Ash Formation Involving Siderite Performance of PF Fired Plants

The mineral form of the iron in coal has a fundamental influence on the observed extent and severity of the ash deposition problems encountered during combustion. There has been extensive research on iron in the form of pyrite (FeS2), the main mode of occurrence for iron minerals in Northern Hemisphere bituminous coals.

Australian bituminous coals generally contain iron in the form of siderite, (FeCO3 with varying levels of Mg, Ca and MnCO3), not pyrite. Consequently there have been difficulties in applying the conventional indices to Australian coals. Research into siderite as the main iron-bearing mineral is sparse.

Theoretical and experimental studies undertaken in this Project investigated the thermal behaviour of pure and substituted siderites along with the effect of siderite associated with clay and quartz in the coal. The effect of the siderite with and without its associated minerals as excluded and included particles was also studied.

The theoretical study involved thermodynamic equilibrium calculations and an analysis of the relevant phase diagrams for the siderite system. Although limitations in the application of the thermodynamic calculations were apparent, it was concluded that the calculations could provide an indication of the expected behaviour for the siderite upon combustion.

The most important factor in relation to production of molten or sticky particles was identified as being the oxidation state for the iron in the combustion residue. Iron present as 2+ forms oxides with lower solidus and liquidus temperatures when compared to iron as 3+.

Investigation of vaporisation of siderite shows that only the iron component would be expected to vaporise, and then only for the included siderite. This is primarily due to the higher temperatures and more reducing environment experienced during combustion by included minerals when compared to excluded minerals.

Combustion experiments were done using a drop tube furnace at temperatures of 1100oC and 1600oC under oxidising conditions. The resultant residues were collected for analysis. In a separate set of experiments the flame transformations of an excluded siderite sample were investigated using a flat flame furnace at temperatures of 1300oC and 1470oC with varying oxygen concentrations for the flame.

The experimental investigations showed that substituted impurities present in the parent siderite influenced the liquidus behaviour for excluded siderite residues. The level of substitution and observed extent of melting for siderite combustion residues correlated well with the range predicted by theory. This resulted in the ranking of siderite compositional types in order of decreasing slagging tendency for combustion products as follows: pure siderite, calcium siderite, manganese siderite, magnesium calcium siderite, and magnesium siderite.

A comparison between the thermal behaviour of siderite as observed in this Project and the published behaviour of pyrite shows that each mineral produces similar final oxidation products.

A case study supported this project. Deposit samples were collected from the Callide B power station which fires a siderite-bearing coal. Samples of deposit were collected from a number of points in the furnace from the burner region through to the economiser. It is known that the mineralogy of the coals fired at the power station is dominated by siderite, kaolinite and quartz.

Deposits collected from the burner regions and furnace walls have a composition that reflects the bulk ash chemistry. With increasing distance from the burner there is an increase in the bulk iron content of the deposits. Analysis of the deposits and combustion experiments using two of the coals fired at the power station show there is a systematic increase in the iron content of the deposits with distance from the burner. This corresponds to decreasing gas temperature through the furnace, correlating with increasing iron content of the "sticky" particles. This is what was expected from the phase diagrams.

The deposit grows in most parts of the furnace by a mechanism of inertial impaction of sticky iron and iron alumino-silicate particles. Deposits on the economiser appear to initiate and accumulate by thermophoresis of fine alumino-silicates (major) and iron oxide (minor) particles. Thermophoresis is also operative in the superheater region.

McLennon32 has developed a slagging index specifically for predicting deposition from sticky particles derived from iron minerals and their interaction with clay minerals under purely oxidising or reducing conditions. The index was tested using data from this Project and found to be valid in the radiant section of the furnace but not in the convective section. This inadequacy was addressed through modifications to the assumptions used for the calculation of the index, in particular the treatment of iron as being solely 3+ for oxidising conditions.

The modified index now provides a basis for predicting the occurrences of sticky particles derived from iron and clay minerals in both the radiant and convective sections of the furnace.

The limitations of the modified index are discussed in the report.