Charles E. Hamrin

Sulphur transformation and removal for Western Kentucky coals

Abstract Inhibition isotherms were measured for Western Kentucky No.9 coal. Crushed and sieved coal (−25 + 140 U.S. mesh) was fluidized in 10-g batches in a 22-mm i.d. quartz reactor up to a temperature of 870 °C. The release of hydrogen sulphide during heatup under nitrogen and at the run temperature (usually 1–2 h) under the same gas (pyrolysis), hydrogen, or hydrogen/hydrogen sulphide mixtures was followed by gas chromatography. The residue or char was analysed for pyritic, organic, sulphide, sulphate, and total sulphur. Inhibition isotherms, which are pseudo-equilibria between sulphur in the char and gaseous hydrogen sulphide, were measured at 600 and 870 °C. At the lower temperature the isotherm was found to be independent of the hydrogen sulphide concentration in the gas stream and the char sulphur content remained constant at 2.6%. At 870 °C the sulphur content of the char was greater than that of the original coal when gas mixtures of 1, 3, and 6% hydrogen sulphide in hydrogen were used, indicating the necessity of maintaining low hydrogen sulphide concentration for sulphur removal. In pure hydrogen, sulphur removal increased continuously from 47% at 600 °C to 84% at 870 °C. For pyrolysis under nitrogen, sulphur removal was 40% at 600 °C and increased to 59% at 740 °C. No further removal occurred above this temperature up to 870 °C. In addition to the inhibition isotherms, sulphur-form transformation diagrams were constructed for coal treated with nitrogen, hydrogen, and hydrogen/hydrogen sulphide mixtures. Pyritic sulphur, which comprised 40% of the sulphur in the original coal, was completely converted to ferrous sulphide at 600 °C in hydrogen and 740 °C in nitrogen. At 870 °C the sulphur content of the char produced under hydrogen was 1.1% made up of 48.4% ferrous sulphide, 43.4% organic sulphur, and 8.2% sulphate.

OXYGEN AND NITROGEN IN COAL BY INSTRUMENTAL NEUTRON ACTIVATION ANALYSIS

The feasibility of using fast neutron (14 MeV) activation analysis techniques for the determination of oxygen and nitrogen in coal has been investigated. Conditions that favor instrumental neutron activation analysis (INAA) include the absence of problems associated with sample dissolution and the capability of extremely rapid analyses as compared to older techniques such as the Kjeldahl method for nitrogen. Most previous oxygen determinations have been by difference after major component analyses. In the present study, oxygen was determined in sized coal and its low temperature ash (LTA) with the difference representing the organic oxygen content. Both the oxygen and nitrogen analyses employ a multiscaling technique with the former based on the16O(n, p)16N reaction, while the latter utilizes the annihilation radiation produced by the product of the14N(n, 2n)13N reaction. The high-energy gamma-radiation associated with the decay of16N was essentially free of spectral interferences for coal analysis, although fluorine could cause a primary interference if the F/O ratio exceeds 0.02. In the nitrogen work, experiments were performed to determine correction factors to account for the effects of the12C(p, γ)13N and13C(p, n)13N “knock-on” reactions and the39K(n, 2n)38K reaction which produce interfering β+ emitting radionuclides. Data are presented for oxygen in Western Kentucky No. 9 and No. 11 coal and coal ash and for nitrogen in eleven different coals.

Comparison of methods for the determination of organic oxygen in coals

Abstract Coals distributed widely in rank and geographical origin have been analysed for organically-bound oxygen by several conventional and fast neutron activation analysis (FNAA) techniques. Pyrolysis of demineralized coal with measurement of evolved CO2 by coulometry was found to be the most reliable of the conventional methods. Direct FNAA determinations of organic oxygen by analyses of demineralized coal (DMC) samples yielded data in excellent agreement with pyrolysis, as did values computed using a modified ‘by difference’ calculation. The convergence of data from these totally independent approaches, suggests a measure of true organic oxygen levels has been achieved. A ‘difference method’ based on FNAA determinations of total dry coal oxygen and inorganic oxygen in low temperature ash yielded organic oxygen data that were typically lower than pyrolysis values, possibly due to oxidation of organic sulphur in the ashing process and/or the presence of non-extractable mineral oxygen in the DMC used in the pyrolysis method. A third FNAA ‘difference method’ based on simultaneous determinations of both total oxygen and silicon contents of dry whole coals, followed by estimation of the inorganic oxygen contents based on the silicon contents, was found to be rapid and adaptable to on-stream analysis. However, observed mineralogy-dependent deviations from a simple inorganic oxygen-silicon relationship suggest that the latter technique would be most successful when applied to coals with similar mineralogy.

Determination of oxygen and nitrogen in coal by instrumental neutron activation analysis

Abstract The purpose of this study was to measure oxygen and nitrogen in coals using instrumental neutron activation analysis. For six U.S. coals total oxygen ranged from 9.4 to 28.7% and total nitrogen varied from 0.72 to 1.61%. To obtain values of organic oxygen and nitrogen either a low-temperature-ashing method or an acid-treatment method was suitable for bituminous coals. The mean difference of the experimentally determined values (Odmmf)LTA − (Odmmf)AT = −0.82, s = 0.51, was found to be statistically significant at the 95% confidence level, but the comparable difference for nitrogen was not. By the LTA method oxygen and nitrogen on the dmmf basis for bituminous coals showed no statistically significant difference with calculated dmmf values. Nitrogen was detected in all the LTAs varying from 0.38 to 1.67%. Formation of insoluble CaF2 in the acid-treatment method caused an interference in the nitrogen determination due to the 19F (n, 2n) 18F reaction but was correctable. In addition, recoil proton reactions on C and O leading to the formation of 13N must be accounted for in all nitrogen determinations in the coal matrix.

Starting materials and superconductor stoichiometry

Abstract Oxygen contents of as-received La 2 O 3 and Nd 2 O 3 were determined by fast neutron activation analysis. X-ray diffraction was used to determine the phases present in these materials. Results of these analyses show that significant amounts of hydroxides (in one case over 90 wt%) were present in all oxide samples. These results carry tremendous implications for the stoichiometry of the superconductor formation reaction and the composition of the final product. The most important conclusion to be drawn from this study is that many of the bulk superconducting ceramics produced to date may be of approximate composition at best.

Studies of the superconducting oxides NdxY1−xBa2Cu3Oy

Abstract We have prepared polycrystalline samples of the superconducting oxides NdxY1−xBa2Cu3Oy with x=0, 0.1, 0.5, 0.9, and 1.0. A SEM investigation indicates, for neodymium containing samples, 20 micron size particles in the interior and a “basket weave” morphology on the pressed surface. EDX and x-ray diffraction reveal no indication of rare earth segregation in the alloys. The in-plane lattice constants increase by 1% with x, but ther is no variation in the orthorhombic splitting (b/a−1 = 2%). All samples have transition (50% resistance) temperatures between 91 K and 92 K.

A study on the performance of the catalytic porous-wall three-phase reactor

A theoretical investigation of a catalytic porous-wall reactor in which gaseous and liquid reactants approach each other from opposite sides of the catalyst is undertaken. Equations for the annular liquid-channel are coupled with those for the catalytic wall and solved numerically and analytically using a simplified model. For the model reaction under study, the main design and operation parameters which affect reactor performance are the Thiele modulus, Peclet number, width of the liquid channel and the inlet concentration of the reactant in the liquid phase. The effect of reactor configuration is peculiar to the cylindrical geometry because the thickness and relative location of the catalytic wall as well as the selection of the liquid and gas channels can influence the reactor performance. Thin-walled catalyst tubes have larger effectiveness factors and as the tube radius approaches that of the reactor, conversion in the reactor increases especially when the liquid is saturated with the gaseous reactan...

Hydrodesulphurization of thiophene by coal mineral matter and a cobalt-molybdate catalyst in a pulse reactor

Abstract A pulse reactor was used to compare the catalytic activity of a commercial desulphurization catalyst (Nalco 471) and of mineral matter from Western Kentucky coal. Hydrodesulphurization of thiophene, a model coal sulphur compound, was the reaction studied. Mineral matter was obtained from the coal in its least altered state by a low-temperature, oxygen-plasma technique commonly referred to as low-temperature ashing. Conversion is determined from the C 4 gases which are separated in a two-column Chromatographic system. At 748 K it was found that thiophene conversion with mineral matter was 12% of that with the commercial catalyst. Relative activities of hydrogenation of intermediate butenes to butane, the effect of presulphiding, and that of pyridine poisoning were also determined.

Application of the general purpose collocation software, PDECOL, to the Graetz problem

Abstract The software package PDECOL is capable of solving coupled systems of parabolic, nonlinear, partial differential equations. Its use of B-spline collocation for spatial discretization makes it valuable for obtaining solutions in which extreme gradients are present. A severe limitation of this package is the requirement that the boundary and initial conditions be consistent. The package can, however, be modified to overcome this limitation. To demonstrate the utility of the modified PDECOL package, the solution of the classical Graetz problem is obtained for different flow geometries and for some common boundary conditions. The results indicate that excellent agreement with the analytical results can be obtained provided that a suitable distribution of a sufficient number of spatial grid points is used and an error criterion compatible with the desired accuracy is chosen for the time integration. For this steady-state problem this integration is with respect to the spatial dimension which appears in the first derivative.

Desulphurization and deagglomeration of Western Kentucky coal using bottom-ash

Abstract Fluidization of bottom-ash from a producer-gas generator with Kentucky No.9 coal at a temperature of 870 °C reduced the hydrogen sulphide emitted and prevented agglomeration (or caking) of the coal at bottom-ash/coal ratios greater than 0.4. The sorption capacity of the ash ranged from 1.6 to 9.4 mg S/g bottom-ash at this temperature. Scanning electron microscopy illustrated the growth of coal particles on heating and the various stages of melting and sphere formation.