February 2026 Volume 8

SCIENTIFIC USE CASE Advanced Microstructural Evolution Prediction Through Multi-Scale Simulation Approaches for Nickel-Based Superalloys By Angela Haykal, PhD FORGING RESEARCH

Objective The primary scientific objective of this research is to develop and validate a comprehensive computational framework for predicting microstructural evolution during thermomechanical processing of nickel-based superalloys, specifically Inconel 718. This investigation aims to bridge the critical gap between computational efficiency and physical accuracy by implementing and comparing three distinct modeling approaches: phenomenological (JMAK), mean-field (NHM), and full-field (Level-Set) methods. The research seeks to establish optimal modeling strategies that enable accurate prediction of dynamic recrystallization (DRX), post-dynamic recrystallization (PDRX), grain growth phenomena, and the influence of second-phase particles (SPP) on microstructural evolution during hot metal forming processes. Background Scientific Context and Challenges The aerospace and power generation industries rely heavily on nickel-based superalloys, particularly Inconel 718, for critical components operating under extreme conditions. These materials must maintain exceptional mechanical properties at elevated temperatures while exhibiting resistance to creep, fatigue, and corrosion. The final microstructure, determined during thermomechanical processing, directly governs these properties. However, predicting microstructural evolution during complex forming operations presents significant scientific and computational challenges. The primary scientific challenge lies in accurately capturing the complex interplay between multiple physical phenomena occurring simultaneously during hot working processes. These include dynamic recovery, continuous and discontinuous dynamic recrystallization, grain boundary migration, nucleation events, and interactions with second-phase particles. Traditional experimental approaches for studying these phenomena are limited by the difficulty of real-time observation at processing temperatures (typically 940-1080°C for Inconel 718) and the prohibitive cost of extensive trial-and-error testing. Current State of Knowledge The field has evolved from purely empirical approaches to sophisticated physics-based models. The Johnson-Mehl-Avrami Kolmogorov (JMAK) equation, developed in the mid-20th century, provided the first mathematical framework for describing transformation kinetics. However, its assumptions of constant nucleation and growth rates, along with its inability to capture spatial heterogeneities, limit its applicability to complex industrial scenarios.

Recent advances in computational materials science have enabled the development of full-field models that explicitly resolve individual grains and their interactions. These models, while physically accurate, suffer from computational demands that can reach several hours for simulating a single material point, making them impractical for industrial-scale process optimization. The emergence of mean-field approaches represents an attempt to balance physical fidelity with computational efficiency. These models retain essential physics while using statistical representations of the microstructure, potentially offering a viable solution for industrial applications. Methodology Multi-Scale Modeling Framework The research employs a three-tier modeling approach integrated within the TRANSVALOR software suite: 1. Phenomenological Approach (JMAK Model in FORGE®) The JMAK model implements the classical Avrami equation to describe the overall kinetics of recrystallization: X(t) = 1 - e^(-b·t^n) Where X represents the recrystallized fraction, and b and n are empirical Avrami coefficients dependent on nucleation and growth rates. This model operates with negligible computational overhead during finite element simulations, making it suitable for rapid industrial assessments. The implementation captures global nucleation, growth, and impingement phenomena, producing characteristic sigmoidal transformation curves.

Figure1: Representative sigmoidal recrystallization curve obtained under isothermal conditions, consistent with the Avrami equation

FIA MAGAZINE | FEBRUARY 2026 66