Computation Assisted Materials Development for Improved Corrosion Resistance: Residual Lifetime Assessment and Multiscale Modeling Methods
Program Organizers: Rishi Pillai, Oak Ridge National Laboratory; Christopher Taylor, Dnv Gl
Monday 8:00 AM
November 2, 2020
Room: Virtual Meeting Room 29
Location: MS&T Virtual
8:00 AM
Introductory Comments: Computation Assisted Materials Development for Improved Corrosion Resistance: Rishi Pillai1; 1Oak Ridge National Laboratory
Introductory Comment
8:05 AM Invited
Assessing High Temperature Durability for Long-term Applications: Bruce Pint1; 1Oak Ridge National Laboratory
To increase efficiency, a wide range of energy-related applications are pushing components to higher temperatures where environmental degradation can be life limiting. Unlike mechanical properties such as creep, there is no simple degradation parameter to capture, for example, the time-temperature-thickness limitations of candidate alloys. Also, the degradation mechanism varies greatly depending on the environment. For complex environments such as molten salts and liquid metals, the first assessment step is determining the degradation mechanism. For simpler high temperature gas reactions, the mechanisms are better understood and the issue becomes quantifying the degradation rate in a manner useful to component designers. Modeling is rapidly evolving and more sophisticated computational methods are now being adopted. Case studies will be presented from various projects to illustrate the stages of compatibility assessments from identifying mechanisms to predicting lifetimes. Research sponsored by the US DOE, Offices of Fossil Energy and EERE.
8:35 AM
Development of a Multiscale Corrosion Model for Valve Steels in a Gasoline Engine Environment: Michael Tonks1; Xueyang Wu1; Simon Phillpot1; Robert Ullberg1; Iman Abdallah2; Adrien Couet2; John Perepezko2; Mark Carroll3; Wen Jiang4; 1University of Florida; 2University of Wisconsin-Madison; 3Tenneco; 4Idaho National Laboratory
As engines are pushed to higher temperatures, valve steels undergo microstructure evolution that sensitizes it to corrosion and can result in premature failure. The goal of this project is to develop the Stainless Steel Alloy Corrosion (SStAC) tool for modeling corrosion of valve steels in an engine environment at temperatures up to 800 C in 1D, 2D or 3D. The tool is being implemented using the open source MOOSE framework, coupling a corrosion model (including the impact of the microstructure and alloy composition) with mechanics and thermal transport. Simulations at the atomic and mesoscales are being used to obtain parameters for the model that include the impact of microstructure and alloy composition. The tool will be validated against new data obtained using laboratory and engine tests. The completed SStAC tool will be used to optimize existing valve steels to improve their corrosion resistance without significantly increasing cost.
8:55 AM
High Temperature Oxidation Lifetime Modeling of FeCr and NiCr Foils in Water Vapor: Marie Romedenne1; Rishi Pillai1; Sebastien Dryepondt1; Bruce Pint1; 1Oak Ridge National Laboratory
Fe-based alloys are typically employed in heat exchanger components for combined heat and power generation systems. However, operating temperatures above 650 °C are desired to improve efficiency of these applications requiring the use of Ni-based alloys (e.g. alloy 625), which typically form a protective Cr2O3 scale. The ability to maintain formation of a protective Cr2O3 scale depends on an interplay of different processes (loss in wall thickness, subsurface Cr depletion, Cr2O3 volatilization). Foil specimens of Fe- and Ni-based alloys were oxidized in dry and wet air for up to 30,000 h at 650, 700 and 800 °C. Recuperator foils exposed for up to 100,000 h at about 650 °C in real engine conditions were also analyzed. The impact of composition, temperature and gas flow rates on Cr depletion and loss of wall thickness of the foils will be discussed with an oxidation kinetics model and coupled thermodynamic-kinetic diffusion calculations.
9:15 AM
Simulation of Dissolution of \Gamma\Prime Precipitates in Ni-base Superalloys during Oxidation: Taiwu Yu1; Christopher Taylor2; Babu Viswanathan1; Brett Tossey2; Yunzhi Wang1; 1Ohio State University; 2DNV GL
Slow-growing oxides like Al2O3 and Cr2O3 form at the surface of Ni-base superalloy heat exchangers during high temperature exposure, which alters the microstructure near the surface due to the depletion of the oxidizing elements. In this presentation we show how to combine density functional theory and DICTRA simulation to calculate concentration variation of the oxidizing elements and dissolution of precipitates of Alloy 282. By DICTRA we predict quantitatively concentration profiles and the volume fraction of precipitates. The changes of average particle size as well as size distribution of the precipitates are conducted through Panprecipitation module in Pandat. Furthermore, a parallel simulation is conducted based on a phase field model to show more details on the change of precipitate microstructure during the dissolution. The simulation results are compared with experimental characterization and they show a good agreement. This work illustrates the usefulness of applying the CALPHAD methods in designing oxidation-resistance alloys.
9:35 AM
Machine Learning to Predict Cyclic Oxidation of NiCr-based alloys: Jian Peng1; Marie Romedenne1; Rishi Pillai1; Govindarajan Muralidharan1; Bruce Pint1; J. Haynes1; Dongwon Shin1; 1Oak Ridge National Laboratory
Machine learning (ML) can offer many advantages in predicting material properties over traditional materials development methods based solely on limited experimental investigations or physical-based simulations with respect to cost, risk, and time. However, thus far limited efforts have been published to predict alloy oxidation resistance via ML. In this presentation, we compare two different oxidation models (a simple parabolic law and a statistical cyclic-oxidation model) to represent the high-temperature oxidation of NiCr-based alloys in dry- and wet-air within the context of data analytics. We successfully trained ML models with highly ranked key features identified from the extensive correlation analysis. The performance of selected oxidation models in ML was compared and discussed. This research was sponsored by the Department of Energy, Vehicle Technologies Office, Propulsion Materials Program.
9:55 AM Invited
Tailoring the Microstructure of Eutectoid Steels during Annealing for Improved Corrosion Resistance: Insights from Phase-field Simulations: Kumar Ankit1; 1Arizona State University
A key issue in the design of corrosion-resistant eutectoid steels is its pearlitic microstructure where the alternating lamellae of ferrite (anode) and cementite (cathode) phases constitute a galvanic couple resulting in accelerated degradation of structural components. Recent investigations suggest that altering the morphology of pearlitic colonies, in particular, the inter-lamellar spacing and ferrite-cementite interfacial area ratio, through the isothermal annealing process, is an effective means of enhancing corrosion resistance. Therefore, in order to tailor the microstructure, a comprehensive understanding of the operative diffusional mechanisms during processing is warranted. In this talk, I will discuss a CALPHAD-informed phase-field model that can be used to simulate the cooperative and non-cooperative evolution of eutectoid phases in multicomponent steels during the inter-critical and sub-critical annealing processes. Implications of annealing time and temperature on the characteristics of evolving microstructure will be discussed.
10:25 AM
Metal-Oxide Bond-energy Models for Bond Energies of Alloy Oxides in Corrosion: Szu-Chia Chien1; Wolfgang Windl1; Gerald Frankel1; 1The Ohio State University
With the goal of developing a scienced-based approach for alloy design in corrosion applications, we herein introduce a metal-oxide bond-energy model for alloy oxides based on pure-phase bond energies and bond synergy factors that describe the effect of alloying on the bond energy between cations and oxygen. The bond energy model is parameterized for a series of binary cation-alloy oxides using density-functional theory (DFT) energies and is shown to be directly transferable to multi-component alloy oxides. We parameterized the model for alloy oxide energies in the corundum and rock salt structures using most common metal cations in stainless steels. The calculated bond energy values give sensible results in comparison to common experience, including the role of Cr in the passive-layer on Fe-Ni-Cr alloys for corrosion applications. Additionally, the bond synergy factors give insights into the mutual strengthening and weakening effects of alloying on cation-oxygen bonds.