Sorbents for SO2 removal from humidified simulated flue gas have been prepared from CaO-fly ash slurries using a pressure hydration technique. The fly ash materials were obtained from the combustion of four bituminous, six sub-bituminous, and two lignite coals. The CaO/fly ash ratio, and the slurry reaction time and temperature were varied in the preparation step. The sorbents were characterized by measuring thermogravimetrically the SO2 uptake from flue gas, and by X-ray diffraction, surface area, particle size, and optical microscopy studies. The X-ray diffraction results indicated that calcium silicate hydrate, Ca2SiO4·H2O, and calcium aluminum silicate hydroxide, Ca3Al2(SiO4)(OH)8, were the principal Ca-containing species formed during the hydrothermal reaction between CaO and fly ash. There was a good correlation between the surface area of the sorbent and the SO2 uptake. Particle size measurements and optical microscopy studies indicated that the glass-like spheres in fly ash remained largely intact during the hydrothermal preparation step.

A thermogravimetric (TG) technique has been developed for studying the effectiveness of Ca(OH)2 sorbents for removal of SO2 from humidified simulated flue gas. Results obtained with the TG technique were in good agreement with earlier fixed-bed reactor studies, indicating that the TG technique is a satisfactory method for screening sorbents. An impregnation procedure was used to investigate the ability of eight inorganic salts to increase the utilization of hydrated lime. The alkali halides were most effective, with more than a twofold increase in utilization. A second approach to preparing the sorbent—dissolution of the salt in the water used to prepare the hydrate from CaO at 95°C—gave less satisfactory results. No correlation could be found between the H2O incorporation and the SO2 uptake on the sorbents. Data from other studies were compared with the TG results, and it was concluded that the manner of combining the additive with the Ca(OH)2 was an important factor in determining additive effectiveness.

Precipitated iron Fischer-Tropsch catalysts were prepared at various constant pH levels with a Kölbel in-line mixing unit using either sodium carbonate or ammonium hydroxide as the precipitant. In microreactor studies at 1.38 MPa and 548 K, the Na2CO3-precipitated catalysts prepared at low pH (3.7 and 4.7) showed unusually high olefin selectivity (C2C4) as well as activity maintenance superior to that of the catalysts prepared at higher pH (5.8, 7.6, and 9.8). While the NH4OH-precipitated catalyst showed lower olefin selectivity, the activities of these catalysts, especially the copper-promoted ones, were quite stable for periods of up to 300 h. Thermogravimetric studies indicated that graphitic carbon deposition occurred to a greater degree on Na2CO3-precipitated catalysts prepared at high pH. The high soda content of these catalysts was identified as one factor accounting for this deposition and presumably also accounted for the higher rates of deactivation. Those factors responsible for the high olefin selectivity of the Na2CO3-precipitated catalysts prepared at low pH remain unidentified.

Copper oxide is a promising sorbent for combined flue-gas-cleaning processes because it is one of the most reactive metal oxides for the sorption of SO/sub 2/ from synthetic flue gas. In addition, CuSO/sub 4/ catalyzes the reduction of nitrogen oxides by ammonia. Preparation and regeneration schemes for unsupported CuO have been investigated to optimize the reactivity of the sorbent material. Sorbent effectiveness for flue gas desulfurization is discussed in relation to physical properties of the CuO powders.

Discrete Fe-N phases which are difficult to prepare by common chemical methods have been synthesized by rf sputtering. Some of these phases are relevant in the Fischer-Tropisch synthesis of synfuels. Single Fe-N phases as well as the desirable mixtures have been synthesized by a careful control of various sputtering parameters. These films showed varied columnar microstructure with the columns perpendicular to the film plane. Magnetic anisotropy measurements for selected samples containing pure Fe4N, ε-Fe3N, and Fe4N+αFe have been carried out. The anisotropy components when the field (H) is perpendicular or parallel to the columns show significant differences in the magnetization versus H curves up to fields of 8 kOe at room temperature and evidence for superparamagnetism. These results are explained on the basis of structural anisotropy of the columnar structure of the films. Mössbauer results for the ε-nitride film are compared with those reported in the literature and again related to the microstructure.

This study was carried out with the objective of preparing more active and stable nitrided iron catalysts for the conversion of synthesis gas to a product with a high alcohol content that could be used directly as an automotive fuel or converted to a gasoline-like product over a shape selective zeolite in a dual reactor unit. The rationale is given for the preparation of a molybdenum-promoted nitrided fused iron catalyst. Catalyst characterization yielded equivocal results as to whether a mixed Fe-Mo nitride was formed. Regardless of the nature of the Mo-containing species, these catalysts were significantly more active than unpromoted nitrided fused iron catalyst. Characterization of fused iron catalysts using thermogravimetric analysis indicated the rates of reduction, carburization, and nitriding were strongly dependent on reaction temperature and particle size as well as the nature of the promoter. X-ray diffraction measurements indicated the metal crystallite size of the reduced catalysts was strongly dependent on reduction temperature. A microreactor unit capable of on-line gas chromatographic analysis was constructed for the screening of a variety of carbided iron catalysts. The carbon number distribution obtained using a carbided fused iron catalyst was found to correlate with the Anderson-Schluz-Flory equation.

When comparing catalysts on a spece-time-yield basis, the potassium-promoted iron-cobalt catalyst and the potassium-promoted iron catalyst are the most active. While the iron-manganese catalyst is quite active on a per gram of iron basis, on a per gram of catalyst basis it is only about 20% as active as the potassium-promoted precipitated iron catalyst. The fused iron catalyst is the least active of any catalyst studied, but this catalyst was reduced at 723/sup 0/K (vs 523/sup 0/K for the precipitated iron catalyst). This catalyst can be reduced at 623/sup 0/K in about ten hours, and considering the large effect of reduction temperature on carbiding rate, there is reason to believe the lower reduction temperature will result in a more active catalyst. While the fused iron, iron-cobalt-potassium, and iron-manganese catalysts showed little change in conversion after being placed on-stream, the precipitated iron catalysts, B and C, declined significantly during the first 12 hours and then stabilized. Olefin selectivity as measured by the propylene/propane ratio decreases with time on-stream, although the degree of polymerization changes little during the first 50 hours. The degree of polymerization decreased in the following order: fused iron > K-promoted ppt. Fe > ppt. Fe, Fe-Mn > Fe-Co-K.more » As far as maximizing gasoline yield is concerned, it appears the Fe-Mn catalyst is to be preferred not only because of the high C/sub 5/-C/sub 11/ selectivity but also because the olefin content is the highest of any catalyst in this study. 15 figures, 3 tables.« less