| Abstract Scope |
Introduction: Process tubing coils in Olefin Furnaces contain gravity cast U-bend fittings from advanced high performance Fe-Cr-Ni alloys. These fittings are exposed to steep temperature variations during their service life, leading to premature failures through thermal fatigue and thermal shock. The internal diameter (ID) of tube coils is subjected to carburization by process fluids, while the outer diameter (OD) is exposed to oxidation from furnace flue gas. With the combination of these extreme conditions, furnace coil failures were experienced after only two years of the projected six-year service life. Additive Manufacturing (AM) of metallic alloys provides an opportunity to improve service life of these U-bend fittings through the production of functionally graded materials (FGMs). Functionally graded metallic alloys contain compositional gradients tailored to the demands of the service conditions. Since producing FGMs using AM is a relatively new concept, design approaches need to be established to fully develop this technology to address challenges in tailoring the compositional, microstructural, and property gradients while avoiding formation of harmful microconstituents. The goal of this project is to develop alloy design and additive manufacturing process optimization approaches to improve service life in U-bend fittings of Olefin furnaces. Experimental Procedure: The current phase of the project has focused on alloy design, validation, and testing. The FGM alloy design process is intended to address carburization on the inner diameter (ID), oxidation on the outer diameter (OD), and thermal fatigue/creep properties in the through thickness (core) of the U-bend fittings. These design considerations will create a compositional gradient from the OD-core and core-ID. This design process is being carried out using CALPHAD software with thermodynamic and kinetic databases in Thermo-Calc. Thermo-Calc projects are being run in Python scripts utilizing the TC-Python library. The scripts are used to run several calculations to dilute three alloy compositions from OD-core and core-ID to determine the expected microstructure. Scheil non-equilibrium simulations are run to predict microstructure after welding (AM), and equilibrium simulations are run to predict microstructure during service. The results of Thermo-Calc simulations will be validated through production of button melts of select compositions, experimental characterization of the solidification behavior and phase transformations using single sensor differential thermal analysis (SS DTA), and advanced metallurgical characterization. Results and Discussion: The current FGM alloy design has a core composition of Fe-37Cr-45Ni-0.4C-1Nb-1Mn-0.6Si with potential microalloying additions of W, Mo, Ti, or Zr based on results from previous work on high performance Fe-Ni-Cr alloys. Up to 2wt.% Si additions are added to the OD composition to address oxidation, and up to 4wt.% Al additions are added to the ID composition to address carburization. Using TC-Python scripts, the resulting microstructural gradients from OD-core and core-ID have been modeled in a novel twin pseudo-binary phase diagram. The x-axis of the plot varies dilution percentage from -100% (OD composition with 2wt.% Si) to 0% (core composition) to 100% (ID composition with 4wt.% Al). Twin pseudo-binary phase diagram plots are generated for both solidification and service using the Scheil and equilibrium models respectively. Compositional data is gathered in an automated process as well to verify the existence of phases of interest, and to understand the progression of certain phase transformations as alloying elements are added. The phases and carbides present in the twin pseudo-binary diagrams generated have been consistent with previous work and literature, but need experimental validation with button melts to understand how to improve the model. The results from the twin-pseudo binary phase diagrams tie into the AM process design development as the presence of carbides and grain interior strengthening phases must be managed to optimize creep rupture and thermal fatigue. TC-Prisma and TC-Dictra will be employed to understand how phases/carbides may progress during fabrication and at service conditions starting from a Scheil solidification (Arc-DED) microstructure. Conclusions: Additive manufacturing of functionally graded high performance Fe-Ni-Cr alloys has the potential to improve service life in U-bend furnace coil fittings in Olefin furnaces. Compositional gradients can be tailored in FGMs to address harsh service environments that differ from the inner to outer diameters of these fittings. Alloy design procedures have been established to create twin pseudo-binary phase diagrams to model compositional gradients from OD to core to ID within the fitting wall. Button melts, SS DTA, and advanced metallurgical characterization will validate these results. Precipitation, diffusion, and finite element models will be used to understand carbide/phase progression in the FGM alloy system and guide selection of welding parameters for Arc-DED processing. Given Olefin furnaces produce hundreds of megatons of polymers annually, improving service life and performance of furnace coils can save millions of dollars in downtime. As a result, the alloy design being developed for FGMs in these furnace applications is critical and can be built upon for other applications in the future. Keywords: Additive Manufacturing, Functionally Graded Materials, CALPHAD, Alloy Design, Computational Modeling. |