Ceramic Energy: Advanced Si3N4 Components for Microturbines
One of the overall goals of the U.S. Department of Energy (DOE) and industry is to improve energy efficiency and reduce emissions in microturbine systems for distributed energy. Advanced ceramics are considered enabling materials to achieve these goals. To that end, researchers at Saint-Gobain's Northboro R&D Center in Northboro, Mass., are in Phase III of a program to develop hot section silicon nitride (Si3N4) components for advanced microturbine systems under a subcontract with Oak Ridge National Laboratory (ORNL).
Si3N4 offers excellent thermo-mechanical properties, including high strength and creep resistance at elevated temperatures, a low coefficient of thermal expansion, and a high thermal shock resistance. These properties make the material a suitable candidate for high-temperature gas turbine components such as integrally bladed rotors (IBR), vanes and combustor liners. The goal of this program is to optimize NT154 high-temperature Si3N4 for elevated temperature hot section (1100-1300°C) microturbine applications for the distributed generation of power (approximately 25-500 kW). The successful use of ceramics as an enabling material for achieving higher temperature microturbine operation will result in improved efficiency, reduced fuel consumption and lower emissions.
Saint-Gobain and ORNL researchers have enjoyed a close and productive collaboration on this program. The technical effort has involved a three-pronged approach-NT154 material development and improvement; development of a recession (weight loss)-resistant solution for Si3N4 in a high-temperature (1100-1300°C), high-gas-velocity (20-50 m/sec), humid (10% H2O) gas turbine environment; and the development of a net-shaped forming technique for rotors and vanes.
Material DevelopmentThe NT154 Si3N4 is a glass-encapsulated, hot isostatic pressed (HIPed) composition containing 4 wt.% yttria. Although the composition was originally developed in the late 1980s, it had been dormant for approximately eight years before the start of the current DOE contract in 2002. Under this contract, the material development effort was initially focused on reestablishing the NT154 Si3N4 process and properties-a milestone that was accomplished and verified at ORNL in 2003 (see Figure 1).
In the ongoing activities to improve the properties of NT154, a major effort was directed toward enhancing the material's as-processed (AP) (or as-fired) strength. Typically, the AP surface properties can be inferior to the properties of the bulk material because of surface defects and reaction layers that can form during manufacturing. The aerofoil sections of net-shaped turbine components are not subjected to dense machining due to the high costs involved. In 2004, researchers achieved a significant programmatic milestone by developing a new densification process that improved the AP strength of NT154 Si3N4 surfaces more than 40% compared to the baseline values, with modulus of rupture (MOR) values exceeding 700 MPa.
Recession ControlMaterial recession due to volatilization in humid gas turbine environments has been a continuing problem for Si3N4. Through detailed thermodynamic computations, Saint-Gobain researchers have been able to define novel environmental barrier coating (EBC) compositions that offer a promising solution. Monoliths of the candidate coating materials were made and evaluated for suitability in a coefficient of thermal expansion (CTE) match with Si3N4, as well as for durability (recession resistance) in a high-temperature (1250°C), high-water-vapor-pressure environment. The candidate EBC compositions exhibited a stable microstructure and negligible recession or weight loss at 500 hours of testing in a high-pressure Keiser Rig* at ORNL (see Figure 2). Additional recession testing is in progress.
*The Keiser Rig is a high-temperature, high-pressure exposure facility developed by Jim Keiser and Irv Federer in the early 1990s to examine corrosion of candidate heat exchanger materials.
Net-Shaped FormingA major hurdle in commercializing complex-shaped NT154 Si3N4 components for the turbine industry is the lack of a proven forming technique. The hard and brittle nature of high-performance ceramic materials makes final dense machining difficult and costly. Ideally, a net-shaped forming technique could be used before densification so that the final component will require minimal or no finishing.
Past efforts using various casting approaches have presented drawbacks such as low yield, degradation of material properties and an inability to hold part tolerances. These issues were magnified as parts became larger.
Under the DOE contract, researchers investigated green machining as a net-shaped forming technique. The experimentation involved evaluating variables related to the green blank properties, cutting tools and machining parameters. Through a significant effort, researchers were able to develop and optimize a novel green machining process.
The optimized process was used to manufacture several axial and radial designed rotors for demonstration purposes. Excellent concentricity of 0.004 in. (within a part) and minimal part-to-part variation were achieved. Additionally, with the help of the new densification process mentioned previously, average surface roughness values as low as 30 æ-in have been attained on the AP surfaces. These promising results have verified the potential of the new green machining process for this application.
Two of the radial rotors were delivered to ORNL for mechanical property evaluation. Disks core-drilled from the blade and hub regions were tested in biaxial flexure to establish the AP and bulk strength of these components. Fully machined biaxial flexure discs cut from the hub section of the rotors were also tested. These AP and machined biaxial strength values from the rotors compare well with the data generated using MOR bars machined from tiles, suggesting that the mechanical property data generated from test bars can be reproduced in large net-shaped components.
An additional radial rotor has been spin-tested at room temperature to 97,000 rpm (the design speed) without failure. Spin testing above 100,000 rpm is planned as soon as a new hub assembly design is completed.
Further EvaluationSaint-Gobain plans to supply prototype NT154 Si3N4 rotors during 2005 and 2006 for evaluation in the DOE advanced microturbine program. Also in 2005 and 2006, Saint-Gobain intends to provide NT154 Si3N4 hot section components to engine builders for other potential applications, such as small military turbine engines and aircraft auxiliary power units.
Authors' AcknowledgementsThe authors would like to acknowledge funding by the U.S. DOE Distributed Energy Program and Oak Ridge National Laboratory (managed by UT-Battelle, LLC) under Prime Contract No. DE-AC05-00OR22725 with the DOE.
For more information about the NT154 Si3N4 components, contact Saint-Gobain High-Performance Materials, Northboro Research & Development Center, 9 Goddard Rd., Northboro, MA 01532; (508) 351-7815; fax (508) 351-7760; e-mail Robert.H.Licht@saint-gobain.com; or visit www.saint-gobain.com/us/businesses/ceramics.html.
More information about Oak Ridge National Laboratory can be found at www.ornl.gov.