TAILORING THE PROPERTIES OF MULTI-PHASE TITANIUM THROUGH THE USE OF IMAGING MODALITY AND MACHINE LEARNING - Thesis Summary

Aug 6, 2026

Upload by Himbo

High-strength alloys that combine strength, ductility, hardness, and toughness are essential for extreme-environment design, yet strength and fracture toughness often trade off due to opposing mechanisms. This work targets a favorable exception in multiphase alloys with mixed-phase microstructures. Machine learning is used to identify and correlate critical microstructural features in a Ti-10V-2Fe-3Al alloy reported to show high strength and fracture toughness. Metallurgical specimens are characterized using SEM, EDS, and EBSD to build a training dataset for a CNN model that segments and classifies microstructures and links them to processing and properties.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages104

Reading Time4 h 22 min

AccessFree to download

Acknowledgment

Abstract

List of Tables

List of Figures

Dedication

Chapter One: Introduction

Chapter Two: Literature Review

Titanium

Convolutional Neural Networks (CNNs)

Chapter Three: Methodology

Titanium Sample Preparation

SEM

Imaging Modality

Chapter Four: Results and Discussion

Introduction

As-Quenched Microstructure

TAILORING THE PROPERTIES OF MULTI-PHASE TITANIUM THROUGH THE USE OF IMAGING MODALITY AND MACHINE LEARNING By GUNNAR BENJAMEN BLASCHKE A thesis submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN MATERIALS SCIENCE AND ENGINEERING WASHINGTON STATE UNIVERSITY School of Mechanical and Materials Engineering MAY 2023 © Copyright by GUNNAR BENJAMEN BLASCHKE, 2023 All Rights Reserved © Copyright by GUNNAR BENJAMEN BLASCHKE, 2023 All Rights Reserved To the Faculty of Washington State University: The members of the Committee appointed to examine the thesis of GUNNAR BENJAMEN BLASCHKE find it satisfactory and recommend that it be accepted. | Field, David P., Ph.D., Chair | | --- | | Merriman, Colin C., Ph.D. | | Boddeti, Narasimha, Ph.D. | Beckman, Scott P., Ph.D. ACKNOWLEDGMENT I want to first acknowledge the financial support from Idaho National Laboratories for funding this Master’s thesis and giving me the opportunity to continue my education in Materials Science and Engineering. Also I must mention the professional experience I gained as a materials science engineer in the form of an internship in the Summer of 2022 where I met countless individuals such as Dr. Tom Mason and Cody Gibson that supported me and helped me achieve my ambitions. Thank you to my committee members, Dr. Narasimha Boddeti and Dr. Scott Beckman. I appreciate the support you have given me as my committee members as well as the challenging courses, such as Topology Optimization with Dr. Narasimha Boddeti and Mechanical Behavior of Materials with Dr. Scott Beckman . Thank you to my Chair, Dr. David Field, for taking me in as a student. I am grateful for the opportunity to work with you and to have learned from you. I will always appreciate your kindness and support that you have given me throughout my time as a graduate student at WSU. Along with all the countless times you had to reexplain to me how SEMs and EBSD systems work and showing me numerous times how to best optimize an SEM for imaging and EDS acquisition. What I have learned from you will be invaluable to me throughout my career. Thank you to my advisor, Dr. Colin Merriman, for choosing me as the student to work on this project. My first class with you was MSE 201 in the Fall 2018 semester, which was also the worst grade I have ever received in my collegiate career. Your classes have always been difficult and rewarding which lead me to decide to attend graduate school after working with you during my undergraduate studies. I will always appreciate the knowledge and advice you have given me throughout my undergraduate and graduate studies. TAILORING THE PROPERTIES OF MULTI-PHASE TITANIUM THROUGH THE USE OF IMAGING MODALITY AND MACHINE LEARNING Abstract by Gunnar Benjamen Blaschke, M.S. Washington State University May 2023 Chair: David P. Field High strength alloys with good ductility, hardness, and toughness are needed to meet rigorous design requirements for extreme environments. However, metals rarely exhibit both high strength and good fracture toughness as the underlying mechanisms work in opposition. An exception can be found in multiphase alloys that form microstructures of mixed phases. Machine learning techniques can be used to identify and correlate critical microstructural features in a Ti-10V-2Fe-3Al alloy that is reported to exhibit high strength and fracture toughness. Metallurgical specimens are characterized in a imaging modality manor using scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), and electron backscatter diffraction (EBSD) . Microstructural features are analyzed and will be used to construct a training dataset for a convolutional neural network (CNN) model to gain the ability to segment and classify the microstructures ofTi 10V-2Fe-3Al and relate microstructure to processing and properties. TABLE OF CONTENTS Page ACKNOWLEDGMENT.........................

FAQs

What material system and properties does the thesis focus on?

The thesis focuses on multiphase titanium alloys, specifically a Ti-10V-2Fe-3Al alloy, aiming to connect microstructure with high strength and fracture toughness while considering properties relevant to extreme environments.

Which imaging and characterization modalities are used?

Microstructural characterization is performed using scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), and electron backscatter diffraction (EBSD) to capture features for analysis and model training.

How does machine learning contribute to the study?

Machine learning methods build a training dataset from analyzed microstructural features and use a convolutional neural network (CNN) to segment and classify microstructures, enabling relationships between microstructure, processing, and properties.

Aug 6, 2026

Upload by Himbo