Energy & Fuels 1991,5,555-561
residence time on CHI and CzH4yields indicates that the cracking reactions of the volatiles (which generate more hydrocarbon gases) begin at an earlier stage during pyrolysis as pressure increases from 100 to 309 psig, consistent with both the weight loss and the tar yield decreasing more rapidly during the early stage of pyrolysis at 309 psig than at 100 psig. The apparent first-order rate constants for devolatilization were dependent on the extent of devolatilization and on the applied pressure. The devolatilization rate decreased significantly with increasing pressure. At each pressure, rate constants decreased as pyrolysis proceeded to greater extents of devolatilization. The first-order rate constants for swelling were similarly dependent on extent of swelling and applied pressure. Swelling rates reached a maximum at 100 psig and decreased as pressure was increased further. At each applied pressure, swelling rate constants decreased as swelling proceeded. The different effects of pressure on devolatilization rate and swelling rate
555
provide evidence that two different mass-transfer mechanisms are operative, depending on applied pressure. At 100 psig or higher, the reduction in both devolatilization and swelling rates suggests that bubble transport was the main mass-transfer mechanism. At atmospheric pressure the devolatilization rate is higher but the swelling rate is lower than those observed at elevated pressures, indicating that a diffusional mechanism is predominant.
Acknowledgment. We are grateful for the financial support for this work provided by the United States Department of Energy. We thank Dr. Paul Painter for assistance in interpretation of the DRIFT spectra, and Dr. Richard Hogg for assistance in using the optical image analysis system. Mohammad Fatemi developed the computer models of particle residence time and temperature used in this study, and Carl Martin and Ronald Wincek assisted in maintaining the high-pressure entrained-flow reactor system.
Viscous Sintering of Coal Ashes. 1. Relationships of Sinter Point and Sinter Strength to Particle Size and Composition Bongjin Jungt and Harold H. Schobert* Fuel Science Program, Department of Materials Science and Engineering, The Pennsylvania State University, University Park, Pennsylvania 16802 Received December 7, 1990. Revised Manuscript Received April 3, 1991
Several tests were performed to evaluate the sintering characteristics of coal ashes prepared from four coals ranging in rank from lignite through bituminous. The ashes were prepared by standard ASTM ashing procedures and then separated into three different particle size fractions. Measurement of electrical resistance as a function of inverse absolute temperature showed that the sinter point decreased with decreasing particle size, as a result of the increased surface area to volume ratio of the ash particles. Compressive strength was studied as a function of temperature in the range 750-950 "C. For any given particle size of each coal ash, the strength of the sintered ash increased with sintering temperature. At a given sintering temperature, the strength was inversely proportional to particle size. The compressive strength test results showed qualitative agreement with the Frenkel sintering equation. X-ray diffraction of the sintered ashes showed that, as sintering temperature increased, there was a depletion of anhydrite and increase in the amount of calcium-containing aluminosilicate phases. This result suggests that chemical reactions accompanying sintering lead to the formation of relatively low-melting calcium aluminosilicates that can act as the "glue" helping to form strong sintered ash deposits.
Introduction Deposition of ash on heat-transfer surfaces is a significant problem in the operation of coal-fired boilers. The problems have been discussed in detail in the coal combustion and ash chemistry literature.'+ In addition to the reduced heat transfer, ash deposits also impede gas flow and can cause physical damage to boiler internah2 The overall process of ash deposition involves transformations and reactions of the inorganic components of coal in the flame; transport of ash particles to deposition surfaces; initial adhesion of the ash particles, both to the deposition surface and to each other; and finally, further Present address: Department of ChemicalnginGing, E i v ersity of Delaware, Newark, DE 19016.
0887-0624/91/2505-0555$02.50/0
interactions among deposited ash particles to form a strong deposit. Particularly with low-rank coals, the initiation of deposition onto clean boiler surfaces begins with the formation of a so-called white layer, usually rich in sulfates.6 The white layer is able to capture ash particles that (1) Austin, L. G.; Benson, S. A.; Schobert, H. H.; Tangsathitkulchai, M. US.Department of Energy Report, DOE/FE-70770,1987. (2) Bono, R. W.; Levassew, A. A. In Mineml Matter and Ash in Coal; Vorrea, K . S., Ed.; American Chemical Society: Washington, DC, 1986; Chapter 21. (3) Raask, E. Mineral Impurities in Coal Combustion; Hemisphere: Washington, DC, 1985. (4) Schobert, H. H.; Conn, R. E.; Jung, B. R o c . 4th Annu. Pittsburgh Coal Conf. 1987,423-427. (5) Singer, J. G.; Combustion: Fossil Power Systems; Combustion Engineering: Windsor, CT, 1981; Chapter 3.
0 1991 American Chemical Society
Jung and Schobert
556 Energy & Fuels, Vol. 5, No. 4, 1991
strike the surface by inertial impaction. Because there is usually a very large temperature differential across the white layer, the first ash particles to impact the surface cool very rapidly! The adhesion of these particles to the surface is a factor in determining the need for sootblowing to remove the deposit. The ash particles captured by the white layer reduce the rate of heat transfer across the accumulating deposit so that eventually newly arriving particles cool slowly enough to allow particle-to-particle interactions to take placee6 The interactions among ash particles can include sintering, chemical reactions, and melting. The extent to which these interactions occur increases as the deposit grows away from the relatively cool deposition surface into the relatively hot gas stream. The sintered strength of the ash deposit ultimately determines whether or not the accumulated mass can be removed from the boiler tubes by sootblowers. Particles that have not sintered strongly can be removed easily, whereas strongly sintered deposits may be so resistant to fracture that normal sootblowing operations will not remove them. The process of ash sintering to form a strong deposit mass is complex. Fly ash is a heterogeneous material and the sintering process is often accompanied or preceded by one or more chemical reactions among the constituents of an ash particle. Chemical reactions and sintering may proceed at different rates at different locations within the deposit, depending on the local timetemperature history. In addition, factors such as ash particle shape, the particle size distribution, furnace temperature, and atmosphere can influence the course of the sintering process. Such factors also complicate the determination of the mechanisms of ash sintering. Several relatively recent investigations have focused on the development of strength in sintered There are three mechanisms by which sintering can occur: solid-state interactions, viscous flow of liquid formed by partial melting, or vapor-phase alkali transport.10 The viscous flow mechanism is considered to be the predominant mechanism for the formation of strong, sintered ash deposits in coal-fired boilers." This is due in part to the fact that most of the vapor-phase alkalies in combustion gases can condense to form low-melting alkali-metal aluminosilicates. This process has two consequences: it forms the low-melting "glue" for the viscous flow sintering; and it substantially reduces the mole fraction of alkali species in the vapor phase in immediate contact with the deposit, hence reducing the participation of vapor-phase alkali transport as a sintering mechanism in the deposit. The viscous flow sintering process has been described by a model developed by Frenkel, taking into account the viscosity and surface tension of the liquid phase.I2 Frenkel's equation is x 2 = 3ryt/2q where x is the radius of the interface between the sintered particles, r the radius of the particles, y the surface tension of the liquid, t the sintering time, and q the liquid vis~ o s i t y .In~ the Frenkel model the interparticle interface (6)Schobert, H.H.Lignites of North America; Elsevier Amsterdam, in press; Chapter 11. (7)COM, R. E. M.S.Thesis, The Pennsylvania State Univeristy, University Park, PA, 1984. (8) Gumming. I. W.; Joyce, W.I.; Kyle, J. H. J.Imt. Energy 1985,169. (9)Jung, B. Ph.D. Diaertation, T h e Pennsylvania State University, University Park, PA, 1990. (10)Kin ery, W.D.; Bowen, H. K.; Uhlmann, D. R. Introduction to Ceramics; &ley: New York, 1976. (11)Benson, S . A. Ph.D. Dieeertation,The Pennsylvania State University, University Park, PA, 1986. (12)Frenkel, J. J. J. Phys. (Moscow) 1946,9,385.
Table I. ASTM Ash Commsitionr (wt 70)of Four Coals component
SiO,
Robinson 34.9 15.6 0.6 11.7 16.4 2.7 4.4
0.0 13.6
Eagle Butte 28.5 15.5
0.8 10.2 22.7 5.6 3.9 0.0 12.7
Beulah 27.6 15.8 0.7 10.8 19.8 5.4 5.5
0.0 14.3
Elk Creek 56.7 30.1 1.7 6.7
1.1 0.9 0.5 2.0 0.3
is assumed to be circular and the particles, spherical. The Barnhart and Williams sintering test has provided an empirical relationship between the development of measurable mechanical strength in sintered, compacted ash samples in the laboratory, and strength of deposits in combustion system^.'^ A study of the sintering of specially prepared spherical particles of a glassy slag from coal ash suggested the applicability of the Frenkel model to ash deposition proce~ses.'~Transport of material by viscous flow is likely to be of particular importance in the sintering of coal ashes since the final product of physical and chemical transformations of the ash is often a glassy slag. Experimental studies of the sintering of spherical glassy particles have provided experimental verification of the Frenkel model.16J6 Laboratory sintering tests extended to brown coal ashes have shown a dependence of sinter strength and temperature of onset of sintering on ash particle size." These results were consistent with the observed ash behavior for formation of strong bonded deposits in boilers. Laboratory-scale evaluation of ash deposition from US.low-rank coals has demonstrated a relationship between the strength of the deposit, and the amount and viscosity of liquid formed and thus available for sintering." To gain a better understanding of the sintering behavior involved in ash deposit formation in coal combustion systems, with particular emphasis on low-rank coal ashes, four coal ashes were sintered as a function of temperature and particle size in an oxidizing (air) atmosphere to simulate the combustion systems. The sinter p~int'J&'~ of the ashes was determined by an electrical resistance method. Different particle size fractions of each ash were used to determine the effect of particle size on the sinter point. The conventional ash fusion tests (ASTM D 1857-87)20are insensitive for detecting small amounts of low-temperature melt phases, whereas the electrical resistance method is sensitive for monitoring the development of low-temperature sinter bonding at temperatures below the initial deformation temperature. The compressive sintered strengths of each ash were measured as a function of sintering temperature as a qualitative test of the Frenkel equation and to interpret and evaluate the physical and chemical changes of the sintered ash particles. The effect of particle size on the compressive sintered strength was also determined. After the compressive sintered strength was measured, the mineralogical phase changes of the sintered ashes were ascertained from X-ray diffraction analysis. (13)Barnhart, D. H.;Williams, P. C. Trans. AZME 1956, 78, 1229. (14)Raask, E. VGB Kraftswerkstechnik 1973,53,248. (15)Kuczynski, G.C. J. Appl. Phys. 1949,20,1160. (16)Kingery, W.D.; Berg, M. J. Appl. Phys. 1955,26,1205. (17)Dering, I. S.;Dubrovskii, V. A; Dik, E.E.Thermal Eng. 1972,19, 48. (18)Raask, E.J. Therm. Anal. 1979,16,91. (19)Cumming, J. W.J. Zmt. Energy 1980,153. (20)American Society for Testing and Materiale. Annual Book of ASTM Standards; Volume 05.06;ASTM Philadelphia, 1990.
Viscous Sintering of Coal Ashes Table 11. Summary of Relationships of Particle Size and Sinter Point coal ash particle size range, Gm sinter point, O C Robinson
