Prediction of ash deposition characteristics under oxy-firing conditions helps to determine how retrofit to oxycombustion might affect boiler performance. To obtain data to help achieve this end, a novel temperature-controlled ash deposition probe was designed to collect temporally resolved deposit samples in a 100 kW rated down-flow test furnace system, firing a Powder River Basin coal. The rig was designed to represent practical units in terms of temperatures and particle and gas concentrations yet still be sufficiently well-defined to allow for controlled systematic studies. The deposit probe/furnace system, which is described in detail, was designed to segregate early "inside" deposits from the average deposits gathered over a period of time. Deposits were gathered under controlled conditions for oxidant input conditions of 50% O 2 /50% CO 2 (once through, with no flue gas recycle) and air. Effects of the deposit holding time, deposit probe temperature, and flue gas temperature at the probe location were investigated. Temporal segregation of deposits was achieved by physically separating deposits gathered on the horizontal probe surface into "inside" and "outside" deposits, where "inside" deposits approximated the initial deposit layers. Furthermore, results showed that deposits gathered over long times on the vertical surface of the probe were similar with respect to both composition and particle size distribution to the inside layer of the horizontal deposits, but different from the bulk horizontal deposits that have typically been reported in the literature. There were no significant effects of holding times greater than 1 h on bulk deposit compositions, although particle size within the deposit did appear to increase with time. There were also no significant differences between compositions of "outside" deposits from oxy-firing (OXY50) and those from air firing. "Inside" deposits from OXY50, however, contained higher Si and Fe and lower S and Na compared to those from air combustion. These results are interpreted in the light of available mechanisms. Tests in which only the deposit surface temperature was changed showed that the mass of deposits on the vertical surface parallel to the flow, shown to be representative of the "inside" deposits on the horizontal surface, was proportional to the temperature difference between the flue gas and the surface. This supported the hypothesis that the early layer was deposited largely by thermophoresis of small particles and not by Fickian or Brownian diffusion or impaction.

No abstract

Ash aerosol and ash deposit formation during oxy-coal combustion were explored through experiments in a self-sustained 100 kW rated down-fired oxy-fuel combustor. Inlet oxidant conditions consisted of 50% inlet oxygen with CO2 (hereafter denoted as OXY50 conditions). A Berner low pressure impactor (BLPI), a scanning mobility particle sizer (SMPS), and an aerodynamic particle sizer (APS) were used to obtain size segregated ash aerosol samples and to determine the particle size distributions (PSD). A novel surface temperature controlled ash deposition probe system that allowed inside and outside deposits to be separated was used to collect the ash deposits. The ash aerosol PSDs given by the BLPI and those produced by SMPS/APS were consistent with each other. Data suggested that oxy-coal combustion under these conditions did not change the formation mechanisms controlling the bulk ash aerosol composition, but it did increase the formation of ultra-fine particles initially formed through metal vaporization, due to increased vaporization of silicon at the higher combustion temperature. The smaller particles contained within the deposits had higher Si and lower Na and S concentrations under OXY50 conditions than for air combustion. Moreover, the ash aerosol composition for particle sizes less than 2.4 μm was related to the composition of the inside deposits. A higher Na in the ash aerosol resulted in higher Na in inside deposits with comparable absolute Na concentrations in both those aerosol particles and those inside deposits particles. The contribution of S and Si to the inside deposits showed that S in the vaporization modes together with Si in the ultrafine vaporization mode, contributed significantly to the composition of the inside deposits. These results provided direct evidence that prediction of the chemistry of the initial deposit layer (but not of the bulk deposits) required knowledge of the size segregated chemistry of the ash aerosol.

No abstract

No abstract