Kumar and Pratt, 1996). The combustion of fixed carbon and ash in representative samples of biomass occurs at 900°C according to these studies.

The data presented in Table 8.7 illustrate the pyrolysis characteristics ob­tained by TG and DTG analysis of several biomass species and parts. This type of data and derived data can be used to project the utility of a given biomass feedstock for thermal conversion. For example, the organic material emitted in the temperature range 320 to 500°C was assumed to be potential tar-forming volatiles (Grover, 1989). Those biomass species having lower emissions of volatiles in this temperature range were judged to be more suitable for conver­sion to tar-free gases on pyrolysis.