No abstract

Erosion damage was estimated for the first stage of a large electric utility gas turbine based on projected particle distributions in the gas leaving the hot gas cleaning system of a pressurized fluidized-bed gasifier system. Based on the assumptions used in making the estimates, cleaning of the turbine expansion gas to a particulate concentration of 0.005 gram per standard cubic metre (0.002 grain per standard cubic foot) with particles larger than 6-μm diameter effectively removed should give satisfactory blade life from an erosion standpoint. Two stages of high-performance cyclone cleanup to 0.1 gram per standard cubic metre (0.5 grain per standard cubic foot) with 0.05 weight percent of 12-μm diameter particles remaining in the gas would wear stator vane trailing edges by 0.25 cm (0.1 in.) thickness (roughly equivalent to full wall thickness in upstream stage vanes) in 10000 h of operation. The numerical results presented in this paper are based on the estimate that coal ash and sulfur sorbent particles will have, when impacting superalloy turbine materials under turbine conditions, 1/25th of the erosivity of silicon carbide particles impacting a nickel alloy at room temperature. The estimates do not account for the appreciable slowing of the 1- to 3-μm particles in the blade boundary layers before they reach the blading, even though these small particles account for most of the damage. The numerical results are in this way conservative. Actual data on the damage which coal gas particulates do to blade materials under turbine conditions are needed to establish the erosion tolerance of the turbine more accurately.

A theory is presented to predict deposition rates of fine particles in two-dimensional compressible boundary layer flows. The mathematical model developed accounts for diffusion due to both molecular and turbulent fluctuations in the boundary layer flow. Particle inertia is taken into account in establishing the condition on particle flux near the surface. Gravitational settling and thermophoresis are not considered. The model assumes that the fraction of particles sticking upon arrival at the surface is known, and thus, treats it as a given parameter. The theory is compared with a number of pipe and cascade experiments, and a reasonable agreement is obtained. A detailed application of the model to a turbine is also presented. Various regimes of particle transport are identified, and the range of validity of the model is discussed. An order of magnitude estimate is obtained for the time the turbine stage can be operated without requiring cleaning.

No abstract

Cobalt chromite is an electronically conductive oxide stable in both air and mixtures of H2-H20 and CO-CO2 having oxygen partial pressures greater than 10 -15 atmospheres at 1000~ It is potentially useful as a material to interconnect cells of a high-temperature, zirconia-electrolyte, fuel-cell battery. The 1000~ resistivities and conductivity-types of cobalt chromite doped with manganese and v a n a d i u m were measured over the range of oxygen partial pressures of interest for fuel cell operation. Vanadium-doped cobalt chromite increased in resistivity on exposure to air. Two mole per cent manganese doping resulted in stable 1000~ resistivities of 6 ohm-cm in air and 50 o h m -c m in H2-H20 mixtures having an oxygen partial pressure of 10 -15 atm. These resistivities are sufficiently low to make manganese-doped cobalt chromite useful for the fuel-cell battery application.The individual cells of a thin-film, high-temperature, fuel-cell battery using zirconia electrolytes are connected in series by a film of an electrically conductive oxide which joins the air and fuel electrodes of adjacent cells, as shown in Fig. 1. Previous work (1) established that cobalt chromite was a possible interconnection material for fuel cells operating at 1000~ Electrical characteristics were studied by C o p p e t al.f2), B r a d b u r n and Rigby (3), and Schmalzried (4). Resistivity at 1000~ in air varied by two orders of magnitude--2, 70, and 200 ohm-cm were reported respectively by these authors.Measurements of a 80~ thick film of undoped cobalt chromite separating the air and fuel atmospheres in the fuel-cell battery configuration gave an "effective double-atmosphere resistivity" of approximately 200 ohm-cm. This is four times higher than desired for practical fuel-cell batteries.In choosing dopants to enhance the electrical conductivity of cobalt chromite, the observation of Bradb u r n and Rigby (3), that the B cations in the normal spinel structure AB204 play an important role in determining the conductivity, was considered. Cobalt aluminate displays resistivities several orders of magnitude higher than both cobalt chromite and nickel chromite. Inspection of a model of spinel reveals possible conducting paths by way of 3d electron interchange between adjacent B ions because of their collinearity and small B-B separation. Both m a n g anese, which has a + 3 ionic radius of 0.62A and four 3d orbital electrons, and vanadium, having a + 3 ionic radius of 0.66A and two 3d orbital electrons, should be able to occupy chromium sites as the Cr +3 ionic radius is 0.64A and has three 3d orbital electrons. Either v a n a d i u m or manganese might be expected to introduce defects in the normal 3d electronic structure of cobalt chromite and hence enhance the conductivity. Based on these considerations, it was decided to employ manganese sesquioxide and v a n a d i u m sesquioxide as dopants to lower the resistivity of cobalt chromite. As discussed later, there is experimental evidence that some manganese actually substituted for cobalt....