J. Phys. Chem. C 2007, 111, 14011-14020
14011
C/H2O Reaction under Supercritical Conditions and Their Repercussions in the Preparation of Activated Carbon Francisco Salvador,* M. Jesu´ s Sa´ nchez-Montero,† and Carmen Izquierdo‡ Departamento Quı´mica-Fı´sica, Facultad de Quı´mica, UniVersidad de Salamanca, 37008 Salamanca, Spain ReceiVed: May 15, 2007; In Final Form: July 4, 2007
Two chars prepared by carbonization of oak wood and anthracite were used to perform a comparative study of the gasification with supercritical water (SCW) and with steam. This work reports the effects of the type of char, the activating agent, temperature, flow rate, and particle size employed on the kinetics, mechanism of reaction, and the characteristics of the activated carbons obtained. The results show that the reactivity of the two chars is much higher with SCW than with steam. Although this increase can be explained in terms of the greater penetration of SCW and diffusional effects in the pore structure of the chars, some aspects suggest a possible change in the mechanism of reaction favored by the formation of clusters in SCW. The evolution of porosity was also found to differ when the char was gasified with SCW and with steam, being governed strongly by the starting material. When the oak char was activated with SCW, the smallest microporosity was broadened from the very first moments due to its very open pore structure, providing carbons with larger micropores and some mesoporosity. In contrast, in the case of the anthracite char, with a narrower pore structure, the evolution of the porosity was slower and less uniform, favoring external gasification of the particle. Accordingly, the carbons had a broader distribution of micropores, and mesoporosity was scarce.
1. Introduction The reaction of carbon with steam to form synthesis gas (CO + H2) is the basis of many industrial processes, such as the gasification of coal, the steam re-forming of natural gas or hydrocarbons, or the regeneration of coked catalysts. It is also the main procedure for the industrial manufacture of activated carbon. Other preparation procedures, such as gasification with CO2 or activation with chemical reagents (H3PO4, ZnCl, NaOH) are less common. The C/steam reaction is endothermic and in the absence of catalysts requires temperatures of 800-1000 °C for the gasification rate to be appreciable. In the 1960s, and especially in the United States, the interest aroused in the development of coal gasification technology led to a plethora of research projects in this field. A summary of such investigations was reported by Johnson.1 The main mechanistic studies of the C/steam reaction were performed by Gadsby et al.,2 Long and Sykes,3 StricklandConstable,4 Johnstone et al.,5 Wike and Rossberg,6 Binford and Eyring,7 and Ergun8 and later compiled in Laurendeau.9 Although this reaction has been addressed in many studies, only two relevant works have been conducted at high pressure: Blackwood and McGrory10 (1-50 bar) and Klaus and Wolfgang11 (1-10 bar). In those works the authors reported that the reactivity and mechanism of reaction under these conditions are different from when gasification is carried out at atmospheric pressure. In addition, although in recent years knowledge about and use of supercritical fluids have increased considerably,12 the C/H2O reaction under supercritical conditions has never been studied. * Corresponding author. Phone: +34 923 294478. Fax: +34 923 294574. E-mail:
[email protected]. † E-mail:
[email protected]. ‡ E-mail:
[email protected].
Two supercritical fluids are of particular interest, carbon dioxide and water. Water has a very high critical point (374 °C and 220 bar) due to its high polarity. Supercritical fluids have transport properties similar to those of gases: i.e., low viscosity, high diffusivity, and very low surface tension. However, the solvating power is similar to that of liquids. Moreover, in the particular case of water the ionic product decreases considerably and the dielectric constant becomes so low that the water is converted into an apolar solvent, dissolving organic compounds and gases and precipitating the salts. All these properties mean that supercritical water (SCW) has found many applications as an extraction solvent, as a reaction medium, and also as a reagent and/or catalyst. Matsumura et al.13 are the only investigators who have carried out a brief study on the gasification kinetics of a carbon activated with SCW. They found that the reaction rate was not affected by variations in total pressure. Their measurements were consistent with the rate equation for the C/steam reaction at atmospheric pressure proposed by Long and Sykes,3 suggesting that the gasification mechanism for high and low pressures is the same. The preparation of activated carbons with supercritical fluids, water and CO2, was initially proposed by Salvador et al.14 Currently, the number of publications addressing the issue continues to be low. Li et al.15 have reported a small development of the surface area and the creation of a certain degree of mesoporosity during the activation of a carbon fiber with CO2 and supercritical water. Cai et al.16 studied the development of porosity during the gasification of a phenolic resin char with SCW. The results of that work, with few data, suggested that SCW is able to develop more mesoporosity than steam. In a previous work,17 we studied the development of porosity in an olive stone char during gasification with SCW and with steam. The differences observed were attributed to the different
10.1021/jp073723e CCC: $37.00 © 2007 American Chemical Society Published on Web 08/28/2007
14012 J. Phys. Chem. C, Vol. 111, No. 37, 2007
Figure 1. Experimental setup.
reactivities of the two activating agents. The greater ease with which SCW penetrates the pore structure of the char affords reaction homogeneity, with a direct effect on the development of porosity. Nevertheless, Matsumura et al.13 have reported that the reaction mechanism for activation with steam at atmospheric pressure and with SCW could be the same. This is not consistent with our own results, since the activation energies of the gasification of olive stone char with SCW and with steam are different. The results of Matsumura et al.13 are also in conflict with those of Blackwood and McGrory.10 In light of the above, further studies are needed to elucidate the C/H2O reaction at high pressure. In the present work, we systematically address the gasification of two types of char, of different pore structures, with SCW and with steam. We analyze the effects of different experimental variables (the raw material, time, temperature, flow rate, and particle size) on the reactivity of the chars and on the characteristics of the activated carbons obtained. 2. Experimental Section Two chars were prepared by carbonization: one of oak wood at 650 °C and another of anthracite provided by Union Minera del Norte (Spain) at 700 °C in a stream of N2 (100 cm3/min) (25 °C; 1 bar) over 60 min. For the gasification of the chars with SCW and with steam, the experimental setup shown in Figure 1 was used. The flow reactor comprises a Hastelloy C276 tube measuring 30 cm in length, 12 mm o.d., and 7 mm thickness. The activation chamber is located in the lower part of the reactor. This is designed to receive a cylindrical sample holder containing a bed of char. Three thermocouples, type K, 1/16 o.d., in contact with the char measure the temperature through the bed (10 cm in length). Graphitized carbon fiber filters (inert) prevent the char from being carried away by the flow of the activating agent. The reactor is placed inside an oven, which maintains it at constant temperature. A fan rapidly homogenizes and stabilizes the oven temperature. A Eurotherm model 902 controller/ programmer measures and maintains the temperature of the char constant, with a precision of (2 °C. Previously degassed water circulates through the char bed, impelled by a Shimadzu model LC-10AS HPLC pump. Before entering the reactor, the water is heated in a preheater up to the working temperature. The stream of SCW crosses the bed of char, taking with it the gasification products. This stream is then cooled outside the oven before crossing the back-pressure regulators. Samples of 4 g of char (size, 0.5-1.40 mm) were gasified at a 4.0 g/min flow of water at atmospheric pressure and at 29
Salvador et al. MPa at different temperatures, 525-775 °C. The time of activation was varied to obtain different burnoffs. The gases generated during the gasification process were separated from the liquid stream with a separator and dried in a desiccator containing magnesium perchlorate for analysis. CO and CO2 were monitored using a Siemens model Ultramat 23 infrared analyzer. Samples of gas were analyzed chromatographically with a carbosphere column (80-100 mesh, 2 m length). The CO2 dissolved in water was also taken into account. Burnoff was determined from the loss of char weight and the ash content. The porous texture of the samples was analyzed from physical adsorption isotherms of N2 at 77 K and of CO2 at 273 K, measured in an automatic Micromeritics ASAP 2010 volumetric adsorption apparatus. The adsorption of the two gases provided complementary information about the micropore structure of the adsorbent. Apparently, at 77 K the penetration of N2 into micropore structures was severely limited by a slow process of activated diffusion, being adsorbed by capillary condensation in micropores larger than 0.7 nm. The adsorption of CO2 at higher temperatures (273 K) facilitated the activated diffusion of this molecule inside the microporous structure, penetrating smaller micropores. These isotherms were used to calculate the following: (a) the volume of the largest micropore (0.7-2.0 nm), V0(N2), and that of the micropores with the smallest size ( 1200 °C). Very few studies using elevated pressures have been published: Blackwood and McGrory10 (1-50 atm), Klaus and Wolfgang11 (1-10 bar), and Matsumura et al.13 (25.5-34.5 MPa (SCW). Under these conditions, the reactivity is greater than at atmospheric pressure, and appreciable amounts of other reaction products, such as CO2 and CH4, are produced, even when working at temperatures far below 1200 °C. Blackwood and McGrory10 investigated the C/steam reaction in a pressure range of 1-50 atm and a temperature range of 750-830 °C, observing an increase in reactivity with pressure and the formation of considerable amounts of CH4. Those authors assumed that higher steam pressure promotes the conversion of adsorbed H2 to CH4, thus increasing the C(O) concentration and hence the global rate. They suggested a modification of the hydrogen inhibition mechanism to explain the formation of CH4. Klaus and Wolfgang11 studied the C/steam reaction at 1000 °C at a total pressure of 10 bar and several water partial pressures, focusing their attention on the gases generated. They concluded that the CO2 was not formed at the surface but was a consecutive product of the CO (shift reaction). In contrast, the CH4 would be formed at the surface of the carbon, but not through the adsorbed hydrogen, C(H2), as proposed by Blackwood and McGrory.10 Matsumura et al.13 studied the gasification of a carbon activated with SCW (600-650 °C and 25.5-34.5 MPa), reporting that in all cases the composition of the gas contained a large amount of H2 and CO2 and a small amount of CH4 and CO (64% of H2, 33% of CO2, 2% CH4, and 1% CO). Results identical to those of Matsumura were found in the activation of the two chars studied in the present work, confirming the idea that during gasification with SCW the water-gas shift reaction must be of great importance. Melius et al.21 investigated the mechanism of the water-gas shift reaction under supercritical conditions and proposed that the reaction would take place in two steps, with the formation of formic acid, which would later decompose into CO2 and H2O:
CO + (n + 1)H2O f HCOOH + nH2O f CO2 + H2 + nH2O Quantum chemistry calculations carried out by those authors indicated that if no additional water molecules participate in the reaction (n ) 0), the activation energy for the first step of the reaction is 61.7 kcal/mol, and for the second step 64.9 kcal/ mol. An additional water molecule (n ) 1) participating in the first step of the reaction could form a transition-state ring structure with the carbon monoxide molecule. Such a structure would require only about half the activation energy, 35.6 kcal/
14016 J. Phys. Chem. C, Vol. 111, No. 37, 2007
Salvador et al.
mol, needed for the reaction when n ) 0. If an additional water molecule also participated in the second step of the reaction, the activation energy for this step would be cut about 40% to 37.3 kcal/mol. If two additional water molecules participated in the reaction (n ) 2), the activation energies would be even smaller. In these calculations, water would act as a catalyst for the reaction by lowering the energies of activation. These catalytic water molecules would be more likely to participate in the reaction if it were carried out in supercritical water because its high compressibility would promote the formation of solutesolvent clusters. These effects of the density of water in the water-gas shift reaction were also investigated by Stevens et al.22 in a study of the reaction in SCW from 410 to 520 °C and 2.0-60 MPa. The results of this experimental work revealed a noticeable increase in the reaction rate with the increase in pressure. The data support the theoretical prediction of the existence of a polar transition-state complex, characterized by an unusually large negative volume of activation that results from a dramatic change in the local density of the SCW. Although supercritical conditions provide a large displacement in the equilibrium of the water-gas shift reaction toward the formation of CO2 and increase its reaction rate, there are serious doubts as to whether all the CO given off from the surface of the carbon during the gasification, reaction R3, is converted into CO2 in the short residence time in the flow-mode gasification reactor. Accordingly, and taking into account the mechanism of the water-gas shift reaction propitiated by the supercritical conditions, this latter reaction could be considered an alternative to reaction R3 and would be responsible for the greater reactivity of the char in SCW:
C(O) + H2O f CO2 + H2 + Cf
Figure 7. Evolution of different sets of micropore volume along the gasification of anthracite char with SCW and steam.
(R6)
This reaction would be favored by the formation of clusters or aggregates on the C(O) of the surface. According to this reaction, the CO2 would be a primary product of the gasification process. Due to the absence of kinetic data concerning the gasification of carbon with SCW, further intense work should be carried out to confirm this hypothesis. 3.4. Evolution of the Porosity. Study of the evolution of porosity with gasification was performed from the adsorptiondesorption isotherms of N2 at 77 K and the adsorption isotherms of CO2 at 273 K. Figures 7 and 8 show the evolution of the volume of the largest, V0(N2,), and of the smallest, V0(CO2), micropores during the activation of the anthracite and oak chars with SCW and steam at 700 and 600 °C, respectively. There are important differences, depending on the type of activating agent and char used. In the case of the anthracite char gasified with SCW, a significant increase was seen in V0(N2,) up to around 50% burnoff. However, the increase in V0(CO2) was much less marked. The values of both volumes were the same up to around 11% burnoff. These data can be interpreted in terms of the existence of an initial increase in the narrowest microporosity, followed by a widening to broader pores. This kind of behavior is similar to that observed for the gasification of olive stone char with SCW.17 When the anthracite char was gasified with steam, the behavior of the char was different: V0(CO2) and V0(N2) increased with the increase in burnoff, this increase being higher for V0(N2). Up to a burnoff of 25%, V0(CO2) was greater than V0(N2).
Figure 8. Evolution of different sets of micropore volume along gasification of oak char with SCW and steam.
If the microporosities of the carbons prepared with the anthracite char with the two activating agents are compared, it may be seen that V0(N2) is greater in the carbons prepared with SCW up to approximately 40% burnoff. This is due to the effect of the SCW, which facilitates penetration to the interior of the char particle, even into the finest pores. Thus a strong increase in the micropore volume occurs at low burnoffs. In the case of the oak char activated with SCW, V0(N2) also increased up to around 40% burnoff. However, unlike the anthracite char, V0(CO2), decreased rapidly as from very low burnoffs. This different kind of behavior can again be attributed to the different porosities of the two chars. The micropore
C/H2O Reaction under Supercritical Conditions
Figure 9. Evolution of mesopore volume along the gasification of anthracite char and oak char with SCW and steam.
Figure 10. Effect of temperature on the evolution of micropore volume (closed symbols) and mesopore volume (open symbols) along the gasification of anthracite char with SCW (9, 700 °C; B, 725 °C; 2, 750 °C).
structure of the oak char is more open and is broadened since the very start of the gasification. The evolution of the porosity of the oak char gasified with steam was similar to that of the anthracite char gasified with steam. The evolution of mesoporosity with gasification is also a function of the activating agent employed and of the char used, Figure 9. For the two chars gasified with SCW, Vmes increased throughout the burnoff range, but increased faster for mediumto-high burnoff. When the chars were gasified with steam, the development of mesoporosity was lower. The oak char developed greater mesoporosity than that of anthracite because its
J. Phys. Chem. C, Vol. 111, No. 37, 2007 14017
Figure 11. Effect of temperature of the evolution of micropore volume (closed symbols) and mesopore volume (open symbols) along the gasification of oak char with SCW (9, 550 °C; B, 575 °C; 2, 600 °C).
Figure 12. Effect of flow rate on the evolution of micropore volume (closed symbols) and mesopore volume (open symbols) along the gasification of anthracite char with SCW (9, 8 g/min; B, 4 g/min; 2, 1 g/min).
micropore structure is broader, favoring the conversion of micropores into mesopores. The effect of temperature in the development of the porosity of the two chars during gasification with SCW is shown in Figures 10 and 11, in the 700-750 and 550-600 °C range. From this information it may be inferred that the gasification temperature has a small influence on the development of the micro- and mesoporosity of the anthracite char but not on that of oak. In the carbon samples prepared from the anthracite char,
14018 J. Phys. Chem. C, Vol. 111, No. 37, 2007
Figure 13. Effect of particle size on the evolution of micropore volume (closed symbols) and mesopore volume (open symbols) along the gasification of anthracite char with SCW (9, 2.00-3.40 mm; B, 1.002.00 mm; 2, 0.50-1.00 mm).
V0(N2) decreased slightly with temperature, whereas the Vmes volume increased slightly. These small variations in the development of porosity contrast with the strong variation in the gasification rate with SCW in that temperature range, Figure 2. The porosity of the oak char was not affected by the gasification temperature and only depended on the burnoff. Figure 12 shows the effect of the flow rate of SCW on the development of the porosity of the anthracite char gasified at 700 °C. The higher V0(N2) corresponds to the series gasified with a flow rate of 4 g/min, the greatest differences being found at
Salvador et al. intermediate burnoff values. The Vmes volume increases with flow, this increase being more pronounced for higher flows. In the case of the oak char, no effect of flow rate on the development of porosity was observed. We also investigated the effect of particle size on the textural characteristics. Again, only the porosity of the anthracite char was seen to be slightly affected by the particle size, Figure 13. For the same burnoff, V0(N2) decreased when the particle size decreased; by contrast Vmes increased when the particle size increased. All the differences observed in the evolution of porosity with the gasification temperature, flow rate, and particle size between the two chars could be explained in terms of their different reactivities. In the anthracite char, the diffusional limitations imposed by its fine microporosity mean that the concentration of SCW inside the particle would decrease and that the reaction outside would predominate. The evolution of the texture of the two chars gasified with steam and with SCW was studied at microscopic scale using scanning electron microscopy. As shown in Figure 14, the particles of the anthracite char had a very compact laminar structure and smooth surfaces. Some of the particles displayed mineral aggregates strongly incrusted in the char. SEM-EDX analyses of these structures revealed that the major components were Fe and Ca. With gasification, the particles of anthracite became very altered. The surfaces were now rougher and some layers were opened, displaying exfoliation. This exfoliation appeared at low burnoffs and was increased with high burnoffs. The exfoliation was also enhanced when high temperatures were used. In all cases, the exfoliation was significantly less pronounced with SCW than with steam. Possibly, the formation of the crack could be due to shrinkage of the layers as a result of greater activity at the surface than in the interior of the char. All this would be propitiated by the difficulty encountered by the activating agent to penetrate the char. Since steam has a lower penetration potential than SCW, it will carry out activation mainly at the outer surface of the layers, thereby increasing the degree of
Figure 14. SEM micrograph of anthracite char: (a) anthracite char gasified with SCW at different burn-offs and (b) anthracite char gasified with steam at different burnoffs.
C/H2O Reaction under Supercritical Conditions
J. Phys. Chem. C, Vol. 111, No. 37, 2007 14019
Figure 15. SEM micrograph of oak wood, oak char: (a) oak char gasified with SCW at different burnoffs and (b) oak char gasified with steam at different burnoffs.
Figure 16. Evolution of the ash content with burnoff during gasification of the anthracite char with SCW and steam.
Figure 17. Evolution of the ash content with burnoff during gasification of the oak char with SCW and steam.
exfoliation. In contrast, SCW, even though it induces activation at the surface of the layers, is able to partially penetrate the internal structure of the layers and develop their porosity. Figure 15 shows the SEM images of oak wood, the char, and the activated carbons prepared with SCW and with steam at different burnoffs. The oak wood exhibited a pattern of lignocellulosic vessels arranged horizontally and vertically. During the process of carbonization, the material became compacted, although the surface of the char clearly displayed the cellular structure of the starting material. It was also possible to detect the presence of mineral matter strongly encrusted within the char. The composition of this mineral material was analyzed with the EDX connected to the SEM, and the results indicated this was mainly Ca and K. In the carbons activated with low burnoff, the external surface showed the same physical aspect as the char, but in the case of
the carbons activated with SCW the mineral matter was seen to have become detached from the char in the form of crystalline structures. For high burnoffs, the alteration of the external surface was appreciable and was more marked in the case of the carbons activated with SCW. In these carbons the elimination of mineral matter continued, although now the crystalline aggregates were smaller. It seemed as though they had been slowly dissolved by the SCW. 3.5. Ash Content Evolution. Figures 16 and 17 show the evolution of the ash content during the gasification of the two chars with SCW and steam. In the case of the anthracite char, the ash content increased with burnoff when it was gasified with SCW and with steam. However, this increase was lower when SCW was used. Regarding the oak char activated with SCW, the ash content decreased dramatically at the beginning of the gasification, thereafter increasing slightly with burnoff. Carbons with 50% burnoff had a lower ash content than the starting char.
14020 J. Phys. Chem. C, Vol. 111, No. 37, 2007 Nevertheless, in the oak char gasified with steam the content of ash increased with burnoff. These findings confirmed the SEM observations, where it was observed that SCW was able to extract the mineral matter with greater ease, above all in the case of the oak char. Thus, gasification with SCW can be said to afford purer activated carbons than those obtained with steam. 4. Conclusions The rate of gasification of the two chars with SCW is much greater than gasification with steam as a result of the better penetration of SCW into the pore structure of chars. The oak char was more reactive than that of anthracite and required lower temperatures (525-600 °C). The effect of particle size and flow rate, together with the activation energies and the preexponential factors of the Arrhenius equation suggest that (i) the process of gasification of oak char with SCW is governed by the chemical reaction, (ii) in the case of the anthracite char/SCW, the process could be partially controlled by transport and diffusion phenomena in the pores, since this char has a narrower pore structure, and (iii) the gasification of both chars with steam is diffusion-controlled. Although the results reported here do not confirm the idea that the mechanisms of reaction with SCW and with steam are necessarily different, there is evidence to suggest that such a change in mechanism would be possible as a result of the special properties of SCW (low polarity, high density, and cluster formation, etc.). The evolution of the porosity of the two chars when activated with SCW and with steam is different, especially in the case of oak char. The porosity of the oak char changes very rapidly during gasification with SCW, the whole of the micropore structure simultaneously and homogeneously becoming wider. This evolution is unaffected by the different experimental conditions employed. In contrast, in the gasification of the anthracite char with SCW the pore development is slower and less uniform, as confirmed with the SEM study. In addition, in this case the evolution of porosity is slightly affected by the experimental variables (temperature, flow rate, and particle size) and external gasification is favored.
Salvador et al. Finally, when SCW is used, the gasification process is accompanied by the extraction of mineral matter, such that purer activated carbons can be obtained. Acknowledgment. Financial support from the Ministerio de Educacio´n y Ciencia, Spain, and the European Regional Development Fund (Project CTQ2006-00759/PPQ) is acknowledged. References and Notes (1) Johnson, J. L. Kinetics of Coal Gasification; Wiley-Interscience: New York, 1979. (2) Gadsby, J.; Hinshelwood, C. N.; Sykes, K. W. Proc. R. Soc. London 1946, A187, 129. (3) Long, F. J.; Sykes, K. W. Proc. R. Soc. London 1948, A193, 377. (4) Strickland-Constable, R. F. J. Chim. Phys. 1950, 47, 356. (5) Johnstone, J. F.; Chen, C. Y.; Scott, D. S. Ind. Eng. Chem. 1952, 44, 1564. (6) Wike, E.; Rossberg, M. Z. Elektrochem. 1953, 57, 641. (7) Binford, J. S., Jr.; Eyring, H. J. Phys. Chem. 1956, 60, 486. (8) Ergun, S. U.S. Bur. Mines Bull. 1962, 598, 38. (9) Laurendeau, N. M. Prog. Energy Combust. Sci. 1978, 4, 221. (10) Blackwood, J. D.; McGrory, F. Aust. J. Chem. 1958, 11, 16. (11) Klaus, J. H.; Wolfgang, F. M. Carbon 1992, 30 (6), 883. (12) Noyori, R., Ed. Supercritical Fluids. Chem. ReV. 1999, 99, 355. (13) Matsumura, Y.; Xu, X.; Antal, M. J., Jr. Carbon 1997, 35, 819. (14) Salvador, F.; Sa´nchez, C.; Mercha´n, M. D.; Salvador, A. Spain Pat. 2155746, 1998; Eur. Pat. 0974553; U.S. Pat. 6.239.067. (15) Li, Y. Y.; Mochidzuki, K.; Sakoda, A.; Suzuki, M. Carbon 2001, 39, 2143. (16) Cai, Q.; Huang, Z. H.; Kang, F.; Yang, J. B. Carbon 2004, 42, 775. (17) Molina-Sabio, M.; Sa´nchez-Montero, M. J.; Jua´rez-Galan, J. M.; Salvador, F.; Rodrı´guez-Reinoso, F.; Salvador, A. J. Phys. Chem. B 2006, 110, 12360. (18) Radovic´, L. R.; Walker, P. L., Jr.; Jenkins, R. G. Fuel 1983, 62, 849. (19) Walker, P. L., Jr.; Rusinko, F., Jr.; Austin, L. G. In AdVances in Catalysis; Eley, D. D., et al., Eds.; Academic Press: New York, 1959; Vol. 11, p 133. (20) Pis, J. J.; Mahamud, M.; Pajares, J. A.; Parra, J. B.; Bansal, R. C. Fuel Process. Technol. 1998, 57, 179. (21) Melius, C. F.; Bergan, N. E.; Shepherd, J. E. 23rd Symposium (International) on Combustion; The Combustion Institute: Pittsburgh, PA, 1991; p 217. (22) Stevens, F. R.; Richard, R. S.; Jason, D. A. J. Phys. Chem. A 1998, 102, 2673.
