The use of crude oil as a backup fuel in gas turbines remains an economical option for power plants in petroleum-rich countries like Saudi Arabia. However, operating gas turbines with crude oil can corrode their Hot-Gas-Path (HGP) components due to the formation of vanadium oxides and alkali metal vanadates, when preventative measures are not employed. While the pre-combustion treatment of crude oil fuel is effective at reducing the content of sodium (Na) and potassium (K) in the fuel and, therefore, their resulting combustion products, any pre-combustion removal of vanadium has remained unviable on a commercial scale. Typically, a best practice limit of 0.5 ppmw vanadium is imposed on liquid fuels when inhibitors are not employed, and the total alkali content is limited to 1.0 ppmw. To evaluate the effects of incremental exceedances of such limits on the structural materials and coatings of gas turbines, a high-pressure hot-corrosion test rig was built at King Abdullah University of Science and Technology (KAUST). Arabian Extra Light (AXL) crude oil with a vanadium content of up to 2 ppm was chosen as the test fuel. This burner rig was designed to accommodate high temperatures (~1173 K/ 900 °C) and the combustion of unrefined crude oils at an elevated pressure of up to 15 bar. Highly reduced flow velocities (~1 m/s) were employed to deconvolute erosional effects from corrosion. This experiment exposed the tested materials to combustion products under these conditions for extended periods (up to 500 hours). The development of the rig included several design iterations and changes to the configuration of various sub-systems, owing to some challenges in maintaining stable combustion at low flow rates, management of coke/ash formation from the burning of crude oil, and adequate control over the pressure and temperature during extended experimental campaigns. The present publication summarizes the design and commissioning of this high-pressure hot-corrosion rig facility, and the successful demonstration of several long duration (400+ hour) experimental runs with AXL crude oil, where the environment of the sample section with test coupons was maintained at 5.5 bar and 1173 K. The pressure and temperature inside the sample section of the rig were continuously monitored. The batch of material coupons was removed from the sample section after the runs and sent for corrosion analysis. Deposits resulting from the AXL combustion and fouling were collected from several sections of the rig for analytical analysis. The fuel wash procedure, the design of components in the burner rig, and best practices of operation were also documented.
KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: The authors would like to thank Dr. Anthony Dean for numerous technical discussions on the rig design and his strong support for the project. We wish to thank Mr. Umit Gorcum and Ms. Chaoqin Chen for lab operations and analytical chemistry support, respectively. The research reported in this publication was supported by joint funding from General Electric (GE) and KAUST.