The investigators conclude that: (1) Turbine erosion resistance was shown to be improved by a factor of 5 by varying the turbine design. Increasing the number of stages and increasing the mean radius reduces the peak predicted erosion rates for 2-D flows on the blade airfoil from values which are 6 times those of the vane to values of erosion which are comparable to those of the vane airfoils. (2) Turbine erosion was a strong function of airfoil shape depending on particle diameter. Different airfoil shapes for the same turbine operating condition resulted in a factor of 7 change in airfoil erosion for the smallest particles studied (5 micron). (3) Predicted erosion for the various turbines analyzed was a strong function of particle diameter and weaker function of particle density. (4) Three dimensional secondary flows were shown to cause increases in peak and average erosion on the vane and blade airfoils. Additionally, the interblade secondary flows and stationary outer case caused unique erosion patterns which were not obtainable with 2-D analyses. (5) Analysis of the results indicate that hot gas cleanup systems are necessary to achieve acceptable turbine life in direct-fired, coal-fueled systems. In addition, serious consequences arise when hot gas filter systems fail for even short time periods. For a complete failure of the filter system, a 0.030 in. thick corrosion-resistant protective coating on a turbine blade would be eroded at some locations within eight minutes.