450 48 17MB
English Pages xvi, 465 Seiten: Illustrationen, Diagramme [483] Year 2019
Additive Manufacturing for the Aerospace Industry
This page intentionally left blank
Additive Manufacturing for the Aerospace Industry Edited by
Francis Froes Light Metals Industry, Tacoma, WA, United States Advanced Materials Industries, Tacoma, WA, United States
Rodney Boyer RBTi Consulting, Bellevue, WA, United States School of Materials Science and Engineering, University of Shanghai for Science and Technology, Shanghai, P.R. China
Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States Copyright © 2019 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN: 978-0-12-814062-8 For Information on all Elsevier publications visit our website at https://www.elsevier.com/books-and-journals
Publisher: Matthew Deans Acquisition Editor: Christina Gifford Editorial Project Manager: Andrae Akeh Production Project Manager: Kamesh Ramajogi Cover Designer: Matthew Limbert Typeset by MPS Limited, Chennai, India
Contents
List of Contributors 1
2
3
Introduction to aerospace materials requirements and the role of additive manufacturing Francis Froes, Rodney Boyer and B. Dutta 1.1 Aerospace materials and their requirements 1.2 Additive manufacturing 1.3 Additive manufacturing fabrication of various types of materials 1.4 Contents of this book References Review of additive manufacturing technologies and applications in the aerospace industry Joel C. Najmon, Sajjad Raeisi and Andres Tovar 2.1 Aerospace requirements and opportunities for additive manufacturing 2.1.1 Design requirements 2.1.2 Manufacturing capabilities and benefits 2.2 Additive manufacturing technologies 2.2.1 Additive metal technologies 2.2.2 Additive nonmetal technologies 2.3 Additive manufacturing applications 2.3.1 Direct digital manufacturing 2.3.2 Rapid tooling 2.3.3 Rapid prototyping 2.3.4 Repair 2.4 Challenges and potential future applications 2.4.1 Challenges 2.4.2 Potential future applications References Qualification and certification of metal additive manufactured hardware for aerospace applications Richard Russell, Douglas Wells, Jess Waller, Behrang Poorganji, Eric Ott, Tsuyoshi Nakagawa, Hector Sandoval, Nima Shamsaei and Mohsen Seifi 3.1 Introduction
xiii
1 1 1 3 6 6
7
7 8 10 12 12 14 17 18 19 19 20 22 22 23 24
33
33
vi
Contents
3.2 3.3
Special considerations for fracture-critical hardware Current qualification and certification state-of-the-art and gap analysis 3.3.1 Standardization gaps related to qualification and certification 3.3.2 Recent directions in qualification, certification, and quality control for additive manufacturing 3.4 Industry qualification and certification approaches 3.4.1 General Electric qualification and certification approach 3.4.2 Lockheed Martin qualification and certification approach 3.5 Government agency approaches 3.5.1 National aeronautics and space administration qualification and certification approach 3.5.2 Federal Aviation Administration qualification and certification approach 3.6 Summary and recommendations Acknowledgments References
4
5
Design for metal additive manufacturing for aerospace applications Manish Kamal and Gregory Rizza 4.1 Introduction 4.2 Methods and approaches 4.2.1 Topological optimization 4.2.2 Part consolidation 4.2.3 Part integration and repair 4.2.4 Other techniques 4.3 Process aspects of design 4.3.1 Part performance 4.3.2 Part quality 4.3.3 Part evaluation: in-situ and after process nondestructive evaluation (NDE) 4.3.4 Post processing 4.4 Cost considerations 4.5 Product and process design tools 4.5.1 Additive manufacturing design software 4.5.2 Additive manufacturing process software 4.6 Conclusions Acknowledgments References Structure formation in A.M. processes of Titanium and Ni-base alloys I.S. Polkin, S.V. Skvortsova, G.A. Turichin and M.B. Novikova 5.1 Evaluation of the structure of powder particles of different sizes
37 39 39 40 43 43 49 51 51 60 61 62 62 67 67 68 68 71 72 73 74 74 78 79 80 81 82 83 83 84 84 85
87 89
Contents
A dependence of the microstructure of powder particles in the initial state on their size 5.3 Determination of changes in the structure using samples produced by different additive technologies 5.4 Testing of mechanical properties of samples of parts produced by direct metal deposition and selective laser melting 5.5 Conclusions Acknowledgement Reference Further reading
vii
5.2
6
7
Measurement of powder characteristics and quality for additive manufacturing in aerospace alloys Thomas F. Murphy and Christopher T. Schade 6.1 Introduction 6.2 Quality control measurements 6.2.1 Particle size and distribution 6.2.2 Apparent density and flow 6.2.3 Tap density 6.2.4 Moisture analysis 6.2.5 Inclusion analysis 6.2.6 Shape factor 6.2.7 Porosity 6.3 Advanced metallographic techniques 6.3.1 Background 6.3.2 Metallographic sample preparation 6.3.3 Light optical microscopy—automated image analysis 6.3.4 Shape and texture analysis 6.3.5 Microstructural analysis 6.3.6 Chemical analysis References The processing and heat treatment of selective laser melted Al-7Si-0.6Mg alloy Jeremy H. Rao, Paul Rometsch, Xinhua Wu and Chris H.J. Davies 7.1 Selective laser melted Al alloy A357 7.1.1 Process control 7.1.2 Density optimization 7.2 Post selective laser melting heat treatment 7.2.1 Tensile properties 7.2.2 Fractography 7.3 Refinement of laser melting and postprocessing parameters 7.3.1 The selection of optimized heat treatment parameters 7.3.2 The selection of optimized laser parameters 7.3.3 Tensile properties 7.3.4 Fractography
92 93 95 97 97 97 98
99 99 99 99 101 105 106 108 110 111 113 114 117 122 124 135 136 140
143 143 143 145 150 150 151 153 153 153 155 156
viii
Contents
7.4 Conclusions Acknowledgments References 8
9
10
11
Superalloys, powders, process monitoring in additive manufacturing Kevin Minet, Ankit Saharan, Anja Loesser and Niko Raitanen 8.1 Applications: materials in gas turbines 8.2 Material and processing challenges in additive manufacturing of superalloys and different approaches for solutions 8.2.1 Challenges with additive manufacturing—theory on weldability issues 8.2.2 Solidification cracking 8.2.3 Liquation cracking (HAZ) 8.2.4 Strain-age cracking 8.2.5 Ductility dip cracking 8.3 Powder material properties 8.4 Process monitoring 8.4.1 Quality assurance in AM 8.4.2 Challenges with tradition postprocess inspection techniques 8.4.3 Novel QA approaches 8.5 Sensor types for in situ process monitoring 8.5.1 Optical in situ process monitoring systems for AM 8.6 Quality assurance tie-in 8.7 Defect data correlation automated, future closed-loop control possibilities References Fusion and/or solid state additive manufacturing for aerospace applications James C. Withers 9.1 Introduction 9.2 Experimental examples References Further reading
157 158 158
163 164 169 169 170 171 172 173 174 177 177 179 179 179 179 181 182 185
187 187 198 210 211
Profile electron beam 3D metal printing Dmytro Kovalchuk and Orest Ivasishin Summary References
213
Additive manufacturing of titanium aluminides Wei Chen and Zhiqiang Li 11.1 Applications of TiAl 11.2 Fundamentals of TiAl
235
231 232
235 236
Contents
11.3
Processings of TiAl 11.3.1 Casting 11.3.2 Wrought processing 11.3.3 Powder metallurgy 11.3.4 Additive manufacturing 11.4 Laser metal deposition of TiAl 11.5 Selective laser melting of TiAl 11.6 Electron beam melting of TiAl 11.7 Summary and prospects References Further reading 12
13
Aerospace applications of the SLM process of functional and functional graded metal matrix composites based on NiCr superalloys Shishkovsky Igor 12.1 Introduction 12.2 Metal matrix composites fabrication via additive technologies 12.3 Metal matrix composites on the nickel alloy based 12.4 Methods and materials 12.5 Results and discussion 12.6 Conclusions Acknowledgments References Further reading Surface roughness and fatigue properties of selective laser melted Ti6Al4V alloy Zhuoer Chen, Sheng Cao, Xinhua Wu and Chris H.J. Davies 13.1 Introduction 13.1.1 Surface roughness of selective laser melted metallic components 13.1.2 Post-selective laser melting surface treatment 13.1.3 Fatigue performance of selective laser melted Ti6AL4V 13.2 Experimental procedure 13.2.1 Material 13.2.2 Selective laser melting of Ti6Al4V specimens 13.2.3 Surface roughness measurements 13.2.4 Postselective laser melting heat treatment of fatigue samples 13.2.5 Fatigue testing 13.3 Results and discussion 13.3.1 Surface roughness 13.3.2 Fatigue properties
ix
237 237 238 239 240 240 245 248 256 258 263
265 265 266 268 269 270 276 276 276 281
283 283 283 285 286 286 286 286 288 288 288 289 289 293
x
Contents
13.4 Conclusions References 14
15
Aluminum alloys for selective laser melting towards improved performance Paul Rometsch, Qingbo Jia, Kun V. Yang and Xinhua Wu 14.1 Introduction 14.2 Processingmicrostructureproperty considerations for current alloys in selective laser melting 14.3 Alloy and process design for improved performance 14.3.1 Design of new alloys for selective laser melting 14.3.2 Adaptation of existing high strength alloys for selective laser melting 14.3.3 Development of composite materials for selective laser melting 14.4 Summary and outlook Acknowledgments References Additive aerospace considered as a business Lawrence Gasman 15.1 3D printing technologies for tooling and prototyping 15.2 Factors driving additive manufacturing in the aerospace industry 15.2.1 Weight reduction 15.2.2 Additive manufacturing and improved aircraft design 15.2.3 Software as a limiting factor on additive manufacturing aerospace 15.3 How additive manufacturing is improving the supply chain the aerospace industry 15.3.1 Service providers 15.3.2 Siemens 15.3.3 SAP 15.4 Materials for additive manufacturing aerospace 15.4.1 Metals 15.4.2 Directed energy deposition 15.4.3 Polymers 15.4.4 Polymer bed fusion 15.4.5 Composites 15.5 Regulatory factors in additive manufacturing aerospace 15.5.1 Europe 15.5.2 United States 15.6 The geography of additive manufacturing aerospace 15.7 Competitive implications of additive manufacturing aerospace
297 298
301 301 302 310 311 316 319 321 322 322 327 328 328 328 329 329 330 330 330 331 331 331 333 334 336 337 338 339 339 339 340
Contents
16
17
Surface texture characterization and optimization of metal additive manufacturing-produced components for aerospace applications Agustin Diaz 16.1 Introduction 16.1.1 The basics of additive manufacturing surface texture 16.1.2 Surface anatomy of additive manufacturing components 16.2 Best practices for surface texture characterization of additive manufacturing components 16.2.1 Introduction 16.2.2 Brief surface texture review 16.2.3 Surface texture characterization of additive manufacturing components 16.3 Surface finishing of additive manufacturing components 16.3.1 Introduction 16.3.2 Basics of surface finishing to take into account before printing 16.3.3 Surface finishing of additive manufacturing components 16.4 Additive manufacturing-built components surface-finished by the Extreme ISF Process 16.4.1 Introduction 16.4.2 Examples of surface treated by the Extreme ISF Process 16.4.3 Improvement of the mechanical properties of additive manufacturing-built components by Extreme ISF Process 16.4.4 Further analysis of surface texture parameters associated with mechanical performance 16.5 Conclusions 16.6 Corollary Acknowledgments References Developing and applying ICME 1 modeling tools to predict performance of additively manufactured aerospace parts Brain W. Martin, Thomas K. Ales, Matthew R. Rolchigo and Peter C. Collins 17.1 Introduction 17.2 Part 1: Process modeling 17.3 Part 2: Predicting chemistry 17.3.1 Solute loss (vaporization) or pickup (gettering) 17.3.2 Solidification partitioning 17.4 Part 3: Predicting microstructure 17.5 Part 4: Predicting properties and performance 17.6 Limitations 17.7 Summary References
xi
341 341 341 342 345 345 345 349 355 355 356 358 361 361 362
364 365 367 367 371 371
375
375 379 384 385 386 387 391 393 395 396
xii
18
19
20
Contents
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition Lucas W. Koester, Leonard J. Bond, Hossein Taheri and Peter C. Collins 18.1 Introduction 18.1.1 Additive manufacturing components in service 18.1.2 Regulatory actions and standardization 18.1.3 Material properties and defects in additive manufacturing 18.2 State of the art 18.2.1 Optical and thermal monitoring 18.2.2 Post-production inspection 18.2.3 Emerging methods 18.3 Practical considerations Acknowledgments References Combining additive manufacturing with conventional casting and reduced density materials to greatly reduce the weight of airplane components such as passenger seat frames Francis Froes Conclusions References Synergetic technologies of direct layer deposition in aerospace additive manufacturing Petr A. Vityaz, Mikhail L. Kheifetz and Sergei A. Chizhik 20.1 Introduction 20.2 System analysis of the processing methods using the concentrated energy fluxes 20.3 Additive synergetic technologies of layer by layer synthesis 20.4 Ion implantation and ion deposition of coatings 20.5 Electron-beam heating of a coated surface 20.6 Conclusion Acknowledgments References
Index
401
401 401 402 404 406 406 409 409 413 414 414
419 425 425
427 427 428 432 434 438 445 446 446 449
List of Contributors
Thomas K. Ales Department of Materials Science and Engineering, Iowa State University, Ames, IA, United States Leonard J. Bond Center for Nondestructive Evaluation, Applied Sciences Complex II, Iowa State University, Ames, IA, United States; Department of Mechanical Engineering, Iowa State University, Ames, IA, United States; Department of Aerospace Engineering, Iowa State University, Ames, IA, United States Rodney Boyer School of Materials Science and Engineering, University of Shanghai for Science and Technology, Shanghai, P.R. China; RBTi Consulting, Bellevue, WA, United States Sheng Cao Monash Centre for Additive Manufacturing (MCAM), Monash University, Notting Hill, VIC, Australia; School of Materials Science and Engineering, University of Shanghai for Science and Technology, Shanghai, P.R. China Wei Chen AVIC Manufacturing Technology Institute, Beijing, P.R. China Zhuoer Chen Department of Mechanical and Aerospace Engineering, Monash University, Clayton, VIC, Australia; Monash Centre for Additive Manufacturing (MCAM), Monash University, Notting Hill, VIC, Australia Sergei A. Chizhik Presidium of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus; A.V. Lykov Heat and Mass Transfer Institute of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus Peter C. Collins Department of Materials Science and Engineering, Iowa State University, Ames, IA, United States Chris H.J. Davies Department of Mechanical and Aerospace Engineering, Monash University, Clayton, VIC, Australia; Monash Centre for Additive Manufacturing (MCAM), Monash University, Notting Hill, VIC, Australia Agustin Diaz REM Surface Engineering, Brenham, TX, United States
xiv
List of Contributors
B. Dutta DM3D Technology, Auburn Hills, MI, United States Francis Froes Light Metals Industry, Tacoma, WA, United States; Advanced Materials Industries, Tacoma, WA, United States Lawrence Gasman SmarTech Markets Publishing Crozet, Virginia, United States Shishkovsky Igor Center for Design, Manufacturing and Materials, Skolkovo Institute of Science and Technology, Moscow, Russia Orest Ivasishin G.V. Kurdyumov Institute for Metal Physics, Kyiv, Ukraine Qingbo Jia Department of Materials Science and Engineering, Monash University, Clayton, VIC, Australia; Monash Centre for Additive Manufacturing (MCAM), Monash University, Notting Hill, VIC, Australia Manish Kamal Arconic Inc., Arconic Fastening Systems, Carson, CA, United States Mikhail L. Kheifetz Presidium of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus; State Scientific and Production Association hhCenterii of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus Lucas W. Koester Center for Nondestructive Evaluation, Applied Sciences Complex II, Iowa State University, Ames, IA, United States Dmytro Kovalchuk JSC NVO Chervona Hvilya, Kyiv, Ukraine Anja Loesser EOS North America Inc Zhiqiang Li AVIC Manufacturing Technology Institute, Beijing, P.R. China Brain W. Martin Department of Materials Science and Engineering, Iowa State University, Ames, IA, United States Kevin Minet EOS Finland Oy Thomas F. Murphy Hoeganaes Specialty Metal Powders LLC, Cinnaminson, NJ, United States Joel C. Najmon Department of Mechanical and Energy Engineering, Indiana UniversityPurdue University Indianapolis, Indianapolis, IN, United States Tsuyoshi Nakagawa JAXA Safety and Mission Assurance, Tokyo, Japan
List of Contributors
xv
M.B. Novikova All-Russia Institute of Light Alloys, JSC, Moscow, Russia Eric Ott General Electric Additive, West Chester, OH, United States I.S. Polkin All-Russia Institute of Light Alloys, JSC, Moscow, Russia Behrang Poorganji General Electric Additive, West Chester, OH, United States Sajjad Raeisi Department of Mechanical and Energy Engineering, Indiana UniversityPurdue University Indianapolis, Indianapolis, IN, United States Niko Raitanen EOS Finland Oy Jeremy H. Rao Department of Materials Science and Engineering, Monash University, Clayton, VIC, Australia; Monash Centre for Additive Manufacturing (MCAM), Monash University, Notting Hill, VIC, Australia Gregory Rizza Arconic Inc., Arconic Fastening Systems, Carson, CA, United States Matthew R. Rolchigo Department of Materials Science and Engineering, Iowa State University, Ames, IA, United States Paul Rometsch Department of Materials Science and Engineering, Monash University, Clayton, VIC, Australia; Monash Centre for Additive Manufacturing (MCAM), Monash University, Notting Hill, VIC, Australia; Rio Tinto Arvida Research and Development Centre, Jonquie`re, QC, Canada Richard Russell NASA Engineering and Safety Center, Kennedy Space Center, FL, United States Ankit Saharan EOS North America Inc Hector Sandoval Lockheed Martin Missiles and Fire Control, Grand Prairie, TX, United States Christopher T. Schade Hoeganaes Specialty Metal Powders LLC, Cinnaminson, NJ, United States Mohsen Seifi ASTM International, Washington, DC, United States Nima Shamsaei National Center for Additive Manufacturing Excellence, Auburn University, Auburn, AL, United States
xvi
List of Contributors
S.V. Skvortsova Moscow Aviation Institute (National Research University), Moscow, Russia Hossein Taheri Center for Nondestructive Evaluation, Applied Sciences Complex II, Iowa State University, Ames, IA, United States; Department of Mechanical Engineering, Iowa State University, Ames, IA, United States Andres Tovar Department of Mechanical and Energy Engineering, Indiana UniversityPurdue University Indianapolis, Indianapolis, IN, United States G.A. Turichin St. Petersburg State Maritime Technical University, St. Petersburg, Russia Petr A. Vityaz Presidium of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus; Joint Institute of Mechanical Engineering of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus Jess Waller NASA-Johnson Space Center White Sands Test Facility, Las Cruces, NM, United States Douglas Wells NASA Marshall Space Flight Center, Huntsville, AL, United States James C. Withers ATS-MER, LLC, Tucson, AZ, United States Xinhua Wu Department of Materials Science and Engineering, Monash University, Clayton, VIC, Australia; Monash Centre for Additive Manufacturing (MCAM), Monash University, Notting Hill, VIC, Australia Kun V. Yang Department of Materials Science and Engineering, Monash University, Clayton, VIC, Australia; Monash Centre for Additive Manufacturing (MCAM), Monash University, Notting Hill, VIC, Australia; CSIRO Manufacturing, Clayton, VIC, Australia
Introduction to aerospace materials requirements and the role of additive manufacturing
1
Francis Froes1, Rodney Boyer2,3 and B. Dutta4 1 Advanced Materials Industries, Tacoma, WA, United States, 2School of Materials Science and Engineering, University of Shanghai for Science and Technology, Shanghai, P.R. China, 3RBTi Consulting, Bellevue, WA, United States, 4DM3D Technology, Auburn Hills, MI, United States
1.1
Aerospace materials and their requirements
Aerospace materials are frequently metal alloys, although they also include polymeric based materials, that have either been developed for, or have come to prominence through, their use for aerospace purposes. Aerospace uses often require exceptional performance, strength or heat resistance, even at the cost of considerable expense in their fabrication or conventional machining. Others are chosen for their long-term reliability in this safety-conscious field, particularly for their resistance to fatigue loading. The field of materials engineering is an important one within aerospace engineering. Its practice is defined by the international standards bodies that maintain standards for the materials and processes involved, such as ASTM, AMS or company specifications (Table 1.1 shows specifications for additive manufacturing [AM]). Generally, not a lot of information is contained in company specs, but with the controls required, a company spec will be mandatory due to the complexity of the process, where the customer will want to know a lot more details than will ever get into a public spec due to protection of proprietary information. A further requirement is observer observation of fabrication of acceptable quality, including microstructures (Fig. 1.1).
1.2
Additive manufacturing
In publications over the past few years [15], the cost of fabricating various titanium precursors and mill products has been discussed (very recently the price of TiO2 has risen to $2.00 per pound and TiCl4 to $0.55 per pound) and it has been pointed out that the cost of extraction is a small fraction of the total cost of a component fabricated by the cast and wrought (ingot metallurgy) approach. To reach a final component, the mill products must be machined, often with very high buy-to-fly ratios Additive Manufacturing for the Aerospace Industry. DOI: https://doi.org/10.1016/B978-0-12-814062-8.00001-7 © 2019 Elsevier Inc. All rights reserved.
2
Additive Manufacturing for the Aerospace Industry
Table 1.1 Specifications released and in-work for additive manufacturing (AM) components Specification no.
Specification title
Status
AMS 4998 AMS 4999
Titanium alloy powder, Ti6Al4V Titanium alloy laser deposited products, Ti6Al4V, annealed Additive manufacture of aerospace parts from Ni-base superalloy 625 via laser powder bed process Ni base 625 superalloy powder for use in laser powder bed manufacturing of aerospace parts Process requirements for production of powder feedstock for use in laser powder bed additive manufacturing of aerospace parts Laser powder bed fusion process Titanium alloy preforms from high deposition rate additive manufacturing on substrate Ti6Al4V stress relieved Quality management systems— requirements for aviation, space, and defense organizations
Released—1977 Released—2002
AMS 7000
AMS 7001
AMS 7002
AMS 7003 AMS 7004
AS9100
In work
In work
In work
In work In work
Issued 1997, Current Rev. D, 2016
The AS (Aerospace Standards) are utilized for details with regard to quality management systems. They do not cover specific material requirements such as properties, inspection, etc.
(which can reach as high as 40:1). The generally accepted cost of machining a component is that it doubles the cost of the component. The feedstock for AM can be a wire or a powder. Using powder, there are two basic approaches to AM: powder bed fusion (PBF) and direct energy deposition (DED), Figs. 1.1 and 1.2. The PBF
Figure 1.1 Schematic showing powder bed fusion technology.
Introduction to aerospace materials requirements and the role of additive manufacturing
Laser beam
3
Final focus optics
Nozzle shielding gas To powder feeder
Feedback sensor 1 Workholding fixture
Feedback sensor 2 Solid free form shape by direct deposition Substract or die preform
Figure 1.2 Schematic showing DMD technology. DMD, Direct metal deposition.
technique allows the fabrication of complex features, hollow cooling passages, high precision parts, and single metal builds. The DED approach allows large build envelopes, high deposition rates, multiple materials, and addition of material to existing components. Mechanical properties are at least at ingot metallurgy levels (including fracture toughness). Examples of AM manufactured parts and parts which could be AM fabricated in an advanced engine are shown in Fig. 1.3.
1.3
Additive manufacturing fabrication of various types of materials
The AM technique has been applied to metals, ceramics, and polymeric materials (Figs. 1.31.5) After the AM build, ceramic parts are porous and, if desired, can be infiltrated or fired in a postprocess step. This method is used for fine art ceramics. The primary advantages for 3-D printing, especially binder jetting (binder jetting is an AM process in which a liquid binding agent is selectively deposited to join powder particles. Layers of material are then bonded to form an object, for example by hot isostatic pressing) are low cost, high speed, scalability, ease of building parts in
4
Additive Manufacturing for the Aerospace Industry
Figure 1.3 (A) Examples of metallic parts fabricated by additive manufacturing. (B) The propulsion system for the F-35 lightening, which contains a substantial number of components that can be fabricated by additive manufacturing.
multiple materials, and versatility for use with ceramic materials. Originally evolved from systems that used thermoplastics, the binder jetting method has been modified to accept ceramic slurries or ceramic powders in wax or liquid binder carriers. Material jetting has significant challenges, including getting materials to flow through nozzles at reasonable speeds without clogging. Work is ongoing to improve the rheology of material systems for ceramic materials, such as alumina and zirconia. This method promises good surface finishes and high tolerances for parts that can be printed and then fired to high density. PBF originated with selective laser
Introduction to aerospace materials requirements and the role of additive manufacturing
5
Figure 1.4 Examples of ceramic parts fabricated by the additive manufacturing technique.
Figure 1.5 Examples of polymeric materials fabricated by additive manufacturing.
sintering (SLS). SLS uses a powder bed layer in a build box, similar to the binder jet method, but it is placed in a system that brings the powder to an elevated temperature and then exposes select areas to a laser beam. This causes localized sintering of the ceramic powder. The part has sufficient strength for handling, but requires a conventional postprocess firing to achieve full density. All PBF methods share certain characteristics, including one or more thermal sources for inducing
6
Additive Manufacturing for the Aerospace Industry
fusion or sintering between particles, a method for prescribing fusion in a region of each layer, and mechanisms for adding and smoothing powder layers. Electron beam melting has become a popular approach to PBF as it is fast, efficient, and can provide fully dense parts. The downside for ceramics is that the powder bed needs to be electrically conductive. Directed energy deposition is also a technique that is better suited to metals. Polymerization is used for polymers that can be loaded with ceramic powders. Polymeric parts range from epoxy-based, ABS (acrylonitrile butadene styrene a lightweight thermoplastic polymer resin of the polyester family), wax, polystyrene, ceramic, and nylon. All the systems follow the principles of building layer by layer, but vary in how the materials are applied (e.g., as a fine powder, liquid polymer, or molten plastic) and how they are cured (e.g., by melting with a laser or activating UV resin with a laser). The material most commonly used in nonimplantable medical applications is nylon 12.
1.4
Contents of this book
In this book an attempt was made to contact all the major AM fabricators and aerospace user (airframes, engines, and missile) world-wide to fully cover all the significant activity in the AM in aerospace field. Not all the organizations contacted expressed an interest in participating in this book. However it is considered that enough information has been gathered to give the reader a comprehensive view of the field. One rumor which was dispelled is that the USAF is very negative on AM, they are not. Read the 40 articles which follow this introduction to get a broad view of the use of AM in aerospace applications.
References [1] B. Dutta, F.H.(Sam) Froes, Additive manufacturing of titanium alloys, Adv. Mater. Proc. (2014) 1823. [2] B. Dutta, F.H.(Sam) Froes, Chapter 24: the additive manufacturing of titanium alloys, in: M. Qian, F.H. (Sam) Froes (Eds.), Titanium Powder Metallurgy, ButterworthHeinemann, 2015. [3] F.H.(Sam) Froes, B. Dutta, The additive manufacturing of titanium alloys, in: Proceedings of the World Conference on Titanium, San Diego, CA, 2015. [4] B. Dutta, F.H. Froes, Additive manufacturer of titanium alloys, in: A. Badiru (Ed.), Additive Manufacturing Handbook: Product Development for the Defense Industry, 2016. [5] B. Dutta, F.H. Froes, Book Additive Manufacturing of Titanium Alloys, Elsevier Publishing, Amsterdam, 2016.
Review of additive manufacturing technologies and applications in the aerospace industry
2
Joel C. Najmon, Sajjad Raeisi and Andres Tovar Department of Mechanical and Energy Engineering, Indiana UniversityPurdue University Indianapolis, Indianapolis, IN, United States
2.1
Aerospace requirements and opportunities for additive manufacturing
Additive manufacturing (AM) is being established as a fabrication technology that brings revenue to the aerospace industry throughout its supply chain and repair operations [1]. For the last 10 years, the aerospace industry has been one of the top sectors leading the AM market (Figs. 2.1 and 2.2). Today, 18.2% of the revenue in the AM industry is received from the aerospace industry. The aerospace sector is also the fastest growing sector, showing an annual increase of 1.6% in 2016, followed by motor vehicles with a growth of 1.0%. The revenues from the AM are estimated at $2.7 billion in 2016 (growth of 12.9% with respect to 2015) and are expected to surpass $100 billion within the next two decades, mostly in the aerospace industry [3]. The market for AM parts in aerospace can be divided into metallic and nonmetallic (mostly polymer) components, which are generally related to critical and noncritical aircraft parts, respectively. Boeing and Bell Helicopter started using polymer AM parts for nonstructural components in the mid-1990s. In March 2015, Boeing fabricated more than 200 unique parts for 10 different aircraft using AM technologies. By that time, more than 20,000 nonmetallic AM parts were installed in airplanes [1]. Today, Boeing has installed tens of thousands of AM parts on 16 commercial and military aircraft [4]. In 2017, Boeing started using at least four AM titanium-alloy parts to produce its 787 Dreamliner aircraft with near-future plans to manufacture almost 1000 parts via AM to save $23 million per airplane [5]. Airbus is also a main player in AM. It has installed AM metal brackets and bleed pipes on the Airbus A320neo and the A350 XWB test aircraft [6]. It also has a multiyear cooperative research agreement with Arconic to produce large-scale AM airframe components (1 m in length) [7]. NASA, the European Space Agency, and SpaceX are exploring the use of AM igniters, injectors, and combustion chambers on their rocket engines. Honeywell Aerospace, Lockheed Martin, and Northrop Grumman are also important users of AM. The unique design requirements from Additive Manufacturing for the Aerospace Industry. DOI: https://doi.org/10.1016/B978-0-12-814062-8.00002-9 © 2019 Elsevier Inc. All rights reserved.
8
Additive Manufacturing for the Aerospace Industry
Figure 2.1 Additive manufacturing market share by industry in 2016. Source: Wohlers, A., Wohlers Report 2016. 3D Printing and Additive Manufacturing State of the Industry. Annual Worldwide Progress Report. 2016: Associates Wohlers. Wohlers Report 2016. Wohlers Associates, Inc. [2].
Figure 2.2 Additive manufacturing market share by industry in 2017. Source: Wohlers, A., Wohlers Report 2017. 3D Printing and Additive Manufacturing State of the Industry. Annual Worldwide Progress Report. 2017: Associates Wohlers. Wohlers Report 2017. Wohlers Associates, Inc. [4].
the aerospace industry and capabilities of AM technologies are discussed in this section.
2.1.1 Design requirements The aerospace industry is constantly demanding lightweight aircraft components with a high strength-to-weight ratio to improve fuel efficiency, reduce emissions, and respond to safety and reliability requirements [8]. To this end, aerospace designers strive to minimize the amount of material used in every component,
Review of additive manufacturing technologies and applications in the aerospace industry
9
which results in increased design complexity with respect to the structure, function, and property [9]. AM allows the fabrication of parts with virtually any shape (freeform fabrication).
2.1.1.1 Structural design Structurally complex designs are characterized by nontraditional (freeform, organiclike) external shapes that provide high mechanical performance with minimum mass. AM allows the fabrication of parts with complex designs with some constraints, depending on the specific AM technology [10]. Frame designs are examples where topology optimization is used. Complex aircraft structures can be also designed to maximize internal space (packing ratio) [11]. When compared with traditional manufacturing processes, like injection molding or milling, AM is significantly more versatile due to its layer-by-layer fabrication process [12]. This allows for the optimization and manufacturing of lightweight, strong, and robust parts [13]. Structural complexity also describes the multiscale internal hierarchical architectures of lattice/cellular arrays (mechanical metamaterials) [1416]. Such complex structures can be obtained with use of topology optimization algorithms [1723]. Commercial software that can provide lattice materials to optimize designs includes Netfabb, Within, Materialise, and Simpleware, as well as topology optimization tools such as OptiStruct, Genesis, and Meshify. The solution of topology optimization problems involves multiphysics simulations and multiscale design methods to find the best compromised solution among multiple objectives [24]. With the use of these software tools, aerospace designers have the ability to reduce material and weight while making the design suitable for AM. These tools also allow the ability to retain or increase the part’s mechanical performance and reduce manufacturing cost. Lightweight parts can deliver substantial cost savings from the fabrication procedure to the end of the airplane’s life [25].
2.1.1.2 Functional complexity Functional complexity refers to the ability of parts to integrate multiple functions (multifunctional parts), including functions that are traditionally not assigned to the part, for example, heat dissipation, electrical circuitry, flexibility, capillarity to a load-bearing component. Examples in aerospace include structural components that also act as conduits, such as airfoils and turbine blades with embedded cooling channels [26,27]. Functional complexity also defines functions that are difficult to achieve by a single component. An example is the design of the swirler inside a jet engine that forces combustion products to recirculate inside the chamber, generating a highly turbulent flow. Since increasing the turbulence can reduce the chamber pressure, the swirler geometry must be designed precisely to deliver the desired turbulence suitably to mix the injected fuel and the airflow [28]. Such a complex function is difficult to achieve without the freeform fabrication capabilities of AM.
10
Additive Manufacturing for the Aerospace Industry
2.1.1.3 Property requirements Finally, behavioral complexity describes changes in the material properties across the component. This includes multimaterial designs and functionally graded materials (FGMs). FGMs can be fabricated by local microstructure modifications from thermally-induced transformations [2933] or by changes in the architecture of the cellular/lattice structure [3436]. The ability to tailor mechanical properties is particularly relevant to additive metal technologies (AMTs), like directed energy deposition (DED) and powder bed fusion (PBF) [37]. In PBF, the powder grain size and distribution can be varied to change the density and corresponding properties [38,39]. The processing parameters can be varied to tailor the quality of the part’s hardness, fatigue strength, surface microstructure, and surface finish [39,40]. This allows for the manufacturing of metal parts with local control of residual stress concentrations [41].
2.1.2 Manufacturing capabilities and benefits The increased complexity of the structure, function, and property of an aerospace component entails fabrication challenges and higher costs [42]. Traditionally, the design is simplified for manufacturing and assembly operations at the expense of part performance. AM, however, allows freeform fabrication and reduces assembly through part consolidation. It also reduces material waste and allows the use of premium materials that are difficult to process with other manufacturing techniques. Unlike conventional manufacturing, AM reduces or even eliminates the need for tooling, which allows the production of small production runs and parts that require quick turnaround time. The benefits of AM for the aerospace industry are discussed in this section.
2.1.2.1 Part consolidation Traditionally, complex aerospace components contain multiple simple parts that are joined together using different types of fasteners (welds, bolts, and brazes). However, such assemblies offer lower reliability and require greater inspection, tooling, and sustainment costs when compared to a single part [43]. In addition, geometric errors and undesired misalignments or deformations may exceed allowable tolerances in aerospace components [44]. Part consolidation can be attained using AM. This enables feature integration as well as increased reliability and performance [45]. When a complex part is fabricated on a single AM machine, the part inventory is reduced and the economies of scale associated with large centralized factories are also reduced [1]. Decreasing the number of parts in an assembly reduces (1) the number of tools held in inventory, (2) the costs associated with documentation, inspection, and production, (3) the assembly line footprint, and (4) the overall manufacturing costs. GE Aviation has reported a reduction from 855 parts produced using conventional manufacturing into a dozen parts using AM technologies. The simplified
Review of additive manufacturing technologies and applications in the aerospace industry
11
design reduced weight, improved fuel efficiency up to 20%, and achieved 10% more power. A bearing support and sump were redesigned to consolidate 80 parts into one. Also, a 20-part nozzle was consolidated into a single AM unit and the weight was reduced by 25% [46]. Similarly, Airbus reduced a 126-part hydraulic housing tank to a single AM part [47] (Fig. 2.3).
2.1.2.2 Material economy A significant contributor to the high fabrication cost of aerospace components is the buy-to-fly ratio, which is defined as the weight ratio between the raw material and the weight of the final component. Aerospace components with large volume envelope-to-volume ratios (e.g., thin-walled structures and turbine blades) have buy-to-fly ratios as high as 2040:1 [48], which makes CNC tool planning time consuming and impractical due to the massive amount of waste material. In these cases, the material use and the manufacturing cost can be significantly improved with AM freeform fabrication capabilities and reduce the buy-to-fly ratio closer to 1:1. For PBF processes, the amount of waste produced is around 5% compared to that of traditional milling, which can produce up to 95% waste [12,49]. Reducing the material waste and the part weight also has a significant positive effect on the environment [50,51]. The aerospace industry constantly demands premium metals, such as titanium alloys, aluminum alloys, nickel-based super alloys, and special steels. Titanium alloys, for example, provide outstanding characteristics, like high strength-toweight ratio, a wide range of operating temperatures, high corrosion resistance, and composite compatibility. However, titanium alloys are highly limited by their relatively high cost compared to other materials [52] and their poor machinability [53]. A growing number of premium materials are available in AM technologies, including titanium. For instance, GE Avio Aero is producing low-pressure turbine blades in titanium aluminide (TiAl) using AM [54]. Today, Norsk Titanium owns the world’s largest titanium AM facility and supplies aerospace-grade AM structural titanium components, approved by the Federal Aviation Administration (FAA) of the United States, to companies like Boeing.
Figure 2.3 (A) AM hydraulic reservoir rack from Airbus consolidating 126 parts. (B) Consolidated design into one part. AM, Additive manufacturing. Source: Photo copyright Airbus Hermann Jansen.
12
Additive Manufacturing for the Aerospace Industry
2.1.2.3 Small production runs and turnaround time In comparison to conventional manufacturing processes, AM tend to be more expensive for large production runs; however, the relatively high investment costs of fixtures and tools are avoided or considerably reduced in AM [12]. Therefore, AM is more cost effective for customized parts and small production runs, which are common in the aerospace industry. Since aircraft have lifespans of over 30 years, maintaining and replacing legacy parts and tooling may involve high inventory costs [55]. AM enables the production of test and replacement parts on demand for rapid shipment and installation [56]. This minimizes downtime and associated costs. According to Airbus, the turnaround for test or replacement parts was as low as 2 weeks in 2014. Finally, AM allows parts to be manufactured at decentralized locations. This lowers transportation and storage costs. This simplification in supply chains also leads to a reduction in maintenance-based downtime and turnaround time [57].
2.2
Additive manufacturing technologies
The American Society for Testing and Materials International Committee F42 classifies AM technologies into seven categories: binder jetting, DED, material extrusion, material jetting, PBF, sheet lamination, and vat photopolymerization [58]. According to the material, AM technologies in aerospace can be classified into AMTs and additive nonmetal technologies. These technologies are described in this section.
2.2.1 Additive metal technologies The two most common AMTs for aerospace applications are DED and PBF. Widespread DED technologies in the aerospace industry include laser metal deposition (LMD), laser engineering net-shaping (LENS), electron beam welding (EBW), electron beam free-form fabrication (EBF3), and wire arc AM (WAAM). PBF technologies include direct metal laser sintering (DMLS), selective laser melting (SLM), and electron beam melting (EBM). Other relevant AMTs for aerospace applications include binder jetting and supersonic particle deposition (SPD), also known as cold spray (Table 2.1). This section describes the most relevant AMTs to the aerospace industry.
2.2.1.1 Directed energy deposition DED technology works by melting material that is fed to a local site on the build layer, usually occurring within an inert gas atmosphere [1,59]. While this can be used for nonmetal materials, it is predominately used with metals and metal alloys [1,6062]. Feedstock usually comes in the form of powder or wire and is melted with a focused energy source, such as laser beams, electron beams, and arcs [1,48].
Review of additive manufacturing technologies and applications in the aerospace industry
13
Table 2.1 Additive metal technologies in the aerospace industry AMTs Directed energy deposition G
G
G
LMD/LENS EBW/EBF3 WAAM
Powder bed fusion G
G
G
DMLS SLM EBM
Other G
G
Binder jetting SPD/cold spray
LMD, Laser metal deposition; LENS, laser engineering net-shaping; EBF3, electron beam free-form fabrication; EBW, electron beam welding; WAAM, wire arc AM; DMLS, direct metal laser sintering; SLM, selective laser melting; EBM, electron beam melting; SPD, supersonic particle deposition; AMTs, additive metal technologies.
DED processes build up the 3D part, layer by layer; however, the technology can be implemented in a multiaxial machine and provide 3D positioning [1]. This allows manufacturing of complex parts without the need of support structures. Due to increased versatility in orientation, DED technologies are ideal for component repair of turbine blades, engine combustion chambers, compressors, airfoils, and blisks [1,62,63]. Turbine airfoils have been repaired to within 0.030 mm accuracy of the original blade and shown that for a repair volume of 10%, there is a at least a 45% improvement in the carbon footprint and a 36% total energy savings [64]. The mechanical properties of an AM part produced by wire-fed laser and arc beam deposition processes rely heavily on process parameters, load direction, and post build-up heat treatment; however, both processes produced parts with comparable properties and are suitable for aerospace applications [60].
2.2.1.2 Powder bed fusion PBF technologies work by locally melting metal powder on a substrate to form layers. After a new layer is formed, a leveling roller distributes a new layer of powder. Un-melted powder is reusable and becomes the support for successive layers, reducing the need for support structures [1,59]. To avoid powder oxidation, SLM and DMLS occur in an inert gas atmosphere, while EBM occurs in a vacuum [65,66]. Consequently, the requirement of a vacuum makes EBM attractive for outer space manufacturing [59,67]. Compared to DED processes, PBF processes have the ability to create internal passages and, typically, produce higher-fidelity build features [68]. Titanium alloy (Ti6Al4V) has been typically used in the fabrication of AM parts via EBM technology [6974]. Topology optimized Ti6Al4V brackets were manufactured with an Arcam Q10 Plus EBM system, demonstrating that geometry and microstructure are dependent on the build time or cross-sectional area of each layer [74].
2.2.1.3 Other relevant additive metal technologies Binder jetting is used for quick and reliable rapid prototyping of metal AM parts, such as impellers and turbine blades [75,76]. It is also used for rapid tooling in the
14
Additive Manufacturing for the Aerospace Industry
fabrication of AM sand cores and molds to produce large metal parts. Typical metals used are aluminum and copper alloys, gray and ductile iron, and magnesium. Binder jetting AM of sand cores and molds allows the removal of patterns used in indirect rapid tooling and simplifies the steps involved in the creation of low volume production parts [77]. Applications of direct rapid tooling through metal binder jetting include complex gear cases and covers, fuel tanks, transmission housings, components requiring draft free walls, lightweight engine parts, and structural hinges [78]. SPD, or cold spray, works through the consolidation of supersonic, microsized, metal particles onto a suitable substrate upon impact (ballistic impingement). This technology is also suitable for ceramic and polymer powders [79]. The particles are accelerated with a spray gun fitted with a convergent-divergent rocket nozzle using a heated high-pressure gas (helium or nitrogen). Since the metal powder is not significantly heated during this process, with consolidation occurring in the solid form, the risks of oxidation, residual stress accumulation, and changes in the powder’s microstructure are avoided [79]. SPD has been utilized for repairing and enhancing the airworthiness and integrity of aging aircraft structures [8082]. Nitrogen-based SPD has been shown limited particle deformation resulting in high degrees of porosity, while helium-based SPD had substantially more microstructural deformation with very little porosity [83]. Aircraft that are often exposed to salt spray, like navy rescue/patrol helicopters, should be repaired and treated with helium-based SPD [55].
2.2.2 Additive nonmetal technologies Additive nonmetal technologies most relevant to the aerospace industry include: selective laser sintering (SLS), stereolithography (SLA), fused deposition modeling (FDM), and PolyJet [84] (Table 2.2). This section presents the main characteristics of these technologies and their relevance to the aerospace industry.
2.2.2.1 Selective laser sintering SLS is a PBF process that typically uses a laser energy source to melt polymer powders [59]. It has the ability to produce large parts with good mechanical strength at a relatively low cost [91]. For aerospace applications, SLS is primarily used for rapid prototyping of nonfunctional parts and direct digital manufacturing (DDM) of noncritical components [92]. Glass-filled nylon is used in SLS fabrication of engine compartments that require heat resistance, for example, tarmac nozzle bezel [85]. Nylon 12 is used in SLS produced parts that require flexibility, such as ducts and bellow directors for airflow [86] (Fig. 2.4). It has been shown that the percent crystallinity of SLS nylon-12 parts is dependent on the degree of the particle melt, to the extent that sufficiently different percent crystallinity part volumes can be treated as different materials [93]. This opens up the possibility of producing FGM parts with SLS. SLS parts produced with acrylic styrene and polyamide (nylon) show nearly the same mechanical properties as plastic injected counterparts [94,95]. In most applications, SLS is known to be cost effective for the production of small volume parts for aircraft [59].
Review of additive manufacturing technologies and applications in the aerospace industry
15
Table 2.2 Additive nonmetal technologies in the aerospace industry Additive nonmetal technologies Application
Example part
Requirements
Recommended process
Recommended material
Engine compartment
Tarmac nozzle bezel [85]
SLS
Glass-filled nylon
Air ducts
Airflow ducting [86]
SLS
Nylon 12
Full size panels
Seat backs and entry doors [84] Brackets and door handles [87]
Heat resistant functional parts Flexible ducts and bellow directors Large parts with smooth surface finish Cast metal parts using 3Dprinted patterns Fully transparent, high-detail models End-use custom screen bezels High strength, lightweight, durability Rigidity, dimensional stability
SLA
Standard resin
SLA and PolyJet
Castable resin or wax
SLA and PolyJet
Clear resin
PolyJet
Digital ABS
FDM
ULTEM
FDM
ULTEM
Customizable and robust handles
FDM
ULTEM
Casted metal parts
Lights
Bezels UAV
Prototypes and tools
Cabin accessories
Headlight prototypes [84] Dashboard interface [84] Wings and fuselage [88] Camera case prototype and tool to install wiring [89] Door handle covers [90]
SLS, Selective laser sintering; SLA, stereolithography; FDM, fused deposition modeling. Source: Adapted from B. Artley, Aerospace 3D Printing Applications, in: D. Hubs (Ed.). ,https://www.3dhubs.com/knowledge-base/ aerospace-3d-printing-applications. [84].
2.2.2.2 Stereolithography SLA, also known as vat photopolymerization, is a method of creating 3D objects using a light-emitting device (laser or digital light processing) that illuminates and cures a liquid photopolymer resin (thermosetting plastic) layer by layer [96]. SLA has the ability to produce fine features and provide good surface finish with minimum stair stepping effect [91]. Several photopolymer resins can be utilized with SLA: standard (rigid, opaque), castable, and clear, as well as flexible, high temperature, and dental, among others [97]. High-fidelity rapid prototypes for testing, verification, and design of aeroelastic airfoils have been produced with low-stiffness resins, where model similarity between prototypes was highly desired [98]. Cabin accessories such as console control parts with functional knobs as well as full size
16
Additive Manufacturing for the Aerospace Industry
Figure 2.4 Air ducts for laminar flow made with polyamide 12 (PA 2200) from EOS [86].
panels, seat backs, and entry doors have been produced with SLA standard resin [84]. Castable and high-temperature SLA resins are used to fabricate mold patterns (indirect rapid tooling) and injection molds (direct rapid tooling), respectively. Prototypes of highly-detailed, fully transparent aircraft headlights are produced with SLA clear resins [84].
2.2.2.3 PolyJet PolyJet, also known as material jetting, uses inkjet printing technology to jet liquid photopolymer droplets onto a build substrate and then cure it with UV light. It has the ability to fabricate parts with fine features and good surface finish, while exhibiting little stair stepping effect [91]. Some PolyJet systems also boast the ability to produce multimaterial parts through FGM, allowing a wide range of material properties selection [99]. The role of PolyJet in the aerospace industry includes rapid prototyping, indirect rapid tooling (mold pattern fabrication), and DDM. Material jetting occurs through two processes: drop on demand (DOD) and continuous inkjet (CIJ). The DOD process offers high part resolution at the expense of build time, making it favorable for applications requiring a fine surface finish, such as prototype light fittings and intricate wing design prototypes (bat-like ornithopter, lattice structure wing struts) [100,101]. The CIJ process offer faster build times at lower part resolutions and is better suited for noncritical, nonmetallic part fabrication, like interface bezels [102].
2.2.2.4 Fused deposition modeling FDM, also known as material extrusion, is currently the most popular AM technology on the market [103]. It allows the fabrication of durable components made of
Review of additive manufacturing technologies and applications in the aerospace industry
17
high-strength thermoplastics such as ULTEM, polycarbonate, polyphenylsulfone, polylactic acid, and acrylonitrile butadiene styrene (ABS) [104]. FDM systems are widely versatile in applications, ranging from quick and inexpensive rapid prototyping to tough and rigid parts suitable for end-use. The aerospace industry has been substituting traditionally metal parts with sufficiently strong FDM-produced parts to reduce weight and turnaround time for part repairs [105]. Approximately 70 production-grade thermoplastic parts, are being implemented in NASA’s Mars rover because they are lightweight, yet durable enough to withstand the rigors of space [106]. For prototype applications that do not require high resolution and surface finish, FDM technology is ideal because it is economical and does not require chemical postprocessing [107]. Stratasys and Aurora Flight Sciences fabricated the largest and fastest AM unmanned aerial vehicle (UAV) using FDM. Low weight and high strength are achieved using ULTEM 9085 as the build material in addition to adding an internal honeycomb structure to the internal wing design.
2.3
Additive manufacturing applications
AM technologies for aerospace applications usually fall under one of the following categories: DDM, rapid tooling, rapid prototyping, and repair. Table 2.3 shows a breakdown of AM technologies into metal and nonmetal application categories. DDM refers to the production of parts to be utilized in the aircraft. Such parts can be critical components required for the operation (e.g., nozzles, combustion chambers) and Table 2.3 Metal and nonmetal additive manufacturing technologies with select aerospace examples Application Metal
G
G
G
Nonmetal
G
G
G
DDM (direct metal part fabrication) Rapid tooling Repair DDM (fixtures and accessories) Rapid prototyping Rapid tooling
AM technology G
G
G
G
G
G
G
G
DED PBF Cold spray Binder jetting SLS SLA PolyJet FDM
Aerospace examples G
G
G
G
G
G
G
G
Helicopter engine combustion chamber fabrication (DED) [108] Blisk airfoil repair (DED) [109] Satellite antenna bracket (PBF) [110] Lap joint reinforcement (SPD) [81] Ratchet wrench printed by NASA on International Space Station (SLS) [111] UAV wing design (PolyJet) [91] Boeing 777-300ER door handle (FDM) [90] Camera case prototyping (FDM) [89]
DDM, Direct digital manufacturing; DED, directed energy deposition; PBF, powder bed fusion; SLS, selective laser sintering; SLA, stereolithography; FDM, fused deposition modeling; UAV, unmanned aerial vehicle.
18
Additive Manufacturing for the Aerospace Industry
noncritical parts (e.g., brackets, fixtures, and accessories). Rapid tooling refers to the fabrication of tools and patterns required for the fabrication of the final part. It can be classified into direct rapid tooling (e.g., molds and dies) and indirect rapid tooling (e.g., mold patterns). Rapid prototyping refers to the fabrication of nonfunctional parts, usually using nonmetal technologies. Finally, repair entails the repair and reinforcement of metallic parts and joints usually through DED and cold spray processes.
2.3.1 Direct digital manufacturing 2.3.1.1 Direct metal part fabrication DED systems from companies like TWI (The Welding Institute) and EOS GmbH (Electro-Optical Systems) are being used to fabricate complex and overhanging metal structures as the multiaxis orientation systems can orient the part so that jutting structures are built off of existing layers. TWI’s five-axis LMD system has been used in the production of an IN718 helicopter engine combustion chamber, featuring an overhanging flange [108] (Fig. 2.5). RUAG Switzerland has used AM with topology optimization to reduce the weight of the Sentinel satellites’ antenna bracket by 40% [110]. The optimized bracket is made of aluminum alloy, AlSi10Mg, manufactured with the EOS M 400, an SLM AM system. Boeing’s 787 Dreamliner is now using four additive manufactured titanium parts that will eventually reduce its production costs by up to $3 million [5]. Multipart assemblies have been consolidated into a single complex part using EOS’s M 290 DMLS system. The ArianeGroup was able to reduce a 248-part upper stage propulsion module to a single part through DMLS [112]. Vectoflow used DMLS to fabricate a compact, one-piece, flow measurement device, designed to withstand the severe stresses of aircraft in the subsonic and supersonic range [113]. GE has used Arcam’s EBM Q20plus to produce Ti6Al4V turbine blades and structural airframe components [114]. A liquid oxygen flange has been additively manufactured, in lieu of traditional manufacturing means, for use on the upper stage of United Launch Alliance Launch Vehicles [42].
Figure 2.5 TWI’s five-axis LMD printer manufacturing an IN718 helicopter engine combustion chamber. LMD, Laser metal deposition. Source: C. Hauser, Case Study: Laser Powder Metal Deposition Manufacturing of Complex Real Parts. 2014, TWI. Image Courtesy of TWI Ltd.
Review of additive manufacturing technologies and applications in the aerospace industry
19
Postprocessing necessities can be addressed through hybrid manufacturing [115]. Hybrid manufacturing is the combination of additive and subtractive manufacturing into a single system [116]. AM builds up the metal part, while traditional subtractive manufacturing (CNC milling) is used for spot milling and surface finishing. This combined process simplifies part fabrication by reducing the number of production steps. Several companies are introducing hybrid manufacturing systems (Hybrid Manufacturing Technologies’ Ambit Series 7 and ELB-Schliff’s millGrind) to the aerospace industry for part fabrication and repair. ELB-Schliff’s millGrind is the world’s first hybrid manufacturing featuring grinding for in-process repair of aerospace components [117].
2.3.1.2 Fixtures and accessories While, most mechanical and structural components of aircraft are made of metal, fixture and interior accessories are often made from nonmetals to save weight and cost. China Eastern Airlines used FDM to manufacture misprinted seat signs on its Boeing 777 aircraft, an expensive and time-consuming fix [118]. Northrop Grumman utilize SLA to fabricate repair kits for specific aircraft repairs simplifying teardown procedures and reducing maintenance-related downtime [119]. Similarly, Moog Aircraft Group has adopted FDM to manufacture component maintenance manual fixtures for internal inspection [120]. AM of these parts have other benefits, like assembly part consolidation. Advanced Aerials uses 3D systems’ SLS technology to fabricate tough parts for its unmanned vehicle systems [121]. Honeywell also used SLS for the manufacturing of control pod casings on its RQ-16 T-Hawk UAV [122].
2.3.2 Rapid tooling Rapid tooling describes any mold-making process that can create tools quickly and with minimum direct labor [123]. Generally, AM rapid tooling processes fall under indirect or direct rapid tooling. Indirect rapid tooling is the use of AM methods to produce a temporary part model. A reverse ceramic or sand mold is created from this model for metal part casting. One subprocess of indirect rapid tooling is reconfigurable tooling [124]. Reconfigurable tooling allows the reuse of molds through the use of state-changing materials for mold creation. This subprocess has been used to produce splashes and tools of wing shapes for aircraft in the field without the need for disassembly [124]. Airbus used indirect rapid tooling to create structural door hinges for the Airbus A320 [1]. With direct rapid tooling, molds and inserts are made directly with AM processes. Thus, direct rapid tooling does not require as many steps as indirect rapid tooling and has the potential to preserve overall part density more effectively [124].
2.3.3 Rapid prototyping Rapid prototyping allows for realization and verification of computer simulation models of aerospace parts and aircraft. Prototypes are useful for identifying
20
Additive Manufacturing for the Aerospace Industry
design drawbacks and opportunities that only become apparent through a physical model. In addition, prototypes are beneficial for wind tunnel testing and model verification of streamlines. Fig. 2.6 shows the complete model of an aircraft made for wind tunnel testing. Depending on the use case, most prototypes do not need to be manufactured with the actual part material, however prototypes should possess sufficient rigidity and fidelity to achieve accurate results in testing [125]. Generally, polymer-based AM processes, like SLA or FDM, produce models with sufficient rigidity and fidelity for prototype testing. Thus relatively expensive metal-based AM processes can be avoided, making rapid prototyping a quick and inexpensive way to validate physical features and computational fluid dynamic models. PolyJet has been used to prototype rapidly and test various wing designs for UAV applications. NASA carried out a study to evaluate design approaches for a next-generation commercial aircraft with AM rapid prototyping. The preparation of a wind tunnel model is a crucial stage of this project, as it is used to assess the aerodynamics, propulsion, operation, and structure of the proposed design. FDM AM technology shorten design cycles and lead times [126]. Furthermore, Airbus has used rapid prototyping to test and develop a small-sized unmanned aircraft prototype, the structure of which is 90% manufactured from plastic polyamide powder. The time-saving measures of rapid prototyping allowed the aircraft to be manufactured in only 8 weeks [127].
2.3.4 Repair Using AMTs for repair has several profitable outcomes. First, AMTs allow for expensive, damaged parts to be repaired instead of scrapping and replacing them. This has been shown to have significant cost savings [68]. Studies have also shown that repair through AM has a significantly smaller environmental footprint when compared to repair through conventional processes [128]. This section provides applications of AMTs for part repair under two categories: geometry restoration and structural integrity restoration.
Figure 2.6 (A) Complete aircraft model for wind tunnel testing [125]. (B) Flapping-wing UAV [91]. UAV, Unmanned aerial vehicle.
Review of additive manufacturing technologies and applications in the aerospace industry
21
2.3.4.1 Geometry restoration Geometry restoration focuses on restoring missing and worn geometry of the part. This is relatively easy to achieve given current DED processes, and provides direct and immediate cost savings [3]. In the commercial sector, the company Optomec has created the LENS Blisk Repair Solution system, which uses DED processes to repair blisks and other complex aircraft parts. Optomec did a case study on a T700 blisk, which had suffered erosion damage on the leading edge of the airfoils [109]. Fig. 2.7 shows the results of this study. RPM International Inc demonstrated a lowwattage repair of titanium components [129]. This type of low-wattage repair exhibited minimal distortion, making this application suitable for structural components of aircraft [129]. Researchers at Pennsylvania State University have successfully repaired compressor blade tips of the F402 engine in an AV-8B, resulting in increased engine performance [130]. LMD has been used to repair a CFM56 HPT Turbine Shroud SX [131]. The repair steps taken include scanning the damaged surface, planning the deposition path, and depositing melted powder to form the renewed SX structure [131].
2.3.4.2 Structural integrity restoration Structural integrity restoration looks at restoring or enhancing the structural integrity of a part. This typically involves repairing cracks and corrosive damage and is usually done through SPD [55,81,82]. RUAG Australia used an aluminum alloy SPD to prevent the corrosion of the magnesium alloy main transmission of a Royal Australian Navy Seahawk [55]. The United States military used titanium SPD to repair chafing damage on an aircraft hydraulic line [132]. SPD can also be applied to overlapping joints to increase structural integrity of the joint and prevent further corrosive damage. Hence, SPD can be used to repair holes and avoids the structural integrity compromises of traditional patch and strap repairs [55,81]. Furthermore, the application of an SPD doubler on a fuselage lap joint has shown significant reduction in peak stress compared to the fuselage lap joint without the SPD doubler [81].
Figure 2.7 (A) Airfoil repair using a LENS additive manufacturing system. (B) T700 blisk after edge repair. (C) T700 blisk after finishing. LENS, Laser engineering net-shaping. Source: Optomec, LENS Blisk Repair Solution.
22
2.4
Additive Manufacturing for the Aerospace Industry
Challenges and potential future applications
2.4.1 Challenges 2.4.1.1 Manufacturing limitations All AM technologies have limitations in terms of resolution, build quality and consistency, and warping, among other defects [9]. Of particular interest in the aerospace industry are the limitations associated with AMT. For metal sintering processes, the part resolution is limited to approximately 90% of the powder’s size [43]. For melting processes, the part resolution is limited by large powder particles, which increase the size of the melt pool [43]. Additional resolution limitations and uncertainties source from minimum incremental length of servo motors and translation quality of the CAD model to nozzle path planning [43]. Similarly, DED processes have shown that build quality can be inconsistent from build to build [133]. This can lead to uncertainty when measuring material performance at different regions of a part. Furthermore, there is significantly high cost associated with qualifying materials in the aerospace industry through traditional methods [133]. While a system’s build quality can be predicted and accounted for, this quality can change with the manufacturing of new parts or subsequent use the AM system. For AMTs, the spot size (area of energy source on powder) can be subjected to large variations across different combinations of energy sources and metallic powders. Depending on the emissivity and reflectivity of the powder, the powder can strongly reflect incoming energy sources, resulting in the powder not melting completely [43]. Additionally, resulting residual stresses and distortions, from sintering and melting processes, cannot always be predicted. Specifically, due to their nature, buckling distortions can be difficult to predict [133]. This limitation is especially relevant to large topology optimized designs containing thin trusses [133]. Furthermore, large build volumes may be expensive or unrealistic for AMTs that require an inert atmosphere or vacuum.
2.4.1.2 Postprocessing realities For aerospace applications, most DED and PBF processes will require some type of postprocessing [43]. Processes that use powder as the feedstock often produce parts with substantial porosity. This can be addressed with a hot isostatic pressing, which reduces porosity and increases part strength and reliability [43,134,135]. Occasionally an annealing postprocess is applied to consolidate the grain structure and obtain desired properties [43]. Surface finishing is often required for AM parts in aerospace applications. While traditional machining tools can finish the surface of most parts, complex parts often require unconventional methods, such as shot peening, chemical etching, and vibrahoning [43]. This need can limit the complexity of topology optimized parts, as the need and feasibility of surface finishing must be considered [133]. This can delay the lead time for parts and reduces the buy-to-fly ratio of the part, however these
Review of additive manufacturing technologies and applications in the aerospace industry
23
are still relatively small compared to traditional manufacturing methods and can be improved with further innovation of hybrid manufacturing systems.
2.4.1.3 Specification and standard development While there have recently been several new standards on AM for aerospace applications [136], there remains a large amount of research to qualify AM components. Furthermore this research is open ended and often lacks consensus [133]. Although the use of AM for aerospace components is not a novel application, the process of standard development for this application is in its infancy and is gradual [137]. Current standards have been developed specifying feedstock details, defect types, and inspection methods for additively manufactured parts; however, surface finish improvement techniques and damage tolerance refinements are potential areas for development [136,138].
2.4.2 Potential future applications Through the vast geometric freedom offered by AM technologies, multifunctional structures are more easily obtainable and have many potential advantages for the aerospace industry. Multifunctional structures are structures that can perform several engineering functions at once. This can be seen through part consolidation or part redesign and innovation. Several examples include embedded electronics within structures or surfaces (composite layers), structures with rigid and soft material gradients, integrated acoustic and thermal insulation, and 4D printing [43]. 4D printing refers to AM of parts with geometries that change with respect to time through environmental parameters, such as humidity, heat, and radiation [43,139]. PolyJet has started gaining a presence in AM for its ability to manufacture functionally complex parts, like multistage responsive 4D and shape-recovering structures [140,141]. The response of a polymer part to external environments, like temperature, can be used for the consolidation and removal of assemblies and servo motors, such as a passive air temperature regulator [141]. Multimaterial DED processes allow designers the ability to tailor a component’s response and behavior under mechanical loading and thermal environments [1,102] and help consolidate parts, simplifying the assembly procedure [11]. While topology optimization methods have been used for weight reduction, aeroelastic wing design, and optimum stiffener layout [11,110,142,143], multimaterial DED processes allow designers to apply multimaterial topology optimization to aerospace component design [1]. As AM technologies further develop, the foreseeable increase in system build volumes will eventually lead to larger and larger part fabrication, with the prospects of fabricating large components like an airplane wing [1]. AM also has the flexibility and convenience to establish an on demand, short lead time, supply chain for aircraft component replacement [1]. Such establishments are currently being studied for space use by Made in Space, Lunar Buildings, and NASA, investigating the potential of AM tools and parts in zero gravity environments [1].
24
Additive Manufacturing for the Aerospace Industry
Similarly, AM can be used for quick, cost-effective repair of high-value components [1,62,109,132,144]. A majority of the time and cost of AM repair is spent preparing the damaged part for AM repair. Automation of these preparation measures would allow for an automated repair process, which would have significantly lower cost and turnaround time when compared to fabricating a new part [1]. Recently, the European project, RepAIR, has researched a new high-batch repair system for the aerospace industry [145]. The system automatically determines geometrical deviations between the damaged part and original part and uses this data to reconstruct the geometry. Further development includes further automation of surface preparation and support for large-scale components. These automated repair systems can be incorporated into short turnaround supply chains that analyze a damaged component and decide if it is eligible for repair or needs an on demand replacement part fabricated.
References [1] R. Liu, et al., 13—Aerospace applications of laser additive manufacturing A2—Brandt, Milan, Laser Additive Manufacturing., Woodhead Publishing, 2017, pp. 351371. [2] A. Wohlers, Wohlers Report 2016, 3D Printing and Additive Manufacturing State of the Industry, Annual Worldwide Progress Report, Associates Wohlers, 2016. [3] L.J. Kumar, C.G. Krishnadas Nair, Current trends of additive manufacturing in the aerospace industry, in: D.I. Wimpenny, P.M. Pandey, L.J. Kumar (Eds.), Advances in 3D Printing & Additive Manufacturing Technologies, Springer Singapore, Singapore, 2017, pp. 3954. [4] A. Wohlers, Wohlers Report 2017, 3D Printing and Additive Manufacturing State of the Industry, Annual Worldwide Progress Report, Wohlers Associates: Fort Collins, CO, 2017. [5] L. Mearian, Boeing Turns to 3D-Printed Parts to Save Millions on its 787 Dreamliner. April 11, 2017. Available from: ,https://www.computerworld.com/article/3188899/3dprinting/boeing-turns-to-3d-printed-parts-to-save-millions-on-its-787-dreamliner.html.. [6] M. Caujolle, First Titanium 3D-Printed Part Installed Into Serial Production Aircraft. Newsroom 2017, September 13, 2017. Available from: ,http://www.airbus.com/ newsroom/press-releases/en/2017/09/first-titanium-3d-printed-part-installed-into-serialproduction-.html.. [7] BusinessWire, in: BusinessWire (Ed.), Arconic, Airbus to Advance 3D Printing for Aerospace Under Multi-Year Cooperative Research Agreement To Produce, Qualify Large-Scale 3D Printed Airbus Airframe Components, BusinessWire, New York and Frankfurt, 2017. [8] D.F.O. Braga, et al., Advanced design for lightweight structures: review and prospects, Prog. Aerosp. Sci. 69 (2014) 2939. [9] I. Gibson, D. Rosen, B. Stucker, Additive Manufacturing Technologies: 3D Printing, Rapid Prototyping, and Direct Digital Manufacturing, Springer, New York, 2014. [10] G.A.O. Adam, D. Zimmer, Design for additive manufacturing-element transitions and aggregated structures, CIRP J. Manuf. Sci. Technol. 7 (1) (2014) 2028. [11] J.-H. Zhu, Wei-Hong Zhang, L. Xia, Topology optimization in aircraft and aerospace structures design, Arch. Comput. Methods Eng. 23 (4) (2016) 595622.
Review of additive manufacturing technologies and applications in the aerospace industry
25
[12] L. Nickels, AM and aerospace: an ideal combination, Met. Powder Rep. 70 (6) (2015) 300303. [13] M.F. Cruz, A.V. Borille, Decision methods application to compare conventional manufacturing process with metal additive manufacturing process in the aerospace industry, J. Braz. Soc. Mech. Sci. Eng. 39 (1) (2017) 177193. [14] M. Alkhader, M. Vural, Mechanical response of cellular solids: role of cellular topology and microstructural irregularity, Int. J. Eng. Sci. 46 (10) (2008) 10351051. [15] C. Beyer, D. Figueroa, Design and analysis of lattice structures for additive manufacturing, J. Manuf. Sci. Eng., Trans. ASME 138 (12) (2016). [16] H.Z. Jishi, R. Umer, W.J. Cantwell, The fabrication and mechanical properties of novel composite lattice structures, Mater. Des. 91 (2016) 286293. [17] M.P. Bendsoe, O. Sigmund, Topology Optimization: Theory, Methods, and Applications, Springer, Berlin, Heidelberg, 2013. [18] A. Asadpoure, J.K. Guest, L. Valdevit, Incorporating fabrication cost into topology optimization of discrete structures and lattices, Struct. Multi. Optim. 51 (2) (2015) 385396. [19] K.A. James, J.S. Hansen, J.R.R.A. Martins, Structural topology optimization for multiple load cases using a dynamic aggregation technique, Eng. Optim. 41 (12) (2009) 11031118. [20] Y.Y. Kim, T.S. Kim, Topology optimization of beam cross sections, Int. J. Solids Struct. 37 (3) (2000) 477493. [21] K. Maute, M. Allen, Conceptual design of aeroelastic structures by topology optimization, Struct. Multi. Optim. 27 (1-2) (2004) 2742. [22] N.M. Patel, et al., Comparative study of topology optimization techniques, AIAA J. 46 (8) (2008) 19631975. [23] M. Zhou, R. Fleury, Fail-safe topology optimization, Struct. Multi. Optim. 54 (5) (2016) 12251243. [24] T. Wu, K. Liu, A. Tovar, Multiphase topology optimization of lattice injection molds, Comput. Struct. 192 (2017) 7182. [25] F.H. Froes, R. Boyer, B. Dutta, Additive manufacturing for aerospace applications-part II, Adv. Mater. Processes 175 (6) (2017) 1822. [26] R. Liu, et al., Aerospace applications of laser additive manufacturing, in: M. Brandt (Ed.), Laser Additive Manufacturing: Materials, Design, Technologies, and Applications, Elsevier Science, Amsterdam, 2016. [27] J. Gao, et al., An integrated adaptive repair solution for complex aerospace components through geometry reconstruction, Int. J. Adv. Manuf. Technol. 36 (11) (2008) 11701179. [28] A.F. El-Sayed, Fundamentals of Aircraft and Rocket Propulsion., Springer-Verlag:, London, 2016. [29] L. Lin, W.U.H. Syed, A.J. Pinkerton, Rapid additive manufacturing of functionally graded structures using simultaneous wire and powder laser deposition, Virtual Phys. Prototyping 1 (4) (2006) 217225. [30] P. Muller, J.-Y. Hascoet, P. Mognol, Toolpaths for additive manufacturing of functionally graded materials (FGM) parts, Rapid Prototyping J. 20 (6) (2014) 511522. [31] T. Niendorf, et al., Functionally graded alloys obtained by additive manufacturing, Adv. Eng. Mater. 16 (7) (2014) 857861. [32] I.T. Ozbolat, A.K.M.B. Khoda, Design of a new parametric path plan for additive manufacturing of hollow porous structures with functionally graded materials, J. Comput. Inf. Sci. Eng. 14 (4) (2014).
26
Additive Manufacturing for the Aerospace Industry
[33] C. Shen, et al., Fabrication of FeFeAl functionally graded material using the wire-arc additive manufacturing process, Metall. Mater. Trans. B: Process Metall. Mater. Process. Sci. 47 (1) (2016) 763772. [34] J. Bauer, et al., Nanolattices: an emerging class of mechanical metamaterials, Adv. Mater. 29 (40) (2017). [35] M. Tabatabaei, D. Le, S.N. Atluri, Nearly exact and highly efficient elastic-plastic homogenization and/or direct numerical simulation of low-mass metallic systems with architected cellular microstructures, J. Mech. Mater. Struct. 12 (5) (2017) 633665. [36] S. Yuan, et al., 3D soft auxetic lattice structures fabricated by selective laser sintering: TPU powder evaluation and process optimization, Mater. Des. 120 (2017) 317327. [37] M. Yakout, et al., The selection of process parameters in additive manufacturing for aerospace alloys, Int. J. Adv. Manuf. Technol. 92 (5-8) (2017) 20812098. [38] A.B. Spierings, G. Levy, Comparison of density of stainless steel 316L parts produced with selective laser melting using different powder grades, in: SFF Symposium, 2009, 2009. [39] A.B. Spierings, N. Herres, G. Levy, Influence of the particle size distribution on surface quality and mechanical properties in AM steel parts, Rapid Prototyping J. 17 (3) (2011) 195202. [40] M.K. Mariusz Krol, L.A. Dobrzanski, T. Tomasz, Influence of technological parameters on additive manufacturing steel parts in selective laser sintering, Arch. Mater. Sci. Eng. 67 (2) (2014) 8492. [41] N. Hoye, et al., Investigation of residual stresses in titanium aerospace components formed via additive manufacturing. in Eighth Australasian Congress on Applied Mechanics, ACAM 2014, as Part of Engineers Australia Convention 2014, Melbourne, VIC, Australia: Engineers Australia, November 25, 2014November 26, 2014. 2014. [42] G.J. Schiller, Additive manufacturing for aerospace. in 2015 IEEE Aerospace Conference, AERO2015, March 7, 2015March 14, 2015, Big Sky, MT, IEEE Computer Society, 2015. [43] A.A. Shapiro, et al., Additive manufacturing for aerospace flight applications, J. Spacecraft Rockets 53 (5) (2016) 952959. [44] M. Saadat, Challenges in the assembly of large aerospace components, Integrated Systems, Design and Technology 2010., Springer, 2011, pp. 3746. [45] T. Duda, L.V. Raghavan, 3D metal printing technology, IFAC-PapersOnLine 49 (29) (2016) 103110. [46] T. Kellner, An epiphany of disruption: GE additive chief explains how 3D printing will upend manufacturing, 3D Printing, General Electric., 2017. [47] A. Bentur, From Multiple to Singular—Consolidating Parts With Additive Manufacturing, LEO Lane, 2017. [48] B. Dutta, F.H. Froes, The additive manufacturing (AM) of titanium alloys, Met. Powder Rep. 72 (2) (2017) 96106. [49] H. Jan, et al., Rapid manufacturing in the spare parts supply chain: alternative approaches to capacity deployment, J. Manuf. Technol. Manage. 21 (6) (2010) 687697. [50] J. Faludi, et al., Comparing environmental impacts of additive manufacturing vs traditional machining via life-cycle assessment, Rapid Prototyping J. 21 (1) (2015) 1433. [51] R. Huang, et al., Energy and emissions saving potential of additive manufacturing: the case of lightweight aircraft components, J. Cleaner Prod. 135 (2016) 15591570. [52] R.R. Boyer, Titanium for aerospace: rationale and applications, Adv. Perform. Mater. 2 (4) (1995) 349368.
Review of additive manufacturing technologies and applications in the aerospace industry
27
[53] F. Nabhani, Machining of aerospace titanium alloys, Rob. Comput. Integr. Manuf. 17 (1) (2001) 99106. [54] T. Kellner, This electron gun builds jet engines, GE Aviation., General Electric, 2014. [55] N. Matthews, Chapter fifteen—additive metal technologies for aerospace sustainment, Aircraft Sustainment and Repair, Butterworth-Heinemann, Boston, MA, 2018, pp. 845862. [56] White Paper, Additive Manufacturing in Aerospace: Strategic Implications, S.M. Publishing (Ed.), 2014. [57] A. Uriondo, M. Esperon-Miguez, S. Perinpanayagam, The present and future of additive manufacturing in the aerospace sector: a review of important aspects, Proc. Instit. Mech. Eng., G: J. Aerosp. Eng. 229 (11) (2015) 21322147. [58] Technologies, A.C.F.o. A.M. and A.C.F.o.A.M.T.S.F.o. Terminology, Standard Terminology for Additive Manufacturing Technologies, ASTM International, 2012. [59] K.V. Wong, A. Hernandez, A review of additive manufacturing, ISRN Mech. Eng. 2012 (2012) 10. [60] Brandl, E., et al. Additive manufactured Ti6A14V using welding wire: comparison of laser and arc beam deposition and evaluation with respect to aerospace material specifications. in: Sixth International Conference on Laser Assisted Net Shape Engineering, LANE 2010, September 2124, 2010, Erlangen, Germany: Elsevier, 2010. [61] K.M. Taminger, Electron Beam Freeform Fabrication (EBF3) for Cost Effective NearNet Shape Manufacturing, National Aeronautics and Space Administration, Langley Research Center, Hampton, VA, 2006 [2006]. [62] R.P. Mudge, N.R. Wald, Laser engineered net shaping advances additive manufacturing and repair, Weld. J. 86 (1) (2007) 4448. [63] L. Portole´s, et al., A qualification procedure to manufacture and repair aerospace parts with electron beam melting, J. Manuf. Syst. 41 (2016) 6575. [64] J.M. Wilson, et al., Remanufacturing of turbine blades by laser direct deposition with its energy and environmental impact analysis, J. Cleaner Prod. 80 (2014) 170178. [65] E. Uhlmann, et al., Additive manufacturing of titanium alloy for aircraft components, Procedia CIRP 35 (2015) 5560. [66] Li, M.X. and W.Q. Qu. Electron beam additive manufacturing and its applications in aerospace field. in: 12th IUMRS International Conference on Advanced Materials, IUMRS-ICAM 2013, September 2228, 2013, Qingdao, China: Trans Tech Publications Ltd., 2014. [67] L.E. Murr, et al., Metal fabrication by additive manufacturing using laser and electron beam melting technologies, J. Mater. Sci. Technol. 28 (1) (2012) 114. [68] W.E. Frazier, Metal additive manufacturing: a review, J. Mater. Eng. Perform. 23 (6) (2014) 19171928. [69] A.A. Antonysamy, L.L. Parimi, Evaluation of residual stress in TI6AL4V parts produced by power bed—EBM process, SAMPE J. 53 (4) (2017) 3845. [70] H. Galarraga, et al., Fatigue crack growth mechanisms at the microstructure scale in asfabricated and heat treated Ti6Al4V ELI manufactured by electron beam melting (EBM), Eng. Fract. Mech. 176 (2017) 263280. [71] N. Hrabe, T. Gnaupel-Herold, T. Quinn, Fatigue properties of a titanium alloy (Ti6Al4V) fabricated via electron beam melting (EBM): effects of internal defects and residual stress, Int. J. Fatigue 94 (2017) 202210.
28
Additive Manufacturing for the Aerospace Industry
[72] N. Hrabe, T. Quinn, Effects of processing on microstructure and mechanical properties of a titanium alloy (Ti6Al4V) fabricated using electron beam melting (EBM), Part 2: energy input, orientation, and location, Mater. Sci. Eng. A 573 (2013) 271277. [73] N. Hrabe, T. Quinn, Effects of processing on microstructure and mechanical properties of a titanium alloy (Ti6Al4V) fabricated using electron beam melting (EBM), part 1: distance from build plate and part size, Mater. Sci. Eng. A 573 (2013) 264270. [74] S. Yoder, et al., Characterization of topology optimized Ti6Al4V components using electron beam powder bed fusion, Addit. Manuf. 19 (2018) 184196. [75] ExOne, Reduced Lead Time and Cost While Ensuring High Accuracy and Quality, 2015. [76] S. Meteyer, et al., Energy and material flow analysis of binder-jetting additive manufacturing processes, Procedia CIRP 15 (2014) 1925. [77] ExOne, Manufacture a 57 Inch Impeller Without a Core Box With ExOne’s Rapid Casting Process, 2014. [78] Industrial Applications—Aerospace, ExOne, February 19, 2015. Available from: ,www.exone.com/Industries-Applications/Industries/Aerospace.. [79] V. Champagne Jr, N. Matthews, V. Champagne Iii, Chapter fourteen—Introduction to supersonic particle deposition, Aircraft Sustainment and Repair., ButterworthHeinemann, Boston, MA, 2018, pp. 799844. [80] R. Jones, et al., On the use of supersonic particle deposition to restore the structural integrity of damaged aircraft structures, Int. J. Fatigue 33 (9) (2011) 12571267. [81] R. Jones, et al., Supersonic particle deposition as a means for enhancing the structural integrity of aircraft structures, Int. J. Fatigue 68 (2014) 260268. [82] R. Jones, et al., On the potential of supersonic particle deposition to repair simulated corrosion damage, Eng. Fract. Mech. 137 (2015) 2633. [83] S. Ngai, et al., Saltwater corrosion behavior of cold sprayed AA7075 aluminum alloy coatings, Corros. Sci. 130 (2018) 231240. [84] B. Artley, Aerospace 3D Printing Applications, 3D Hubs, December 18, 2017. Available from: ,www.3dhubs.com/knowledge-base/aerospace-3d-printing-applications.. [85] L. Langnau, Better Materials for High Temperature SLS, 2011. [86] 3D-printed functional lightweight components with maximum flexibility in product design, EOS GmbH, March 4, 2013. Available from: ,www.eos.info/industries_markets/aerospace/interior.. [87] E.K. Do, P.K. Wright, C.H. Sequin, Latest-generation SLA resins enable direct tooling for injection molding, SME 5 (3) (2006) 115. [88] W. Hardin, Additive Manufacturing Boosts Aerospace Applications, January 05, 2016. Available from: ,http://insights.globalspec.com/article/1885/additive-manufacturingboosts-aerospace-applications.. [89] S. Moore, 3D Printing Deployed in Aerospace Sector for Tooling, Prototyping, January 21, 2018. Available from: ,https://www.plasticstoday.com/3d-printing/3d-printingdeployed-aerospace-sector-tooling-prototyping/32485232158131.. [90] W. Hipolite, China Eastern airlines successfully 3D prints airplane parts for Boeing 777-300ER Aircraft, 3D Design, 3D Printing, April 13, 2015. Available from: ,https:// 3dprint.com/57751/china-eastern-airlines-3d-print/.. [91] G.D. Goh, et al., Additive manufacturing in unmanned aerial vehicles (UAVs): challenges and potential, Aerosp. Sci. Technol. 63 (2017) 140151. [92] J. Quincieu, et al., Case study: selective laser sintering of the USUSat II small satellite structure, Assembly Autom. 25 (4) (2005) 267272.
Review of additive manufacturing technologies and applications in the aerospace industry
29
[93] C. Majewski, H. Zarringhalam, N. Hopkinson, Effect of the degree of particle melt on mechanical properties in selective laser-sintered Nylon-12 parts, Proc. Inst. Mech. Eng., B: J. Eng. Manuf. 222 (9) (2008) 10551064. [94] H.-H. Tang, M.-L. Chiu, H.-C. Yen, Slurry-based selective laser sintering of polymercoated ceramic powders to fabricate high strength alumina parts, J. Eur. Ceram. Soc. 31 (8) (2011) 13831388. [95] M. Krznar, S. Dolinsek, Selective laser sintering of composite materials technologies, in: 21st International DAAAM Symposium, DAAAM International, Vienna, Austria, 2010. [96] F.P.W. Melchels, J. Feijen, D.W. Grijpma, A review on stereolithography and its applications in biomedical engineering, Biomaterials 31 (24) (2010) 61216130. [97] The Ultimate Guide to Stereolithography (SLA) 3D Printing, FormLabs, March 28, 2017. Available from: ,www.formlabs.com/blog/ultimate-guide-to-stereolithographysla-3d-printing/.. [98] W. Zhu, et al., Design and fabrication of stereolithography-based aeroelastic wing models, Rapid Prototyping J. 17 (4) (2011) 298307. [99] T. Wohlers, T. Gornet, History of Additive Manufacturing, 2016. [100] S.K. Moon, et al., Application of 3D printing technology for designing light-weight unmanned aerial vehicle wing structures, Int. J. Precis. Eng. Manuf.—Green Technol. 1 (3) (2014) 223228. [101] J.F. Stephen, B. George, S. Stefan, Design and fabrication of a bat-inspired flappingflight platform using shape memory alloy muscles and joints, Smart Mater. Struct. 22 (1) (2013) 014011. [102] M. Vaezi, et al., Multiple material additive manufacturing—Part 1: a review, Virtual Phys. Prototyping 8 (1) (2013) 1950. [103] Report, W.S., Additive Manufacturing and 3D Printing State of the Industry Annual Worldwide Progress Report, Wohler’s Associates, Inc., 2016. [104] Brickipedia, Acrylonitrile Butadiene Styrene, 2017. Available from: ,http://lego. wikia.com/wiki/Acrylonitrile_Butadiene_Styrene.. [105] Stratasys, Preparing for Takeoff—FDM Helps Bell Helicopter Build Quality Prototypes, 2015. [106] Stratasys, 3D Printing a Space Vehicle—NASA’s Human Supporting Rover Has FDM Parts, 2015. [107] H. Klippstein, et al., Fused Deposition Modeling for Unmanned Aerial Vehicles (UAVs): A Review, 2017. [108] C. Hauser, Case Study: Laser Powder Metal Deposition Manufacturing of Complex Real Parts, TWI, 2014. [109] LENS Blisk Repair Solution, Optomec, March 28, 2018. Available from: ,www. optomec.com/3d-printed-metals/lens-emerging-applications/blisk-repair/.. [110] RUAG, Certified for Universal Success: Additive Manufacturing of Satellite Components, 2017. [111] J. Harbaugh, Space Station 3-D Printer Builds Ratchet Wrench to Complete First Phase of Operations, 2014. August 07, 2017. Available from: ,https://www.nasa.gov/ mission_pages/station/research/news/3Dratchet_wrench.. [112] EOS, Future Ariane Propulsion Module: Simplified by Additive Manufacturing. 2017. [113] EOS, Durable up to the Sound Barrier and Beyond, 2017. [114] EBM Electron Beam Melting—in the forefront of Additive Manufacturing, Arcam, February 18, 2013. Available from: ,www.arcam.com/technology/electron-beammelting/..
30
Additive Manufacturing for the Aerospace Industry
[115] Seven Families of Additive Manufacturing—A Quick Reference Guide, Hybrid Manufacturing Technologies, January 5, 2018. Available from: ,www.hybridmanutech.com/resources.html.. [116] M. Rouse, Hybrid Manufacturing, 2017. [117] T. Halterman, ELB-Schliff’s New Mill Grind Device Melds 3D Printing and Grinding, 2015. [118] Stratasys, High-Quality Flying—China Eastern Airlines Explores Additive Manufacturing for Aircraft Maintenance, 2017. [119] Northrop Grumman Gets Quick Results From 3D Printed Fixture Tools, 3D Systems Corporation. Available from: ,www.3dsystems.com/customer-stories/northrop-grumman-gets-quick-results-stereolithography.. [120] Stratasys, Insourcing Delivers Impressive Efficiency Moog Aircraft Group Cuts Fixture Costs and Lead Times with FDM 3D Printing, 2018. [121] Advanced Aerials Streamlines Unmanned Vehicle System Development with 3D Printed Parts, 3D Systems Corporation. Available from: ,www.3dsystems.com/ learning-center/case-studies/advanced-aerials-removes-mystery-unmanned-vehiclesystem-development.. [122] SLS 3D printing delivers the lighter, better UAV with 3D Systems on Demand Manufacturing, 3D Systems Corporation. Available from: ,www.3dsystems.com/ learning-center/case-studies/lighter-better-uav.. [123] SME, Rapid tooling design, in: Fundamentals of Tool Design Study Guide. [124] S. Michael, Direct Versus Indirect Tooling and Beyond. 2007; cited 2018. Available from: ,https://www.additivemanufacturing.media/articles/direct-versus-indirect-tooling-and-beyond.. [125] V. Vashishtha, R. Makade, N. Mehla, Advancement of rapid prototyping in aerospace industry—a review, Int. J. Eng. Sci. Technol. 3 (3) (2011). [126] B. Stackpole, 3D Printed Plane Propels Wind Tunnel Testing to New Heights, June 1, 2014. Available from: ,http://www.digitaleng.news/de/3d-printed-plane-propelswind-tunnel-testing-new-heights/.. [127] Commercial Aircraft, Airbus Tests High-Tech Concepts With an Innovative 3DPrinted Mini Aircraft in: Commercial Aircraft, 13 June, 2016. Available from: ,http://www.airbus.com/newsroom/news/en/2016/06/airbus-tests-high-tech-conceptswith-an-innovative-3d-printed-mini-aircraft.html.. [128] F. Walachowicz, et al., Comparative energy, resource and recycling lifecycle analysis of the industrial repair process of gas turbine burners using conventional machining and additive manufacturing, J. Ind. Ecol. 21 (S1) (2017) S203S215. [129] Laser Repair Technology (LRT), RPM Innovations Inc, October 27, 2010. Available from: ,www.rpm-innovations.com/laser-repair.. [130] E.W. Rich Martukanitz, S. Kelly, T. Donnellan, Penn State Applied Research Lab Laser Additive Manufacturing Technology, 2010. [131] A. Gasser, et al., Laser additive manufacturing, Laser Tech. J. 7 (2) (2010) 5863. [132] M. Nikodinovski, Additive Manufacturing Repair Within the ARMY, 2015. [133] S.S. Babu, et al., Workshop Report on Additive Manufacturing for Large-Scale Metal Components—Development and Deployment of Metal Big-Area-AdditiveManufacturing (Large-Scale Metals AM) System, United States, 2016. [134] C. Qiu, et al., Fabrication of large Ti6Al4V structures by direct laser deposition, J. Alloys Compd. 629 (2015) 351361. [135] L.E. Murr, et al., Microstructural architecture, microstructures, and mechanical properties for a nickel-base superalloy fabricated by electron beam melting, Metall. Mater. Trans. A 42 (11) (2011) 34913508.
Review of additive manufacturing technologies and applications in the aerospace industry
31
[136] H. Krueger, Standardization for additive manufacturing in aerospace, Engineering 3 (5) (2017) 585. [137] S. Mellor, L. Hao, D. Zhang, Additive manufacturing: a framework for implementation, Int. J. Prod. Econ. 149 (2014) 194201. [138] J. Mardaras, P. Emile, A. Santgerma, Airbus approach for F&DT stress justification of additive manufacturing parts, Procedia Struct. Integrity 7 (2017) 109115. [139] T.-H. Kwok, et al., Four-dimensional printing for freeform surfaces: design optimization of origami and kirigami structures, J. Mech. Des. 137 (11) (2015) 111413111413-10. [140] Y.L. Yap, W.Y. Yeong, Shape recovery effect of 3D printed polymeric honeycomb: this paper studies the elastic behaviour of different honeycomb structures produced by PolyJet technology, Virtual Phys. Prototyping 10 (2) (2015) 9199. [141] J.E.M. Teoh, et al., Multi-stage responsive 4D printed smart structure through varying geometric thickness of shape memory polymer, Smart Mater. Struct. 26 (12) (2017). [142] M. Tomlin, J. Meyer, Topology optimization of an additive layer manufactured (ALM) aerospace part, The Seventh Altair CAE Technology Conference 2011, 2011. [143] B. Stanford, P. Ifju, Aeroelastic topology optimization of membrane structures for micro air vehicles, Struct. Multi. Optim. 38 (3) (2009) 301316. [144] N.K. Dey, F.W. Liou, C. Nedic, Additive manufacturing laser deposition of Ti6Al4V for aerospace repair applications, in: 24th International Solid Freeform Fabrication Symposium—An Additive Manufacturing Conference, SFF2013, August 1214, 2013, Austin, TX, University of Texas at Austin (freeform), 2013. [145] J. Pottebaum, et al., Future Repair and Maintenance for Aerospace Industry, RepAIR, 2016, p. 89.
This page intentionally left blank
Qualification and certification of metal additive manufactured hardware for aerospace applications
3
Richard Russell1, Douglas Wells2, Jess Waller3, Behrang Poorganji4, Eric Ott4, Tsuyoshi Nakagawa5, Hector Sandoval6, Nima Shamsaei7 and Mohsen Seifi8 1 NASA Engineering and Safety Center, Kennedy Space Center, FL, United States, 2NASA Marshall Space Flight Center, Huntsville, AL, United States, 3NASA-Johnson Space Center White Sands Test Facility, Las Cruces, NM, United States, 4General Electric Additive, West Chester, OH, United States, 5JAXA Safety and Mission Assurance, Tokyo, Japan, 6 Lockheed Martin Missiles and Fire Control, Grand Prairie, TX, United States, 7National Center for Additive Manufacturing Excellence, Auburn University, Auburn, AL, United States, 8ASTM International, Washington, DC, United States
3.1
Introduction
Additive manufacturing is revolutionizing the traditional aerospace part design and manufacturing paradigm. For existing designs, additive manufacturing offers the ability to reduce cost, especially for one-of-a-kind or limited production run quantities. For new designs, high cost and long lead times associated with the production of complex hardware by conventional manufacturing routes have convinced manufacturers to rely on meticulous analyses to mitigate or eliminate the chance of failure. With the advent of additive manufacturing, prototype hardware designs can be iterated early in the design cycle with minimal cost and impact to schedule, restoring the role of incremental testing and iterative redesign. It is anticipated that by using metal additive manufacturing processes, aerospace companies will be able to produce essential, but otherwise unavailable, on-demand parts of simple design in days rather than months once process and part qualifications have been performed. Significant reductions are also anticipated for the design, development, test, and evaluation (DDT&E) time for more complex aerospace components and systems such as commercial aviation gas turbine engines [1]
Disclaimer: The views presented in this paper are those of the authors and should not be construed as representing official rules interpretation or policy of ASTM International, General Electric (GE), Lockheed Martin Corporation (LMCO), National Aeronautics and Space Agency (NASA), Japanese Space and Exploration Agency (JAXA), or the Federal Aviation Administration (FAA).
Additive Manufacturing for the Aerospace Industry. DOI: https://doi.org/10.1016/B978-0-12-814062-8.00003-0 © 2019 Elsevier Inc. All rights reserved.
34
Additive Manufacturing for the Aerospace Industry
or spaceflight rocket engines [2]. For example, in National Aeronautics and Space Administration (NASA) rocket engines, which can contain a variety of metal additive manufactured (AM) components including turbopumps, injectors, combustion chambers, and nozzles, reductions are anticipated in the DDT&E time (710 years to 24 years), hardware lead time (36 years to 6 months), and costs (order-ofmagnitude reductions possible) [2]. Other benefits are also anticipated, namely reduced weight, lower subassembly part counts, smaller inventories, high levels of geometric complexity, and better performing topology of optimized parts [3]. However, to realize these gains and exploit the full potential of additive manufacturing, robust quality control and qualification procedures along with clear interpretation of certification requirements are needed, especially for fracture critical metal AM hardware. Because the concepts of qualification, certification, and quality control are related to one another and are frequently not easily separable in the literature on this subject, this chapter will use the terminology Q&C (qualification and certification) when the three concepts are being considered as a group, but will occasionally refer to the individual terms when differentiation is needed. Despite the lack of publicly available Q&C procedures for AM hardware, some aerospace companies and organizations have begun, or are planning, to install AM hardware in aircraft and spacecraft. To enable this advancement, internal proprietary processes for qualification and quality control have been developed by some organizations. On the commercial aviation side, General Electric (GE) has received Federal Aviation Administration (FAA) approval to fly a metal AM compressor inlet temperature sensor housing in its GE90 jet engines [4], has dual certified with the FAA and the European Aviation Safety Agency metal AM fuel nozzles for its Leading Edge Aviation Propulsion (LEAP) engine and has subsequently ramped up component production and engine shipment [5]. On the noncivilian side, NASA is actively considering the use of metal AM components in its rocket engines [6,7]. In these types of applications, it is essential to manage risk with appropriate Q&C procedures. Some examples of AM metal hardware used or slated for use in civilian and noncivilian aerospace applications are shown in Figs. 3.1 and 3.2, respectively. While additive manufacturing is increasingly being used in the development of new products in the aerospace sector, a universal understanding of the contributions and control methodologies for dimensional tolerances, anomalies such as pores and voids, microstructural variations, higher than desired surface roughness, and residual stress along with their potential effect on part acceptability and mechanical properties are still developing. If not fully understood for the material and part being implemented by the component and system owner, efforts to manage performance risks can limit a part’s use especially in high-value or mission-critical applications. In some cases, engineering inexperience in the technology area, by itself, may also contribute to a system owner’s conservatism in design and application. Like many other conventional manufacturing processes, variation in part quality and mechanical properties exists but can be minimized by implementing appropriate process controls. These process controls are preemptive measures that mitigate or eliminate factors known to influence part quality and the finished part’s properties. Process control methodologies are known and widespread in the production of
Qualification and certification of metal
35
Figure 3.1 FAA-approved T25 compressor inlet temperature sensor (left), and fuel nozzle for the CFM International (partnership between GE Aviation and Safran Aircraft Engines) LEAP jet engine (right), both employing GE additive manufacturing processes. GE, General electric; LEAP, leading edge aviation propulsion. Source: Adopted from T. Kellner, The FAA cleared the first 3D printed part to fly in a commercial jet engine from GE, in: GE Reports, ,https://www.ge.com/reports/post/ 116402870270/the-faa-cleared-the-first-3d-printed-part-to-fly-2/., 2015; T. Kellner, Mind meld: how GE and a 3D-printing visionary joined forces, in: GE Reports, ,https://www.ge. com/reports/mind-meld-ge-3d-printing-visionary-joined-forces/., 2017.
Figure 3.2 Additive manufactured injectors (left) and combustion chambers (right) fabricated and tested by the NASA Marshall Space Flight Center. Source: Adopted from P.R. Gradl, S.E. Greene, C. Protz, J. Buzzell, C. Garcia, J. Wood, et al., Additive manufacturing of liquid rocket engine combustion devices: a summary of process developments and hot-fire testing results, in: 2018 Joint Propulsion Conference, AIAA Propulsion and Energy Forum (AIAA 2018-4625). ,https://arc.aiaa.org/doi/abs/ 10.2514/6.2018-4625..
36
Additive Manufacturing for the Aerospace Industry
high-quality parts and it comes as no surprise that these are key to manufacturing processes such as additive manufacturing. Because of the intimate link between process conditions and the development of material structure and quality during the additive deposition operation, the necessity of defining appropriate levels and types of process control becomes important for many additive parts. The factors (and metrics) most commonly implicated for variation of quality and properties in additive manufacturing are: G
G
G
G
G
G
G
G
G
Feedstock attributes (purity, powder particle shape and size distribution, and chemistry). Processing conditions and controls (laser or electron beam power, hatch width, and scan rate). Thermal conditions during build (layer thickness and platform preheating). Build atmosphere and purity (shield gas or high vacuum). Post-processing [Hot Isostatic Pressing (HIP), heat treatment, and machining]. Finished part properties (microstructure, discontinuities, roughness, and nondestructive and destructive test methods). Equipment (machine to machine variation, calibration, and maintenance). Personnel (training and certification). Facilities (certification).
While a robust Q&C program will address the above factors, it must be pointed out that several technology areas are being actively developed that hold the promise of more effective control of the processingstructureproperty envelope for a given AM process. These areas are: (1) integrated design approaches for materials, processes, and parts; (2) physics-based models relating to process, microstructure, and properties; and (3) closed-loop, in-process monitoring methods and improved process analytics for detecting real-time material and process anomalies [8,9]. While these areas are highly desirable for the qualification path and in an ongoing quality control plan, additional research is needed before being fully incorporated into future metal AM Q&C methods. At the core of this future Q&C push, lies the premise that the source(s) of process-induced discontinuities and microstructural heterogeneity must first be understood, thresholds of acceptability determined, and then mitigated by control of the AM process(es). Also, since location-dependent properties in as—built AM parts are affected by complex interactions between discontinuity—dominated and microstructure-dominated failure mechanisms, the interaction between discontinuities and microstructure needs to be better understood [10,11]. This highlights the need for a comprehensive examination of various factors (e.g., discontinuities, microstructure, and residual stress, etc.) controlling the mechanical behavior of AM materials, while also integrating modeling and experimentation efforts to produce AM parts with the desired location-specific properties [1214]. In the past five years, road mapping activities by the National Institute of Standards and Technology (NIST) [15] and NASA [16] have echoed and expanded on the above Q&C needs and challenges related to metal AM materials, processes, and parts. For example, the 2013 NIST metals-based AM roadmap identified four key Q&C needs impeding largescale deployment of AM; namely, closed-loop
Qualification and certification of metal
37
process control, standardized Q&C guidelines, shared databases, and established protocols for making reproducible parts. The 2014 NASA roadmap identified needs specific to nondestructive evaluation (NDE), such as developing standardized NDE methods tailored specifically to metal AM parts, obtaining a better understanding of effect-of-defect on finished part properties, and adopting quantitative NDE accept/ reject criteria. Although significant progress has been made in each of these areas since 2014, NDE-based Q&C of metal AM hardware is still evolving. Existing NASA quality documentation [17,18] calls out NDE for Q&C of spaceflight hardware made by laser-powder bed fusion (L-PBF) as part of the general part production quality control requirements. However, procedural details of the actual NDE methods used on the shop floor are not given. This is despite the fact that NDE is a means to accept/reject a L-PBF part and can determine whether a part is approved for service or returned to an internal material review board for final disposition (which will be addressed in Section 3.6). Fortunately, the state-of-the-art of currently available NDE methods that have application to AM hardware is rapidly evolving. Some of the more established methods are computed tomography (CT), eddy current (ET), metrology (MET), process compensated resonance testing (PCRT), penetrant testing (PT), radiographic testing (RT), and ultrasonic testing (UT). These methods have been captured in a soon-to-be-completed draft of the American Society of Testing and Materials (ASTM) standard [19]. While production of an AM part can be conducted within one company (from initial product design through to processing and post-processing, and finally to the finished product), qualification of such a part for use in civilian and noncivilian applications can require the collection of a large amount of labor-intensive processing, microstructure, and property-based information. In many cases, data must be collected by the Original Equipment Manufacturer (OEM) from many different companies, who must then make the information available to pave the path for eventual Q&C. However, there are many potential intellectual property hurdles in both the generation and analysis of such data [10]. One approach that can provide the necessary infrastructure to accelerate Q&C involves the use of an Integrated Computational Materials Engineering (ICME)-based platform (MiCloud.AM) to develop an understanding of the microstructure, potential defects, and source(s) of defect generation [20]. These are all influenced by process control while linkages between those variables and melt pool geometry are required to eliminate/minimize such defects and produce the desired properties. While ICME is not covered in this chapter, it does warrant mention.
3.2
Special considerations for fracture-critical hardware
The level of criticality of AM aerospace parts is expected to increase as this technology matures and gains widespread acceptance [21]. As the use of metal AM parts in fracture-critical applications becomes more accepted, the need for Q&C
38
Additive Manufacturing for the Aerospace Industry
standards covering all aspects of the part lifecycle becomes more prevalent. Fracture-critical properties are affected by a variety of factors and two of the dominant factors are discontinuity density and microstructural variation. While the goal of producing defect-free parts in as-deposited materials remains an area of extreme interest, inspection procedures (e.g., CT, MET, PCRT, and UT, etc.) and various post-processing techniques (e.g., HIP and heat treatment, etc.) may continue to be needed in fracture-critical applications. However, noncritical locations with no structural demand (i.e., low stress or strain) in fracture-critical parts may not require the same damage tolerance or properties as required in highly stressed areas. The type and rigor of Q&C procedures will be driven by the relevant industrial sector and end-use application (aerospace, defense, medical, energy, or automotive). For example, the Q&C requirements for high-value, limited quantity production run aerospace parts will be more stringent than for high quantity production run commodity parts. These requirements will, in large part, be attributable to the unique safety concerns associated with human space flight and with military and commercial aviation which impose added rigor and stringency than is justified in other sectors. As is noted elsewhere [3], perhaps the key challenge confronting NASA [17,18], the FAA, Department of Defense (DoD), and the commercial aerospace sector [21] is the qualification of fracture-critical AM parts using either inspection (NDE) or testing, especially in applications where structural margins are low and the consequence of failure is high. Such parts use a damage-tolerant rationale and require careful attention. Currently, it is not clear that defect sizes from NASASTD-5009 [22], which were derived from conventionally made metal hardware, are applicable to AM hardware, particularly when the as-built AM part surface is still present and surface-sensitive NDE techniques such as ET and PT are used. To quantify the risks associated with these parts, it is incumbent upon the structural assessment community, such as the ASTM Committee E08 on Fracture and Fatigue, to define critical initial flaw sizes (CIFS) for the part in order to establish the objectives of the NDE. Specifically, more effort needs to be focused on the characterization and understanding of fatigue and fracture properties of AM materials and the corresponding testing methodologies. In addition to “conventional” crystallographic fatigue crack initiation mechanisms in homogeneous substrate materials, crack initiation due to the presence of inherent AM material anomalies such as porosity, lack of fusion defects, or inclusions also needs to be considered [2329]. It should be noted that the characterization of flaws in fracture-critical AM parts needs to be based on realistic variation in material properties, microstructure, and material defect characteristics representative of the full-scale production environment. Failure to do so may result in “lessons learned” similar to those experienced during the early days of powder metallurgy (PM) when an undetected nonmetallic inclusion in a PM turbine disk was found responsible for the failure of a fracturecritical component that caused the crash of an F-18 aircraft [21].
Qualification and certification of metal
3.3
39
Current qualification and certification state-of-theart and gap analysis
Broader Q&C standardization needs for AM have been recently addressed in a voluntary consensus standards (VSC) gap analysis performed by the America MakesAmerican National Standards Institute (ANSI) Additive Manufacturing Standardization Collaborative (AMSC) [30]. Federal agencies, including NIST, NASA, the DoD, FAA, and others, as well as several standards development organizations (SDOs) were instrumental in the formation of this collaborative. SDOs whose scope of work directly or indirectly relates to aerospace AM standardization include the American Society of Mechanical Engineers (ASME), ASTM International, American Welding Society (AWS), International Organization for Standardization (ISO), Metal Powder Industries Federation (MPIF), and the Society for Automotive Engineers (SAE) International. This collaborative recognized the need for AM standards and conformance procedures to advance the adoption of AM technologies in the United States which, to date, has been largely dependent on proprietary OEM specifications. Unfortunately, AM does not have the benefit of years of incremental refinement by third-party practitioners and OEMs. To counter some of the inefficiencies and shortcomings associated with a reliance on OEMs and to develop technologies having a broader national impact, America Makes and the Penn State Center for Innovative Materials and Processing-3D (CIMP-3D) cosponsored a technical exchange meeting in October 2015 to coordinate United States’ standards development activities for AM [31]. This meeting and initiative led to the formation of America Makes-ANSI AMSC, culminating in a recently released standardization roadmap [30]. The AMSC roadmap can be viewed as a tool to help focus and combine resources where possible, by bringing together stakeholders in the planning and development of industrywide standards and related research and development (R&D) activities to the extent R&D is needed. The standards identified by the AMSC are deemed important for use by industry during qualification of AM materials, processes, and systems, as well as use by regulatory bodies during certification of AM parts. To develop the roadmap, the AMSC took the approach of conducting a lifecycle assessment of an AM part, from initial design, through production, and ending with post-production testing, qualification, and maintenance. Thus, the AMSC initially organized itself around five primary working groups covering Design, Process and Materials, Q&C, NDE, and Maintenance. The Process and Materials group was further divided into four subgroups covering Precursor Materials, Process Control, Post-Processing, and Finished Materials Properties. This approach allowed Q&C and Q&C-related standardization gaps to be identified.
3.3.1 Standardization gaps related to qualification and certification Whereas metal AM parts are tested for performance much the same as traditionally manufactured cast, wrought, and forged items, there are aspects unique to AM that
40
Additive Manufacturing for the Aerospace Industry
must be addressed before such parts are deployed for service. This is especially true for mission- and safety-critical parts and applications. For example, a critical part may be required to be built from qualified materials, using qualified processes, etc. There are many types of qualifications that can be discussed within the scope of AM. As such, Q&C is a major focus area for AM parts [15]. In the most recently released roadmap [30], 95 standards are identified that are either in progress or identified for development. Of those standards, 11 are listed in Table 3.1 according to priority that deal specifically with Q&C of aerospace AM parts. While this list is noninclusive, it is representative of the standardization needs of the aerospace sector. Standardization needs for the medical sector [30] which may have some application to the aerospace sector such as protocols for image accuracy (QC7), personnel training for image data set (QC9), and verification of the 3D model (QC10) have not been included in Table 3.1.
3.3.2 Recent directions in qualification, certification, and quality control for additive manufacturing Input was invited from AMSC participants on relevant qualification procedures [30], allowing current Q&C guidance documentation to be identified. For example, aviation, space, and defense organizations must provide, and continually improve, safe and reliable products and services that meet or exceed customer and applicable statutory and regulatory requirements. To accomplish this goal and to help improve and sustain its overall quality performance with respect to the products and services provided, an organization can adopt a quality management system (QMS). A QMS commonly used by the aerospace industry is SAE AS9100 [33] (or an approved equivalent). A QMS is required to ensure necessary process controls and to mitigate risks associated with noncompliance, especially in cases where there is significant reliance on process controls for the reliability of the product. In addition to a QMS, another key resource that aerospace companies use to control their vendor supply chains is the National Aerospace and Defense Contractors Accreditation Program (NADCAP) [34]. NADCAP is an industry-managed program administered by the Performance Review Institute (PRI) devoted to improving quality and reducing costs of special process accreditations throughout the aerospace and defense industries. Prior to NADCAP, aerospace companies audited vendors in their supply chain using their process requirements to verify compliance. Since the processes used by vendors were often similar or identical, this led to duplicate audits that added to the workload without adding value. In October 2013, the Welding Task Group was assigned the responsibility to assess AM industry needs and develop audit criteria specifically for AM vendors, focusing on L-PBF and EBPBF processes [30]. Draft checklists were developed, culminating in the approval of the checklist AC7110/14 [35] in early 2017. Since then, audits have been performed and suppliers accredited to the AC7110/14 checklist. Based on early comments, future revisions are planned. In addition to the AC7110/14 checklist, a core checklist relevant to AM, AC7110, is also available [36]. The NADCAP checklists
Qualification and certification of metal
41
Table 3.1 Prioritized list of standardization gaps related to quality control, qualification, and certification of aerospace partsa Gapb
Recommendation
Priorityc
Organization(s)
Harmonization of AM Q&C Terminology (QC1)
Adopt uniform terminology for the terms qualification, certification, verification, and validation captured in ISO/ASTM 52900 [32] Develop a part classification system to describe the level of risk associated with a part and may be used as a metric to gauge appropriate qualification requirements Develop standards identifying the means to establish statically validated minimum mechanical properties for metals made using a given set of AM parameters for a given design Develop standardized terminology for process parameters for use across all AM equipment, incorporated into ISO/ ASTM 52900 [32] Issue guidelines that require consistent post-processing for the various AM processes to be applied for qualification and production builds Starting with the most mature technologies, such as L-PBF, develop standards that assess required checks for levels of criticality and safety as part of the DoD procurement process. DoD should participate in the development of such standards and specify the certification requirements needed
High
ASTM F42/ISO TC 261, ASME, SAE
High
NASA, AWS
High
ASTM F42/ISO TC 261, SAE, AWS, MMPDS, NIST
Medium
ASTM F42/ISO TC 261, AWS, SAE
Medium
AWS, ASTM 42/ ISO TC 261, SAE
Medium
ASME, ASTM F42/ISO TC 261, DoD, Industry, SAE
AM Part Classification System (QC2)
Material properties (FMP1)
Harmonizing Q&C terminology for process parameters (QC3)
Post-processing qualification and production builds (P1)
Process approval for DoD-procured parts (QC4)
(Continued)
42
Additive Manufacturing for the Aerospace Industry
Table 3.1 (Continued) Gapb
Recommendation
Priorityc
Organization(s)
Machine qualification (PC4)
Develop guidelines for control of machine-to-machine variability using broader part-specific, processspecific, material-specific, and application-specific practices beyond fit and fit Develop industry standards to allow combination of additional or complementary NDE data to provide a simple, unified interpretation of results Develop materials- and process-specific guidelines for creating and using phantoms used to check the accuracy of a given process, based on what is being imaged and the modality in use (for example, CT or UT) Develop AM operator training and qualification standards or guidelines. Training should cover the various AM materials and processes available in the market and be performance based to ensure consistent AM part quality Develop a standard AM file format building on STL and AMF that can represent all applicable slice files, build path, materials, density, orientation, etc., into a single file format
Medium
NIST, AWS, SAE, ASTM F42, DoD, NASA
Medium
ASTM
Medium
ASTM, ISO
Low
NASA, SAE, AWS, OEMs, ASTM F42/ ISO TC 261
Low
ASTM F42 and ISO/TC 261
Data fusion (NDE5)
Phantoms (QC8)
Machine operator training and qualification (QC5)
Neutral build file dormat (D20)
a
AM, Additive manufacturing; ANSI, American National Standards Institute; ASME, American Society of Mechanical Engineers; ASTM, American Society for testing and Materials; AWS, American Welding Society; CT, computed tomography; DoD, Department of Defense; ISO, International Organization for Standardization; NASA, National Aeronautics and Space Administration; NDE, nondestructive evaluation, NIST, National Institute of Standards and Technology; OEM, Original Equipment Manufacturer; Q&C, qualification and certification; SAE, society for automotive engineers; STL, stereolithography; TC, technical committee; UT, ultrasonic testing. b Analogous gaps in Ref. [27] given in parentheses. c Rankings based on four criteria: (1) criticality, (2) achievability, (3) scope, and (4) effect [30].
Qualification and certification of metal
43
are used by some aerospace prime vendors. Basic elements of the Q&C approach used by aerospace primes GE and Lockheed Martin are discussed in Sections 3.4.1 and 3.4.2, respectively. The Metallic Materials Properties Development and Standardization (MMPDS) Handbook [37], which was formerly published as MIL-HDBK-5 (obs.), is an accepted source for metallic material and fastener system “A” and “B” basis design allowables recognized by the FAA, DoD, and NASA. The MMPDS has had limited exposure to a few AM materials, beginning with AMS 4999 [38]. However, at present, no metal AM alloys are included in the Handbook. Inclusion of an alloy and material form in the MMPDS is predicated on two major factors: (1) the existence of public industry specifications; and (2) the expectation that a single set of defined property limits representative of common process knowledge can be reliably established. In the case of AM, public industry material and processing specifications have not generally been available, and datasets submitted to Battelle have been judged inadequate for deriving publishable design allowables. Specifications for materials and processes are now beginning to emerge from ASTM, SAE, and ISO, for example. Common process knowledge sufficient to enable universally accepted (nonproprietary) capability minimums for AM processes have not yet been forthcoming. Currently, the MMPDS organization is assessing whether AM materials could be published in a separate volume or document with special guidance more suited to AM materials. Lastly, the objectives and schedules of programs has created a significant pull on engineering organizations in industry and government to establish AM requirements as the process understanding evolves. While it is clear that industry standards from SDOs will eventually play a key role in governing the AM process for NASA spaceflight hardware, for example, none have yet become sufficiently mature to adopt independently. Given the significant need to frame the AM requirements in the context of NASA’s overarching standards for materials, structures, and fracture control, NASA Marshall Space Flight Center (MSFC) has developed two quality documents to meet existing materials, structures, and fracture control requirements for AM parts produced with the L-PBF [17,18]. Some of the more salient features of the MSFC quality documents will be addressed in more detail in Section 3.5.1.
3.4
Industry qualification and certification approaches
3.4.1 General Electric qualification and certification approach As part of the industrialization of manufacturing processes, materials and ultimately components in most fields of use, the qualification of the material, process, and part are necessary steps in defining, validating, and establishing a production product. Qualification processes are typically formal validation and approval processes that follow process and material development stages. The ongoing verification of continued acceptability with time relies on the closely connected quality control plan. In some fields of use such as aviation, defense, medical, healthcare, power
44
Additive Manufacturing for the Aerospace Industry
Figure 3.3 Qualification and certification processes relating to materials, processes, and parts in the manufacturing environment.
generation, and others, an additional certification process may also be required. The qualification process and subsequent quality control processes typically extend across organizational/functional boundaries within a producer and between a supplier and customer/part integrator. Certification is more often between a product producer and a regulatory body. A schematic of Q&C processes is shown in Fig. 3.3. The attributes of an ongoing quality control regimen for a given AM part are made up of the QMS required by the system OEM, definition of critical process control variables, definition of product control measures, agreement on quantifiable and meaningful process, and product attribute control levels which differentiate between acceptable and unacceptable products, and measurement/monitoring frequencies.
3.4.1.1 Qualification of additive materials In new technologies such as additive manufacturing, it is essential to validate and demonstrate process maturity and controls that would be the basis of generating specifications and process documents. This will facilitate and accelerate the development of analysis methods and associated design allowables that can be constructed using statistical analysis techniques common in the industry (e.g., the MMPDS Handbook [37]) or by company-specific property data representation rules. In the additive manufacturing field, similar Q&C processes are being used to define requirements and to establish that attributes meet internal, customer, and regulatory requirements. The procedures used are similar to those for other manufacturing processes like
Qualification and certification of metal
45
casting, forging, and machining, etc., and just like in these conventional manufacturing approaches, each has particular product and process requirements, methodologies, and sensitivities that must be considered in assuring that high quality and repeatable products are produced. As additive processing has begun penetrating mainstream production, particularly in high-technology products, the industry has begun to develop a more complete understanding of the scope of qualifications and certifications that are needed. In additive manufacturing, the feedstock (powder or wire material) transforms as the build progresses. Therefore, material qualification should involve the feedstock, melting-solidification transformation processes, and postprocessing (heat treatment, HIP, etc.) requirements. Recently, the FAA released a draft advisory circular [39] for public comment relating to certification approaches for additive-produced aviation engine components. This document highlights a potential path (but not the only path) to certification of flight hardware. Since the certification process is essentially the endpoint of a series of subtler qualification processes, the content offers insight into the qualification steps that are anticipated for AM processes and materials. A summary of typical qualification steps that a part needs to include from this document are: G
G
G
G
G
G
G
G
G
G
G
Feedstock attributes, chemistry, morphology versus acceptability windows and specific vendor qualifications. Powder handling, use, reprocessing, and reuse limits and controls. AM process parameters, controls, and windows for each material and machine type. Deposition environment controls and windows for each facility and machine type. Validation of material sample performance versus design allowables and specification requirements. Surface modification process evaluation. Support structures and support and powder removal process qualification. Post-processing controls including heat treatment. Component and witness part inspection using NDE. Component acceptance tests. Component cut-up assessment.
The strong linkage of AM processing to the structure and properties of the material and part has driven the need for more careful consideration of raw materials, processes, and equipment controls. An example is shown in Fig. 3.4 of the variations in structure of nickel alloy UNS N07718 produced by an L-PBF process, followed by differing post-processing treatments helps to demonstrate the range of behaviors that may result (and variability if processes are not controlled) [40,41]. These effects may be perpetuated through heat treatment. Overall, these effects reflect layer-by-layer processes creating structure and performance capability, not merely part shape. A necessary, but insufficient, condition for the production, qualification, and certification process is the establishment of specifications that control methods and acceptability standards for materials, processes, and parts are key to the definition and control for production additive manufacturing. Frequently these specifications
46
Additive Manufacturing for the Aerospace Industry
Figure 3.4 Examples of microstructural variation observed in laser powder bed nickel alloy UNS N07718 depending on deposition and post-processing treatment showing: (A) an as-deposited structure; (B) a mixed directional and equiaxed structure; and (C) an equiaxed structure produced by variations in thermal post-processing [40,41].
are proprietary to a company, but are shared with a company’s supply base. In the additive field, SDOs have initiated significant activities described in Ref. [30] to identify needed publicly available specifications for materials, processes, and parts. Some examples of these include efforts by the SAE-AMS Additive Manufacturing Committee, ASTM International Committee F42, ISO Technical Committee 261, and others. The establishment of these types of accessible, consensus-driven requirements documents are beneficial for the more widespread implementation of new processes and for setting baseline expectations on level of capability and reliability to bolster technology adoption and prevalence. An outline of typical qualification steps for an aviation-type, high-quality component is: 1. Completion of development of material (alloy) and process leading to a review and approval of technology and method maturity and fitness for production using a gated review process, such as a tollgate or readiness level review. An example may be the “technology readiness level” and “manufacturing readiness level” type ratings similar to those defined by NASA and other organizations [42,43]. This is typically a predecessor to a qualification process, but is necessary to provide the technical understanding to define specifications, requirements, and essential controls to assure repeatability. The development process is also key to providing mechanical performance data and representations of the variation in data contributed by raw materials, processes, structure, size, and
Qualification and certification of metal
2.
3.
4.
5.
6.
7.
47
geometry, etc., suitable for part design and analysis for service. This portion of the development and initial qualification process may involve thousands of hours of technical effort, tens to hundreds of process trials, hundreds to thousands of test specimens, and demonstration parts to complete. Timelines may be a few months to several years depending on intensity of effort, historical experience, complexity, and field of application. Qualification of the raw material (powder, wire, etc.), process, and specific production vendor to validate both the long-term capability to meet any raw material specification requirements, but also to demonstrate the fitness of the raw material for use by the additive process and to establish sufficiency of a vendor’s fixed/controlled process limits in meeting the overall engineering/manufacturing intent. Qualification of a manufacturing source’s compliance in terms of quality system, controls, procedures, and facilities for production of certain types of components to industry norms, such as through ISO, SAE AS9100 [33], and NADCAP [44]. Qualification of individual machines and procedures/process limits by trained process experts to assure that specific equipment is capable of being used for production manufacturing. The criteria may extend the gamut from spatial precision of optics, settings, and tolerance/control limits to demonstration material specimens for evaluation. Qualification of critical process steps and related machines and procedures for steps that are considered significant in terms of helping to define the part acceptability, performance, and quality. For AM, this includes qualification of specific additive manufacturing machines, operating procedures, deposition parameter settings and ranges, and even maintenance and calibration requirements. Qualification of the final component or subcomponent attributes versus engineering and quality requirements, drawings, and specifications. This also includes the definition and approval of fixed or frozen processes which assures that processes, once qualified and approved, are not changed without consideration of potential partial or full requalification. Frequently, this qualification step involves nondestructive and destructive evaluations of initial production parts that have been produced to meet all the relevant requirements. Attributes may also be evaluated and monitored over time to assure ongoing quality and reproducibility as part of the qualification process. Regardless, key evaluations are also then applied to the ongoing quality control plan to assure acceptability of parts over time. Once fully qualified, any changes to the above qualification steps require review and potentially may need requalification.
Although more extensive than may be necessary in some application fields outside of aviation, elements of the above qualification process for additive (and other nonadditive) parts are equally applicable in other fields.
3.4.1.2 Certification of additive materials Downstream certification steps may make use of some or all of the qualification steps’ results to demonstrate acceptability to certifying organization or agency requirements. Some example regulatory requirements for certification of aviation airframe materials and structures include 14 Code of Federal Regulations (CFR) Part 33 [45] for aircraft engines, 14 CFR Part 23 [46] for general aviation airplanes, and 14 CFR Part 25 [47] for commercial transport airplanes, Boiler and Pressure Vessel Code [48] for power generation and pressure vessel type components, and US Food and Drug Administration Federal Food, Drug, and Cosmetic Act [49] for medical devices. In general, the
48
Additive Manufacturing for the Aerospace Industry
certification requirements of a regulator pertain to the final product attributes and performance and are less often related to the details of the processes by which products are made. As a result, the continual evolution and advancement of manufacturing process technology does not necessarily result in a change or addition to regulatory requirements. The potential for significant disparities in the understanding, application, and level of control of new processes can create a need for an increasing level of awareness and technical engagement by regulatory bodies and may also lead to the development of regulatory body minimum expectations on the level of disclosure by the part owner to allow validation of meeting the regulatory requirements.
3.4.1.3 Quality control in additive materials Primary objectives of quality control are to assure that the process and part meet the acceptance requirements and that the capability level of the process and part are controlled and represent those verified in the qualification process and do so repeatably over time. Additional process control guidance from GE for AM is shown in Figs. 3.5 and 3.6 [50]. The process controls shown in Fig. 3.5 parallel analogous controls exercised by NASA MSFC’s quality documents for equipment, materials, and processes discussed in Section 3.5.1.1. Similarly, the use of nondestructive and destructive tests by GE in its Q&C activities, as shown in Fig. 3.6, mirrors NASA MSFC’s combination of inspection and test to develop an integrated rationale for structural integrity of a part, also discussed later. Measurement and monitoring of process and part quality occurs at a frequency determined by the sensitivity of the process and material, initial process/part repeatability, and the application type, criticality, and risk management needs. Transition to a more integral use of process monitoring, analytics, in situ
Explicit controls on machine, material and process (Examples shown, not an all inclusive list) Powder specification
Laser parameters
• Powder source • Powder size • Powder composition • Powder reuse procedures
• Spot size • Laser power • Laser travel speed • Laser dwell time
Calibration & maintenance • Preventive maintenance • Pre-build calibration • Factory environ controls
Post processing • Mechanical finishing • Thermal exposure
Hatch strategy • Contour pass • Sky writing • Line spacing or boundary overlap
Recoat parameters • Layer thickness • Recoater arm material • Recoater arm design
Thermal processing
Build chamber
• HIP cycle parameters • Heat treat/solution atmosphere • Braze HT parameters • Solution temperature
• Build atmosphere • Purge gas • Airflow • Preheat temp • Interpass temp
STRESS CAPABILITY
Qualify
GE
’s m
ate
ria
Ind
l
us
try
CYCLES TO FAILURE Contour Hatch
Laser parameters
Figure 3.5 Explicit Q&C controls for equipment, materials, and processes [50]. Q&C, Qualification and certification.
Qualification and certification of metal
Validate
49
Inspection and quality touch points
Qualification Machine Material Process
Inspection Part
Non-destructive
Functional checks
CT/VCT/CMM
Defect Dimensional recognition
Validate
Destructive
Fuel flow airflow other
Prof pressure ultimate test
Part
Cut-ups
Specimen
Grain size porosity surface finish
On-going material testing
In-situ inspection, modeling and analytics
Computational materials engineering
Meltpool process monitoring
Faster • Material development • Process qualification • Part certification
PREDIX Analytics
Figure 3.6 Summaries of GE Q&C activities for additive manufactured parts, including qualification, inspection, and testing (top) and computational modeling, in-process monitoring, and analytics approaches (bottom) [50]. Q&C, Qualification and certification; GE, General Electric.
inspection, and material modeling are being pursued and the technologies are advancing rapidly to support future integration with Q&C methodologies.
3.4.2 Lockheed Martin qualification and certification approach Lockheed Marin Corporation (LMCO) was an early adopter of additive manufacturing. Uses today range in technical complexity from trade-show models or rapid prototyping, tooling and test fittings, to production applications. The pace for implementing additive manufacturing has been driven by the program requirements,
50
Additive Manufacturing for the Aerospace Industry
Figure 3.7 A Lockheed Martin engineer inspects a 3D printed 1.16 m-wide dome prototype made by an Electron Beam Additive Manufacturing process [51].
cost target, and customer needs. In several cases, use of AM technology has reduced risk, cut costs, and helped LMCO get new products into the market more quickly. LMCO’s accomplishments in additive manufacturing are notable. Most recently, LMCO fabricated 1.16 m-wide titanium domes (Fig. 3.7) to cap off satellite fuel tanks and in so doing reducing waste, cost, and time of production [51]. The domes, which were made by an Electron Beam Additive Manufacturing [EBAM (EBAM is a registered trademark of Sciaky, Inc., Chicago, IL 60638)] process, are some of the largest AM parts made to date and completed final rounds of quality testing in July 2018. Since a small leak or flaw could be catastrophic for a satellite’s operations, LMCO engineers and technicians rigorously evaluated the structure, conducting a full suite of tests to demonstrate high tolerances and repeatability [51].
Qualification and certification of metal
51
Details describing the basic elements of LMCO’s Q&C approach are given elsewhere and are driven by part functionality (primary or secondary structure, or nonstructural), criticality (fracture critical or nonfracture critical), and customer requirements (variable) [52,53]. For example, a part used in a critical structural application would require more stringent part acceptance testing than a part used in a nonstructural application or a prototype part with no adverse consequence of failure [53]. This is similar to NASA’s approach, which will be discussed in Section 3.5.1, which requires a comprehensive volumetric and surface NDE for all metal L-PBF parts regardless of the consequence of failure or structural demand [17]. In NASA’s case, only for parts with a low consequence of failure that could fit into a “do no harm” category designated as Class C would NDE be waived. This NASA “do-no-harm” Class C part category or, similarly, the Japanese Space Exploration Agency (JAXA) nonflight, nonstructural Class C and D part categories, may be intrinsically similar to LMCO’s nonstructural, noncritical, or prototype part categories where only a basic quality of workmanship and/or visual inspection is performed [53]. Similar to industry peers such as GE, LMCO relies on internal and industry standards along with certifications to SAE AS9100 [33] and NADCAP [34] to determine compliance of suppliers’ quality systems, controls, procedures, and facilities for production of AM parts relative to industry norms. Within LMCO there are various checklists covering applicable part category classes similar to the checklists developed by NADCAP for metal L-PBF and EBPBF parts used in critical structural applications [53]. However, since LMCO is using a wider variety of AM processes for a diverse range of applications, a need arose to develop additional checklists suitable for its manufacturing operations.
3.5
Government agency approaches
3.5.1 National aeronautics and space administration qualification and certification approach While additive manufacturing offers the ability to manufacture complex part designs rapidly at a reduced cost, the extreme pace of implementation introduces risks to the safe adoption of this developing technology. The development of aerospace quality standards and specifications is required to balance the benefits of additive manufacturing technologies properly with the inherent risks. Many companies have developed or are developing in-house standards and specifications for additive manufacturing. However, as a certifying authority, NASA’s design and construction standards do not yet include specific requirements for controlling the unique aspects of the additive manufacturing process and the resulting hardware. A significant national effort is now focused on developing standards for additive manufacturing [30]. However, the content and scheduled release of many of the consensus standards needed to ensure proper Q&C of AM hardware do not support the near-term programmatic needs of NASA.
52
Additive Manufacturing for the Aerospace Industry
NASA and its commercial partners in manned spaceflight (i.e., the Commercial Crew, Space Launch System, and Orion Multi-Purpose Crew Vehicle programs) are actively developing AM parts for flight as early as 2018. To bridge this gap, NASA MSFC personnel authored a Center-level standard (MSFC-STD-3716 [17]) to establish standard practices for the L-PBF process. The MSFC standard has been used as a basis for L-PBF process implementation for each of the manned spaceflight programs. This standard and its companion specification [18] will provide a consistent framework for the development, production, and evaluation of additively manufactured parts for spaceflight applications. Based on the principles of MSFC-STD-3716 [17], the development of agencylevel standards to meet NASA AM-related program needs is underway. A team with representatives from nine NASA centers with consultants from other government agencies has been formed. The goal of this team is to develop standards that will apply to AM processes being used by NASA and be readily adaptable to all NASA centers, programs, and projects. Three standards are currently under development for manned spaceflight, noncrewed spaceflight, and aeronautics. As part of this effort, several additional specifications may be required to address raw materials, parts procurement, and processes to supplement these three standards if the determination is made that no existing VCS standards exist or if existing VSC standards do not serve the public interest or are incompatible with NASA’s missions, authorities, priorities, and budget resources [54]. The standards under development will introduce requirements with guidance that can then be used to develop manufacturing plans and provide product specifications for both general and specific applications. The standards will not specifically dictate how to manufacture or certify a component, but the requirements will identify factors that must be addressed for all phases of design, manufacture, and qualification. The NASA standards under development will be applicable to mature technologies. Specific technologies will be discussed in the documents, but to allow for expansion, the documents will not be limited to only these technologies. The standards will concentrate specifically on metals (powder fed PBF, and wire and powder fed Directed Energy Deposition [DED]) and polymers (wire fed FDM/DED). Materials determined to be out of scope include ceramics, composites, regolith, and printed circuits.
3.5.1.1 General qualification requirements The MSFC Technical Standard [17] and Specification [18] provides the framework, and establishes general requirements, for the development and production of spaceflight hardware produced using the L-PBF AM processes. While structural design criteria are not dictated, these documents accommodate requirements from NASA’s governing design and safety standards [5558], thus, providing the necessary controls to ensure safe implementation of AM technology. Fig. 3.8 illustrates the key products and processes controlled by MSFC-STD-3716 and, figuratively, how each product or process is related. The general requirements of an Additive Manufacturing Control Plan (AMCP) that governs foundational process controls
Qualification and certification of metal
53
General requirements
AMCP
QMS
MSFC-SPEC-3717 =Requirements levied by MSFC-STD-3716
Foundational process controls
Feedstock specification
Definition of metallurgical process
Fusion process Thermal process Machine 1 Master QMP/R
Machine 2
Qualification maintenance calibration Training
Machine 3
Machine 4
Machine “n”
QMP/R
QMP/R
QMP/R
PCRD data MPS data
Part production controls
Training plan
Qual. of met. process
QMP/R
Part
ECP
SPC criteria
Design properties
Design process
Classify part
Pre-prod article plan
PPP
Pre-prod article evaluation
Pre-prod article report
AMRR MRB QPP
Production engineering controls
Production
Witness SPC, NDE, acceptance tests Service
Controlling document, requiring NASA approval. Controlling document(s), not requiring NASA approval, but available for review.
Active database, not requiring NASA approval, but available for review.
Action or process. Decisional action or process, with result available for review. Part
Representation of part entering process. Requirements with procedural details contained in MSFC-SPEC-3717.
Service Representation of part entering service. Identifies key points of QMS involvement. Identifies PBF requirements levied by MSFC-STD-3716 with procedures in MSFC-SPEC-3717 Negative outcome of decisional action
Figure 3.8 Key products and processes in MSFC-STD-3716 (top), and symbol legend (bottom) [17].
54
Additive Manufacturing for the Aerospace Industry
and part production controls is also shown. These controls provide the basis for reliable part design and production. Foundational controls include a Qualified Metallurgical Process (QMP) for each machine, an Equipment Control Plan (ECP), personnel training, and material property development via a Material Property Suite (MPS). The MPS is a collection of L-PBF material property data specific to a material and process that includes all test data, design values, and criteria needed to implement and maintain Statistical Process Control (SPC) for the L-PBF process. Part production controls are typical of aerospace operations and include part categories, a Part Production Plan (PPP), a Qualified Production Plan, and other miscellaneous production controls. The symbols in Fig. 3.8 (bottom) indicate the type of product or action, such as internal documents, documents requiring approval, databases, or decisional actions. Fig. 3.8 further illustrates the flow of the products and processes through the general, foundational, and part production controls. While showing the figurative relationships of the key products and processes, Fig. 3.8 cannot be read as a serial flow chart, particularly in the prerequisite foundational controls. The AMCP also defines how active QMS is integrated throughout the process. Key points of QMS integration are illustrated with a green triangle symbol in Fig. 3.8. The AMCP and the QMS govern the engineering and quality assurance disciplines, respectively, from start to finish. An essential element for all AM parts manufactured for the aerospace industry is the creation of an AMCP. The three NASA standards will be based on the principles put forth in MSFC-STD-3716. This can only be accomplished through tailoring; therefore, an AMCP is needed to document the requirements that are based on the appropriate NASA standard. The AMCP will also include the means by which subcontractor and vendor engineering compliance will be managed. In addition to an AMCP, a QMS that conforms to SAE AS9100 [33] or an approved equivalent is also required. The QMS will ensure that QA controls are properly implemented and noncompliance is properly managed.
3.5.1.2 Additive manufacturing part categories To allow requirements to be tailored for a particular NASA application, a classification system has been established to define and communicate the risk associated with a given AM part and to levy appropriate levels of process control, qualification, and inspection. The current NASA MFSC classification system [17], shown in Fig. 3.9, is based on three key decisions: consequence of failure, structural demand, and AM risk. Additive manufacturing risk is calculated based on five ratings criteria: (1) Can all surfaces and volumes be reliably inspected, or does the design permit adequate proof testing based on stress state? (2) Can as-built surface be fully removed on all fatigue-critical surfaces? (3) Are surfaces interfacing with sacrificial supports fully accessible and improved? (4) Are structural walls or protrusions $ 1 mm in cross-section? (5) Critical regions of the AM part do not require sacrificial supports. This decision tree leads to eight distinct classifications. To aid in tailoring the three proposed NASA standards for manned spaceflight, noncrewed spaceflight, and aeronautics a slightly different approach was developed,
Qualification and certification of metal
55
Figure 3.9 MSFC-STD-3716 additive manufactured part classification system [17].
Figure 3.10 Proposed classification system for new NASA standards.
as shown in Fig. 3.10. The new classification system has three levels of primary classification (A, B, and C) and allows for a secondary classification for certain cases for Class A and B parts. The primary classification drives the tailoring of requirements for each part and the secondary classification acts as a communication tool to allow for effective risk management when necessary.
56
Additive Manufacturing for the Aerospace Industry
The first decision gate is the same as given in [17], but the criteria are modified to address all NASA applications. Initially, a part will be designated as Class A (high consequence of failure) if one or more of the following criteria are met: G
G
G
Fracture-critical as per NASA-STD-5019A [22]. Failure would lead to a catastrophic hazard (i.e., loss of life, disabling injury, or loss of a major national asset). Failure would lead to the loss of one or more primary/minimum mission objectives.
Note that in the event of part redundancy, Class A may still be applicable if the project decides that the risk of a common mode failure is credible. Unlike the classification scheme used in [17], the new classification shown in Fig. 3.9 will allow parts with a low consequence for failure that could fit into a “do no harm” category to be designated as Class C. Parts that do not meet this criterion will be assigned as Class B. The exact definition of which attributes would allow for a Class C designation is under development. QA activities requiring inspection (NDE) are only required for parts used in spacecraft and in applications where there is a structural demand. For spaceflight hardware subjected to structural loads, but for which NDE cannot be used or is difficult to perform, a new category A0 is introduced. For spaceflight hardware not subjected to structural loads, a category C1 is introduced analogous to the proposed NASA “do no harm” Class C designation. For category C1 parts, no NDE is currently required by JAXA. Lastly, for hardware not used in spaceflight application, a category D1 is introduced. These parts are only used during the development phase or in ground applications. Like category C1 parts, no NDE is currently required by JAXA for category D1 parts. Analogously, no NDE is required by LMCO for Class II nonstructural parts and Class I noncritical, prototype, or model parts [53]. Considering the overall similarities and differences between the NASA, JAXA, and LMCO part classification systems, perhaps the greatest strength of the NASA system is the premium placed on assigning and communicating the risk associated with a given AM part (safety constraint). Similarly, perhaps the greatest strength of the JAXA system is the premium placed on part design, complexity, and application as primary drivers of the type and requisite accuracy of any post-process NDE performed (NDE capability constraint).
3.5.1.3 Integrated structural integrity rationale The largest latent risk associated with AM parts used in critical spaceflight applications lies in the limitations to verify individual part integrity (see Appendix B in [17]). Currently, process control methods implemented during build are not sufficiently mature or qualified to verify part integrity independently. In these cases, structural integrity must be verified through inspection (NDE) or test (proof and functional acceptance testing) performed in finished parts (or witness coupons) after build. For parts classified with high AM risk, a combination of inspection and tests may be needed to achieve full coverage of the part, thus, ensuring its structural integrity. However, given the extreme diversity of AM parts slated for NASA
Qualification and certification of metal
57
applications, it is not practical to levy specific QA requirements stipulating the degree of inspection or tests needed for a given part. Despite the vagaries associated with the manner in which QA activities such as process control methods, and post-process inspection and tests are imposed, some simplifying rules do exist. For example, AM parts in NASA Classes A1 through to B2, as shown in Fig. 3.9, must be subjected to a qualification test program that demonstrates part performance and functionality meet the design mission requirements, life factors, and lifecycle capability [17]. This program also includes comprehensive surface and volumetric NDE within the limitations of the NDE method used and part complexity. Stated differently, the rigor of the inspection and testbased QA activities used shall be commensurate with the part’s classification. For example, Class A fracture-critical parts typically require a quantitative rationale involving, in the case of NDE, the detection of flaws with a known probability of detection as stipulated in NASA-STD-5009 [22]. The rationale also identifies areas or volumes of the part relying solely on process controls, that is, not verifiable by post-build NDE as risk areas for further consideration.
3.5.1.4 Influence of mission classification For NASA science missions and payloads, a risk-based mission classification is assigned as per NASA Procedural Requirement (NPR) 8705.4 [59]. Fig. 3.11
Figure 3.11 Mission classes. Source: Appendix Bin NASA Procedural Requirement, Risk classification for NASA payloads, in: NPR 8705.4, 2012.
58
Additive Manufacturing for the Aerospace Industry
provides a description of the mission classes. To capture all the missions that would be covered by the three NASA standards, a total of six mission classes should be considered: 1. 2. 3. 4. 5. 6.
Manned Spaceflight Class A (per NPR 8705.0004) Class B (per NPR 8705.0004) Class C (per NPR 8705.0004) Class D (per NPR 8705.0004) Associated ground support equipment and test hardware
The NASA team considered three possible approaches to how part classification and mission classification could interact. These three cases are: 1. Part class determines the requirement set independent of mission class (similar to the JAXA approach). 2. Mission class influences part class through consequence of failure or other criteria. 3. Part class and mission class requirements are combined into a common risk matrix.
The team consensus and recommended approach was that the part classification and the mission classification should be considered independently. This decision led to the recommendation to develop three separate NASA standards.
3.5.1.5 Tailoring approach For each NASA standard, the requirements summary given in Appendix F, Table VIII, in [17] will be used as the basis for tailoring. Table VIII contains Additive Manufacturing Requirements (AMRs). For each NASA standard, a unique requirements matrix will be developed that modifies each AMR to make it applicable for its application. Fig. 3.12 shows an example of a requirements matrix. The matrix will designate each requirement based on part classification to be used Requirement
AMR-3
Requirement description Additive Manufacturing Control Plan
MSFC-STD-3716
NASA-STD (proposed language)
[AMR-3] The CEO responsible for the design and manufacture of L-PBF hardware shall provide an AMCP that accomplishes each of the following: a.Documents the implementation of each of the requirements of this MSFC Technical Standard. b. Documents and provides rationale for any tailoring of the requirements of this MSFC Technical Standard. c. Documents the methods used to control compliance with these requirements by subcontractors and vendors.
[AMR-3] The CEO responsible for the design and manufacture of AM hardware shall provide an AMCP that accomplishes each of the following: a. Documents the implementation of each of the requirements of this NASA Technical Standard. b. Documents and provides rationale for any tailoring of the requirements of this NASA Technical Standard.
d. Provides for complete governance for the implementation of L-PBF such that, once approved by the procuring authority, the AMCP becomes the document used for verification of L-PBF requirements.
Figure 3.12 Requirements matrix example.
B
C
Notes
c. Documents the methods used to control AW T compliance with these requirements by subcontractors and vendors.
O
See Appendix for Tailoring Guildlines
d. Provides for complete governance for the implementation of AM such that, once approved by the procuring authority, the AMCP becomes the document used for verification of AM requirements.
A
Qualification and certification of metal
59
as-written, optional, or tailorable. Tailoring guidelines will be written and provided in either the body of the specification or an appendix.
3.5.1.6 Industry standards The NASA standards will be written to the utilized industry-developed AM standard when appropriate. Because industry standards are being developed and are rapidly changing, a separate NASA specification will be created to list these standards and provide revision-level configuration controls. This will allow the NASA AM community to make adjustments without revising the actual standards.
3.5.1.7 Process specifications The use of process specifications will be defined in the requirements matrix. In cases where no industry standard exists or the standard is considerably substandard, NASA may decide to author a unique specification. For areas covered by MSFCSPEC-3717 [18]; namely, the procedural requirements for foundational process control in L-PBF (Fig. 3.8), including, for example, virgin powder and reuse requirements, process parameters, restart procedures, post-processing, QMPs, equipment calibration and maintenance, facilities qualification, and personnel training, tailoring guidance will be provided either in the body of the text or in an appendix.
3.5.1.8 Procurement specifications Guidance will be provided on how to write a procurement specification. Certain procurement specifications may be written to make it as easy as possible to buy a “good” part from a “proven” manufacturer. These may not be appropriate for Class A1 parts on Class A missions. Industry standards will be leveraged as much as possible. These procurements will focus primarily on raw material requirements, vendor quality/process controls, historical material property trends, and limited part-specific requirements. When appropriate, procurement specifications will intentionally be written in a nonspecific manner to allow a vendor to control proprietary processes.
3.5.1.9 Additional guidance In addition to tailoring and procurement specification guidelines, additional appendices may be required to cover topics such as: G
G
G
Guidance in writing an AMCP. Guidance in writing a PPP. Guidance on what should be in a process qualification or feedstock specification.
3.5.1.10 Warnings There is growing concern in the NASA Durability and Damage Tolerance (D&DT) community [59] that technology gaps exist that may lead to the use of D&DT tools
60
Additive Manufacturing for the Aerospace Industry
beyond their capability. These shortfalls could be accelerated with the use of AM parts where complex designs and unknown local material properties exist. Several research studies and development or testing programs have been proposed or are already underway to close these gaps.
3.5.2 Federal Aviation Administration qualification and certification approach Q&C requirements for commercial aircraft parts have been traditionally linked to the level of part criticality, defined with various degrees of specificity [3,21]. For instance, the FAA rule for materials (14 CFR 25.603) defines its applicability as “parts, the failure of which could adversely affect safety” [47]. The FAA Advisory Circular 25.571-1D defines principal structural elements (PSEs) as elements “whose integrity is essential in maintaining the overall structural integrity of the airplane,” including “structures susceptible to fatigue cracking, which could contribute to a catastrophic failure.” [60] The FAA rule 14 CFR 37.70 for Engine Life-Limited Parts (LLPs) [61], defines LLPs as parts “whose primary failure is likely to result in a hazardous engine effect.” Throughout the FAA rules and guidance materials for various product types, a common denominator emerges in that an appropriate damage tolerance assessment needs to be performed on parts of high criticality such as PSEs or LLPs. Two elements of such assessment are crack growth analysis using fatigue testing and inspections using NDE. These two elements are also evident in NASA’s [17,18] quality processes and documents. Due to the random nature of material anomalies (not specific to AM materials), the FAA Advisory Circular 33.70-1 defining damage tolerance requirements for engine LLPs states that “the probabilistic approach to damage tolerance assessment is one of two elements necessary to appropriately assess damage tolerance” [62]. To support such an assessment, the appropriate characterization of material anomalies is needed, in addition to conventional fatigue and fracture properties of substrate materials. Such characterization should focus on developing the size distribution and frequency of occurrence of material anomalies. As discussed in [21], this information can be used to define an exceedance curve for a given class of material defects which is the key input into probabilistic fracture mechanics-based assessment, such as the one defined in the FAA Advisory Circulars 33.14-1 [63] and 33.70-2 [64] for specific types of material or manufacturing defects. The FAA Part 25 rule for “Fabrication Methods” (14 CFR 25.605) [47] states that “the process must be performed under an approved process specification” and that “each new aircraft fabrication method must be substantiated by a test program.” However, the rule-level certification requirements often do not define the specific acceptable testing or inspection procedures. This level of detail has been left open for the OEM to define as a part of the means to determine compliance with the rules (i.e., means of compliance or MoC), as reviewed and approved by
Qualification and certification of metal
61
certification authority. Developing effective MoC for new technologies such as AM is especially challenging. For this reason, standardization of the methods for testing and inspections can be viewed as enablers for efficient and robust (i.e., consistent) Q&C processes across the aviation industry. For AM, being a relatively new technology, the specific testing and inspection procedures are still under development or are proprietary, and must reflect the unique materials, processing constraints, and design attributes of metal AM parts such as anisotropy, inherent material and microstructural anomalies, residual stresses, and locationspecific properties, etc. [3,21]. Due to the broad range of potential AM applications in civil aviation that include design and production of new parts, replacement parts, and repairs, the FAA has recently issued several documents. These include an internal memorandum providing guidance to the regional aircraft certification offices and Manufacturing Inspection District Offices regarding the engineering and manufacturing considerations for certification of AM parts, as well as the Notice for the Flight Standards District Offices inspectors [65], to provide an introduction and awareness regarding the use of AM technology in maintenance, alterations, and repairs of aircraft and engine components. The latter notice [65] also cites the “lack of industrywide standards for AM” as one of the current challenges, certain aspects of which were summarized in Section 3.3.1.
3.6
Summary and recommendations
State-of-the-discipline Q&C approaches used by industry and government to control AM materials, processes, and parts have been summarized in this chapter. As noted, the slow pace of adopting much-needed voluntary consensus Q&C standards is forcing companies and government agencies with procuring authority for AM parts to establish internal Q&C requirement documentation suited to their needs, programs, and resources. Aspects of current best Q&C practices adopted by the industry (GE and LMCO) and governments (NASA, JAXA, and the FAA) have been summarized. Similarities among these best practices have emerged, such as the need to establish AM part categories to define and communicate the risk associated with a given AM part and the need to adopt specific testing and inspection procedures suited to the procuring authority’s requirements and OEM’s capabilities. In addition to levying Q&C requirements for testing and inspection, requirements must be imposed to control the design, input materials, fabrication processes, equipment, personnel training, and post-processing. All of these requirements issued by the various procuring authorities are likely to evolve as more information and field experience become available in the near future. To facilitate the adoption of future Q&C standards and to further the advancement of Q&C protocols tailored to AM, it is proposed that companies and agencies
62
Additive Manufacturing for the Aerospace Industry
adopt a model of multi-organizational collaboration with peers toward addressing some of the more complex and labor-intensive Q&C challenges. Possible areas of collaboration include: G
G
G
G
G
G
Establishment of consensus understanding of the terms qualification, certification, and quality control and their differences and interrelationships. Develop and adopt Q&C-related industry standards summarized in Table 3.1, especially those deemed to be high priority. Produce and test new NDE capability standards with intentional features and defects to mature current NDE detection capabilities for technologically important AM defects, including improved metrological capability for intricate internal features. Produce and test sacrificial defect standards with known loadings of specific defect types, sizes, and distributions to determine effect-of-defect and facilitate determination of quantitative accept/reject criteria. Advance current in-process monitoring state-of-the art, focusing on commercially available monitoring and sensing technologies. Increase industry/government collaboration to advance process analytics, modeling, and simulation tools.
Acknowledgments The authors wish to acknowledge Jim McCabe and Sarah Bloomquist of ANSI and members of the ANSI-America Makes AMSC Qualification and Certification Working Group for their efforts to identify and prioritize Q&C standardization gaps. The authors also wish to acknowledge Deborah Whitis of GE Additive for her contributions to the GE section.
References [1] G.E. Aviation, GE Aviation Announces First Run of the Advanced Turboprop Engine, 2017. [2] R.G. Clinton, NASA’s in space manufacturing initiative and additive manufacturing development and quality standards approach for rocket engine space flight hardware, in: Additive Manufacturing for Defense and Aerospace 2016 Summit, London, UK, March 2930, 2016. ,https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20160004218.pdf. [3] M. Seifi, M. Gorelik, J.M. Waller, N. Hrabe, N. Shamsaei, S. Daniewicz, et al., Metal additive manufacturing standardization in support of qualification/certification, J. Mater. Miner. Met. Mater. Soc. (JOM) 69 (2017) 439. [4] T. Kellner, The FAA cleared the first 3D printed part to fly in a commercial jet engine from GE, in: GE Reports, ,https://www.ge.com/reports/post/116402870270/the-faacleared-the-first-3d-printed-part-to-fly-2/., 2015. [5] T. Kellner, Mind meld: how GE and a 3D-printing visionary joined forces, in: GE Reports, ,https://www.ge.com/reports/mind-meld-ge-3d-printing-visionary-joined-forces/., 2017. [6] S. Draper, I. Locci, B. Lerch, D. Ellis, P. Senick, M. Meyer, et al., 66th International Astronautical Congress, 2015, pp. 19.
Qualification and certification of metal
63
[7] P.R. Gradl, S.E. Greene, C. Protz, J. Buzzell, C. Garcia, J. Wood, et al., Additive manufacturing of liquid rocket engine combustion devices: a summary of process developments and hot-fire testing results, in: 2018 Joint Propulsion Conference, AIAA Propulsion and Energy Forum (AIAA 2018-4625). 2018. ,https://arc.aiaa.org/doi/abs/ 10.2514/6.2018-4625.. [8] W.E. Frazier, Metal additive manufacturing: a review, J. Matls, Eng. Perform. 23 (6) (2014) 19171928. Available from: https://link.springer.com/article/10.1007% 2Fs11665-014-0958-z. [9] Penn State CIMP-3D and Nexight Group, Strategic roadmap for the next generation of additive manufacturing materials, in: Consortium for Additive Manufacturing Materials (CAMM), 2015. [10] M. Seifi, A. Salem, J. Beuth, O. Harrysson, J.J. Lewandowski, Overview of materials qualification needs for metal additive manufacturing, J. Mater. Miner. Met. Mater. Soc. (JOM) 68 (2016) 747. [11] M. Seifi, M. Dahar, R. Aman, O. Harrysson, J. Beuth, J.J. Lewandowski, Evaluation of orientation dependence of fracture toughness and fatigue crack propagation behavior of as-deposited arcam EBM Ti 2 6Al 2 4V, J. Mater. Miner. Met. Mater. Soc. (JOM) 67 (2015) 597. [12] M. Seifi, A. Salem, D. Satko, J. Shaffe, J.J. Lewandowski, Defect distribution and microstructure heterogeneity effects on fracture resistance and fatigue behavior of EBM Ti6Al4V, Intl. J. Fatigue 94 (2017) 263287. Available from: https://www. sciencedirect.com/science/article/pii/S0142112316301451. [13] M. Seifi, A.A. Salem, D.P. Satko, U. Ackelid, S.L. Semiatin, J.J. Lewandowski, Effects of HIP on microstructural heterogeneity, defect distribution and mechanical properties of additively manufactured EBM Ti48Al2Cr2Nb, J. Alloys Compd. 729 (2017) 11181135. Available from: https://www.sciencedirect.com/science/article/pii/S0925838817332127. [14] J. Dzugan, M. Seifi, R. Prochazka, M. Rund, P. Podany, P. Konopik, et al., Effects of thickness and orientation on the small scale fracture behaviour of additively manufactured Ti6Al4V, Matls. Charact (April 4, 2018). Available from: https://www.sciencedirect.com/science/article/pii/S1044580317328802. [15] National Institute of Standards and Technology, Measurement Science Roadmap for Metal-Based Additive Manufacturing. Prepared by Energetics Incorporated, Columbia, MD, for NIST, U.S. Department of Commerce, 2013. ,https://www.nist.gov/sites/ default/files/documents/el/isd/NISTAdd_Mfg_Report_FINAL-2.pdf. [16] J.M. Waller, B.H. Parker, K.L. Hodges, E.R. Burke, J.L. Walker, E.R. Generazio, Nondestructive evaluation of additive manufacturing state-of-the-discipline report, in: NASA/TM—2014218560, 2014. [17] Douglas Wells, Standard for Additively Manufactured Spaceflight Hardware by Laser Powder Bed Fusion in Metals, MSFC-STD-3716, NASA Marshall Spaceflight Center, Huntsville, AL, 2017. [18] Douglas Wells, Specification for Control and Qualification of Laser Power Bed Fusion Metallurgical Processes, MSFC-SPEC-3717, NASA Marshall Spaceflight Center, Huntsville, AL, 2017. [19] ASTM WK47031, Standard Guide for Nondestructive Testing of Metal Aerospace Additive Manufactured Parts After Build, ASTM International, West Conshohocken, PA, in Committee E07 concurrent ballot, 2018. [20] J. Allison, B. Cowles, J. DeLoach, T. Pollock, G. Spanos, “Integrated Computational Materials Engineering (ICME): Implementing ICME in the Aerospace, Automotive, and Maritime Industries,”, The Minerals, Metals & Materials Society, 2013.
64
Additive Manufacturing for the Aerospace Industry
[21] M. Gorelik, Additive manufacturing in the context of structural integrity, Int. J. Fatigue 94 (2017) 168177. [22] NASA-STD-5009, Nondestructive Evaluation Requirements for Fracture-Critical Metallic Components, available from the NASA Technical Standards System at the NASA website www.standards.nasa.gov, 2018. [23] J.J. Lewandowski, M. Seifi, Metal additive manufacturing: a review of mechanical processes, Ann. Rev. Mater. Res. 46 (2016) 151. [24] H. Gong, K. Rafi, H. Gu, G.D. Janaki Ram, T. Starr, B. Stucker, Influence of inherent surface and internal defects on mechanical properties of additively manufactured Ti6Al4V alloy: comparison between selective laser melting and electron beam melting, Mater. Des 86 (2015) 545554. [25] P. Li, D.H. Warner, A. Fatemi, N. Phan, Critical assessment of the fatigue performance of additively manufactured Ti6Al4V and perspective for future research, Int. J. Fatigue 85 (2015) 130143. [26] N. Hrabe, T. Gnaupel-Herold, T. Quinn, Fatigue properties of a titanium alloy (Ti6Al4V) fabricated via electron beam melting (EBM): effects of internal defects and residual stress, Int. J. Fatigue 94 (2016) 202210. [27] G. Nicoletto, Anisotropic high cycle fatigue behavior of Ti6Al4V obtained by powder bed laser fusion, Int. J. Fatigue 94 (2016) 255262. [28] D. Greitemeier, F. Palm, F. Syassen, T. Melz, Fatigue performance of additive manufactured TiAl6V4 using electron and laser beam melting, Int. J. Fatigue 94 (2016) 211217. [29] S. Beretta, S. Romano, A comparison of fatigue strength sensitivity to defects for materials manufactured by AM or traditional processes, Int. J. Fatigue 94 (2016) 178191. [30] Additive Manufacturing Standardization Collaborative (AMSC), Standardization roadmap for additive manufacturing, in: Version 2.0, ANSI and NDCMM/America Makes, registration required, ,https://www.americamakes.us/america-makes-ansi-publish-version-2-0-standardization-roadmap-additive-manufacturing/., 2018 [31] National Center for Manufacturing Sciences, Technology Exchange on Coordination of US Standards Development for Additive Mfg, ,https://www.ncms.org/technologyexchange-on-coordination-of-us-standards-development-for-additive-mfg/., 2015. [32] ISO/ASTM 52900, Additive Manufacturing—General Principles—Terminology, ASTM International, West Conshohocken, PA, 2015. [33] SAE International, SAE AS9100D, Quality Management Systems—Requirements for Aviation, Space, and Defense Organizations, revised 2016-09-20. [34] Performance Review Institute, About Nadcap, ,https://p-r-i.org/nadcap/about-nadcap/. (accessed 15.10.2018). [35] Performance Review Institute, AC7110/14, Nadcap Audit Criteria for Laser and Electron Beam Metallic Powder Bed Additive Manufacturing, Nadcap Aerospace Standard, available for downloading at no charge to any person registered in eAuditNet, www.eauditnet.com, 2017. [36] Performance Review Institute, AC7110, Audit Criteria for Welding/Torch and Induction Brazing and Additive Manufacturing, Nadcap Aerospace Standard, available for downloading at no charge to any person registered in eAuditNet, www.eauditnet.com, 2017. [37] Batelle Memorial Institute, Metallic Materials Properties Development and Standardization (MMPDS-12), 12th Edition, available at www.mmpds.org, 2017
Qualification and certification of metal
65
[38] AMS 4999, SAE International, Titanium Alloy Direct Deposited Products 6Al 2 4V Annealed, SAE International, reaffirmed 2016. [39] U.S. Department of Transportation Federal Aviation Administration, Advisory Circular Public Comment Phase, Guidance Material for Turbine Engine Parts and Repairs Produced by Powder Bed Fusion Additive Manufacturing Process, in: AC draft No. 33.15-4, 2018. [40] Kelkar, et al., Alloy 718: laser powder bed additive manufacturing for turbine applications, in: 2018 Proceedings of the Ninth International Symposium on Superalloy 718 & Derivatives: Energy, Aerospace, and Industrial Applications, Ott, E.A, et al., Eds., TMS, 2018, pp. 5368. [41] Metal Additive Manufacturing Standardization in Support of Qualification/ Certification, GE internal communication, 2018. [42] NASA Procedural Requirement, NASA systems engineering processes and requirements, in: Appendix E, Technology Readiness Levels, NPR 7123.1B, 2013. [43] Manufacturing Readiness Level (MRL) Deskbook, Version 2017, Department of Defense, Manufacturing Technology Program, ,http://www.dodmrl.com/MRL_Deskbook_2017. pdf., 2017. [44] ASTM News Release, 15 proposed Additive-Manufacturing Standards to support Key Industry Accreditation, ASTM International News Release, ,https://www.astm.org/ newsroom/15-proposed-additive-manufacturing-standards-support-key-industry-accreditation., 2017. [45] FAA, Code of Federal Regulations Part 14, Airworthiness Standards: Aircraft Engines, Part 33, 2018. [46] FAA, Code of Federal Regulations Part 14, Airworthiness Standards: Normal, Utility, Acrobatic, and Commuter Category Airplanes, Part 23, 2018. [47] FAA, Code of Federal Regulations Part 14, Airworthiness Standards: Transport Category Airplanes,” Part 25, 2018. [48] American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code, Section II—Materials, 2017. [49] U.S. Food and Drug Administration, Federal Food, Drug, and Cosmetic Act, Title 21, Chapter 9 of the United States Code, Subchapter V—Drugs and Devices, 2018. [50] D. Whitis, Additive manufacturing material implementation at GE additive, GE Additive Smart Manufacturing Seminar Series, The Changing Landscape of Additive Manufacturing Materials, ,http://smartmanufacturingseries.com/wp-content/uploads/ 2017/08/Whitis.pdf, http://smartmanufacturingseries.com/addapril/., 2017. [51] Lockheed Martin, Giant Satellite Fuel Tank Sets New Record for 3-D Printed Space Parts, ,https://news.lockheedmartin.com/2018-07-11-Giant-Satellite-Fuel-Tank-SetsNew-Record-for-3-D-Printed-Space-Parts#assets_all., 2018 [52] J. Kerr, B. Fenolia, “Imagine The Future: Additive Manufacturing in Aerospace,” presentation, Lockheed Martin Space Systems, Denver, CO, 2015. [53] Additive Manufacturing Standardization Collaborative (AMSC), Lockheed Martin AM supplier quality checklist overview, in Standardization Roadmap for Additive Manufacturing, Version 1.0, Section 2.3.2.2, ANSI and NDCMM/America Makes, ,https://share.ansi.org/ Shared Documents/StandardsActivities/AMSC/AMSC_Roadmap_February_2017.pdf., 2017 [54] Office of Management and Budget, OMB Circular A119; Federal participation in the development and use of voluntary consensus standards and in conformity assessment activities. , Federal Register, February 19. 63 (33) (1998) 85458558.
66
Additive Manufacturing for the Aerospace Industry
[55] NASA-STD-5001, Structural Design and Test Factors of Safety for Spaceflight Hardware, available from the NASA Technical Standards System at the NASA website www.standards.nasa.gov, 2016. [56] NASA-STD-5017, Design and Development Requirements for Mechanisms, available from the NASA Technical Standards System at the NASA website www.standards. nasa.gov, 2015. [57] NASA-STD-5019A, Fracture Control Requirements for Spaceflight Hardware, available from the NASA Technical Standards System at the NASA website www.standards.nasa.gov, 2016 [58] NASA-STD-6016, Standard Materials and Processes Requirements for Spacecraft, available from the NASA Technical Standards System at the NASA website www.standards.nasa.gov, 2016. [59] NASA Procedural Requirement, Risk classification for NASA payloads, in: NPR 8705.4, 2012. [60] FAA, Damage Tolerance and Fatigue Evaluation of Structure, Advisory Circular 25.571-1D, January 13, 2011, https://www.faa.gov/regulations_policies/advisory_circulars/index.cfm/go/document.information/documentID/865446. [61] FAA, Code of Federal Regulations Part 14, Airworthiness Standards: Engine LifeLimited Parts,” Part 37.70, January 1, 2018. [62] FAA, Guidance Material for Aircraft Engine Life-Limited Parts Requirements, Advisory Circular 33.70-1, February 24, 2017, https://www.faa.gov/documentLibrary/ media/Advisory_Circular/AC_33_70-1_Chg_1.pdf. [63] FAA, Damage Tolerance for High Energy Turbine Engine Rotors - Including Change 1, Advisory Circular AC 33.14-1, January 8, 2001, https://www.faa.gov/regulations_policies/advisory_circulars/index.cfm/go/document.information/documentID/22920. [64] FAA, Damage Tolerance of Hole Features in High-Energy Turbine Engine Rotors, Advisory Circular 33.70-2, August 28, 2009, https://www.faa.gov/documentLibrary/ media/Advisory_Circular/AC.33.70-2.pdf. [65] U.S. Department of Transportation-Federal Aviation Administration Notice N 8900. 391, Additive Manufacturing in Maintenance, Preventive Maintenance, and Alteration of Aircraft, Aircraft Engines, Propellers, and Appliances, Washington, DC, 2016.
Design for metal additive manufacturing for aerospace applications
4
Manish Kamal and Gregory Rizza Arconic Inc., Arconic Fastening Systems, Carson, CA, United States
4.1
Introduction
Additive manufacturing (AM), is a manufacturing process the use of which has been largely driven by its ability to create relatively unhindered design customization while containing fewer process constraints when compared to traditional manufacturing processes. In traditional design for manufacturing, most steps to part design assume for simplicity an isotropic material, and the production of part definitions using any variety of conventional methods, such as forging, casting, or subtractive machining. AM as a process separates itself from these traditional methods in that it adds an increased amount of design flexibility when choosing the shape and geometry of a desired part. As well, the additive capabilities of the process, involving the fusing of material layer-by-layer, allow for the application of AM features to a wrought substrate with a differing microstructure. This permits the creation of custom functionally graded materials through AM, and introduces the possibility of materials with tailored microstructures designed for specified applications and performance. When describing design for AM (DFAM), the authors apply a holistic approach to the design process, where the method in developing a part that utilizes AM comprises equally the process of designing the part, as well as the method of its fabrication. In this view, this chapter will focus on both design and process considerations for AM, and their interrelationship for producing metal components for aerospace applications. The topics discussed in detail are: G
G
G
G
Methods and approaches to AM design, identifying the different design techniques currently in use for AM part design, Process aspects of AM design, illustrating the relevance of material properties, part performance, and post processing operations (part evaluation and inspection will also be discussed in some detail), Current design tools available for AM, Economic considerations when evaluating AM as a production process.
Additive Manufacturing for the Aerospace Industry. DOI: https://doi.org/10.1016/B978-0-12-814062-8.00005-4 © 2019 Elsevier Inc. All rights reserved.
68
Additive Manufacturing for the Aerospace Industry
The objective of this chapter is to give the reader a thorough summary of the key considerations when deciding whether to manufacture an aerospace product using AM, and the importance of the interrelationship between part and process. When selecting AM as a manufacturing method, there are several types of AM processes that can be used to produce metallic parts. Of these processes, the two most common AM fabrication systems currently in use are the laser powder bed systems and directed energy deposition systems [1,2]. Both methods are currently used today to produce AM parts in aerospace, with the percentage of AM parts on both aircraft and spacecraft increasing each year [3]. To limit the scope of this chapter, these two methods for producing AM parts will be the only methods touched upon in detail.
4.2
Methods and approaches
There is a strong coupling that exists between process and design when evaluating AM as a method of production. The design freedoms allotted from the various processes that currently exist for producing AM parts allow for new flexibility when selecting the geometric design of a part. For this reason, it is difficult to design a part that utilizes fully the capability of AM without understanding the flexibility and limitations the process provides. The following section will describe various methods that are currently being used in the design and development of AM parts.
4.2.1 Topological optimization One of the major focal points of AM design is the capability of the process to create geometry and features that cannot be produced easily or economically using conventional manufacturing techniques. The AM process lends itself very naturally to a type of “free-form” design in which complex geometries and features can be fabricated without inflicting additional process costs outside of the standard costs associated with the metal AM process. For this reason, many of the AM parts reported in literature usually have very intricate geometries and shapes related to the operational performance of the part [14]. To reflect this, Fig. 4.1 shows an
Figure 4.1 Topologically optimized hinge for AM. (A) Initial hinge design, (B) new hinge designed using topological optimization, (C) additive manufactured hinge. AM, additive manufacturing.
Design for metal additive manufacturing for aerospace applications
69
example of how an aerospace hinge product has been redesigned to function by altering the part’s topology, and then fabricated using an AM laser powder bed process. The topology of the hinge has been modified to meet existing structural performance requirements, and optimized to reduce the overall weight of the part. Both performance and cost are key parameters when deciding if AM is appropriate for part fabrication and large-scale production of a specific part. Cost will be discussed in greater detail later in the chapter, but in terms of design, factors that directly affect the cost of producing a metal AM part are: G
G
G
Overall part volume and weight, The AM system selected for the printing process, The required secondary operations needed to transition an “as printed” part to a finalized form that meets geometric, surface and performance specifications.
All these parameters should be considered when deciding to design a part to be produced using AM. Focusing on the first design parameters mentioned, part weight and volume, a major advantage of AM is the ability of the process to allow existing components within an aircraft or spacecraft to be redesigned resulting in a new, operationally comparable AM part with some degree of weight reduction over its predecessor [5,6]. Since weight is always a critical factor for applications related to flight, weight reduction is a common practice when it comes to AM design methodologies. However, just reducing part weight through material reduction and expecting comparable performance is not a straightforward process. When redesigning an existing part or designing a brand-new part to be fabricated using AM, the utilization of some form of topological optimization is usually applied as a guide in the design process. Topological optimization is a process that focuses on the modification of the topology or surface of a structure by altering that surface using a predefined objective along with applied design constraints. Typically, with metal AM part design, the objective of a topological optimization is to reduce part weight/volume with the constraint that the part not yield or fail under the part’s prescribed operational requirements. In performing a topological optimization of a part, there are several software tools that can be used to aid in the design process (some of these tools will be briefly discussed later in the chapter). Topological optimization through simulation is usually conducted in collaboration with a finite element (FE) software package in which the optimization algorithm uses outputs from a FE model such as stress, strain, contact force, and displacement. Through an iterative process, the optimization algorithm modifies the topology of a part until the desired objective is completed within the constraints of the application [4]. Fig. 4.2 shows an optimization example of an aero-structure bracket. The optimized part in this case was 2 3 stiffer than the originally designed part under the same load conditions, and was also 15% lighter in weight. This was achieved from the bracket being redesigned to meet function rather than to meet an existing manufacturing process, while striving to reduce as much weight as possible in the part.
70
Additive Manufacturing for the Aerospace Industry
Figure 4.2 FE topological optimization of a bracket. (A) Initial bracket design, (B) bracket design after FE topological optimization, (C) additive manufactured bracket. FE, Finite element.
The numerical algorithms used in the topological optimization processes usually act upon an element’s mass or stiffness, such that through each optimization iteration, individual element stiffness or mass is altered and adjusted to achieve the desired optimization objective. To generalize the process briefly, for a structural component undergoing topological optimization, noncrucial areas outside the load path will see a reduction in element’s stiffness or mass, while elements in the load path will be unaffected, since reduction in stiffness could result in not meeting the desired objective and problem constraints. The ideal scenario for utilizing topological optimization as a design tool for AM would be using an iterative optimization process in which many design iterations are generated, eventually leading toward an optimized final design, which can be numerically validated and then fabricated using an AM process. In practice, this method is usually more complicated than this simplified description, and the process of topological optimization has several attributes that are difficult to generalize for all cases. There are several challenges that exist when applying topological optimization as a design tool for AM. Some key challenges that should be considered in the optimization process are: G
G
G
G
Topological optimization software, when it is applied in conjunction with a FE software package, is highly dependent on the boundary conditions and problem setup within the FE environment. For this reason, if a FE model is not sufficiently representative of the application of interest, the resulting topological optimized design may not be useful or adequately meet the desired application requirements. Features or geometries created from a topological optimization may not be easily fabricated using existing AM processes. Examples of such features could be thin walls, overhangs, or features that require complex support structures that cannot be easily removed using secondary operations. Features or geometries created from a topological optimization may have structural limitations due to the creation of stress concentration sites, and may be more susceptible to cyclic load failure conditions such as fatigue cracking or partial structural damage when loaded beyond the material’s yield strength [7]. An example of these types of features are internal lattice structures used as a replacement for a solid body. Features or geometries created from a topological optimization process may be prone to a greater collection of defects and material anomalies. This can be due to the specific AM build process being used, the part’s susceptibility to environmental contamination during a build, the specific fabrication material, location of the feature within the part, or some combination of the before mentioned.
Design for metal additive manufacturing for aerospace applications
71
Even though challenges do exist, and the method of topological optimization applied to AM is not yet a fully streamlined design process, it is still an effective tool when being applied to AM design. The limitations of the topological optimization process, the inputs necessary for the process, and the relevance of generated outputs based on input assumptions must all be understood to ensure the effectiveness of the tool. In the end, topological optimization is only one tool of many that can be applied to AM design, and is best used as a guide when being utilized for the design or redesign of parts and structures for AM.
4.2.2 Part consolidation Part consolidation in the framework of AM is a process applied specifically to multiple part assemblies and structures. It is the method of using the competencies of the AM fabrication process to reduce a multiple part assembly made up of many components into a redesigned part that has the same operational functionality, but designed to include fewer overall components. The benefits of part consolidation in AM is part simplification, potential performance improvement, and reduction in necessary tooling and fabrication time. From a manufacturer’s standpoint, fewer components can be a substantial cost driver in reducing overhead cost associated with labor, tooling, part traceability, and inventory needed for that assembly. From an operator or user standpoint, it usually means easier use and maintenance of the product. For these reasons, part consolidation and simplification directly correlate to cost reduction for both suppliers and customers. The method of part consolidation in AM is straightforward in that an existing part or structure made up of multiple components is redesigned to minimize the number of components needed for functional operation of the part. Fig. 4.3 shows an example of part consolidation in an aero-engine cowl latch, where the components encompassing the handle assembly (Fig. 4.3A) have been consolidated into a single component (Fig. 4.3B). The difficulties with this method are not usually in the execution of the consolidation process, but in the selection of parts for the process that are appropriate. Typically, cost is the primary driver of part applicability for this process.
Figure 4.3 AM component consolidation of a jet engine cowl latch (A) initial latch handle assembly made up of 5 components, (B) AM redesigned handle with components consolidated into a single component, (C) additive manufactured jet engine cowl latch. AM, additive manufacturing.
72
Additive Manufacturing for the Aerospace Industry
Part consolidation through AM can be categorized, as can many applications of AM, as a disruptive process looking to displace an incumbent process for producing a part. For example, assume a part is currently made of 3 components, and the manufacturer of that part has a known order volume and cost of the part and its associated components. To displace the current manufacturing process using part consolidation through AM, the goal would be to consolidate the 3 components into a single new component with the same/comparable functionality. Still, for this to be an effective process for the manufacturer of this part, the new AM part made up of a single component must be at the minimum comparable in cost and time needed for fabrication to the incumbent process and product. Applying this general requirement has a limiting effect when evaluating possible parts and structures for AM redesign. As AM technology improves, and cost drivers to produce AM parts decrease, the family of parts to which this process can be applied should grow.
4.2.3 Part integration and repair The concept of part integration is not new to structure fabrication, but in the context of AM, it involves the process of combining components or subassemblies together using AM material as a medium. The benefits of part integration are application dependent, but a general benefit is utilizing this process to integrate separate parts using AM to reduce assembly issues or labor. A proposed example of this application is shown in Fig. 4.4 for aero-engine casing. The existing process for fabricating a turbine engine case includes complex machining steps whereby boss and pad features are machined out, holes are drilled and tapped, and specialty fasteners called inserts or studs are then threaded into these tapped holes. Drilling and tapping a hole on a large curved surface of a casing with accuracy is inherently challenging. Additionally, since superalloys are commonly used, this further increases the difficulty of machining such features. Machinists often invest significant time and resources machining a case only to reject the casing in the final stages of machining due to a faulty hole drilling/tapping or a faulty fastener
Figure 4.4 An aero-engine casing (A) a typical boss feature in a casing, (B) a typical issue during installation of inserts into boss hole, (C) a hybrid casing with an AM boss feature encapsulating a machined receptacle added onto a wrought base casing using AM. AM, additive manufacturing.
Design for metal additive manufacturing for aerospace applications
73
installation because of improper drilling/tapping. In this case, a specially designed receptacle would be manufactured and prepared to accept an insert or stud. The receptacles would be placed on the base casing and the boss or pad would be built around the receptacles using AM, tying the receptacles in place and making them an integral feature of the boss. This would likely need to be finish machined to meet the specified surface finish requirements, and inserts/studs would be added to the receptacle after machining. Since the receptacle has already been prepared to accept the inserts/studs, installation would be much easier and would reduce the risk associated with preparing a tapped hole to accept an insert or stud. It should be recognized that the process is not inherently limited to studs and inserts, but, in fact, can be used to add other types of hardware, such as bearings, bearing journals, bushings, inspections ports, or sensors. Another benefit of the process is its application in maintenance and repair operations, where a structure or part is repaired using AM when it has incurred some sort of damage to one of its components that is not easily repaired and replaced using conventional methods. Rather than scrapping the entire part or structure, AM material can be added to the damage site as replacement material reattaching a new component to the structure, and then performing any necessary secondary operations needed to return the overall part or structure to operation [8]. As stated initially, the use of part integration utilizing AM is application dependent, and should not be used for all applications involving the combination or repair of components. Characteristics of AM part integration that affect its application are material properties and microstructures seen along the interfaces between the AM material and substrate material. The microstructure of the interface is highly dependent on the AM process and parameters used for fabrication, and usually the AM material microstructure will typically not have the fine grain size seen in wrought materials. So, when applying this method to wrought parts, there will usually be some transition in microstructure and grain size at the interface, which may affect the mechanical performance of the structure or part depending on the specifics of the application (Fig. 4.5). In most applications, though, it is typically possible to manage AM process parameters to achieve performance characteristics of the assembly [9].
4.2.4 Other techniques One application of AM worth briefly mentioning is the use of the process as a means of increasing the bearing strength of a part or structure. Specifically using AM as a plating or cladding process to increase the wear resistance of a part, or
Figure 4.5 Inconel 718 alloy material interface between AM material printed atop a wrought substrate. AM, additive manufacturing.
74
Additive Manufacturing for the Aerospace Industry
using it to create a sacrificial layer along the bearing surface of a part, which is in contact with a secondary part or structure. This usually involves adding material onto a substrate of different base material utilizing an AM process. Challenges is this method usually stem from the quality of the additive and base material interface. Depending on the material choices and AM process used, such issues can arise from varying residual stresses during printing or the presence of anomalies and voids at the interface.
4.3
Process aspects of design
Design opportunities with AM, as discussed in the previous section, continue to be one of the key propellants for interest in AM. Regardless of the manufacturing techniques used for fabrication, all parts need to perform some function, and this is where the process aspects of design become critical. Is it possible to print a part successfully? What kind of performance can be expected from the parts? Is it economically feasible? The following section touches on these concerns focusing on the process aspect of AM part design.
4.3.1 Part performance With any part design, it is important to understand the underlying behavior of the material being used to simulate part performance. Traditionally, most designs assume for simplicity an isotropic material, and part sizing is then done accordingly. Compared with traditional manufacturing processes, AM does not just provide a great deal of flexibility with design, but it also creates new challenges. With the possibility of tailored microstructure, functionally graded material, and the ability to add features to a wrought substrate with a different microstructure, the traditional realms of design would need to be updated to include design based on location specific properties in the same part. Also, like most other manufacturing processes, the microstructure and resultant mechanical properties in AM parts are process route dependent. Here laser power, scanning strategy, or build orientation can significantly impact the properties of the part, or the occurrence and magnitude of defects. Fig. 4.6 illustrates this point. Ti6Al4V properties are shown here as reported in literature [1013]. Such variation in properties of the material is not observed with typical manufacturing techniques. The pedigree of the material data used for part design needs to be fully understood to ensure its similarity to the intended AM process in order to be confident in the simulation predictions of the optimized part.
4.3.1.1 Microstructure There are several public and private entities engaged in AM. Process parameters and their effects on mechanical properties or distribution of defects tend to be
Design for metal additive manufacturing for aerospace applications
75
Figure 4.6 Boxplots of Ti-64 static mechanical properties at room temperature as reported in literature [1013].
Figure 4.7 Typical 15-5PH microstructure, with no heat treatment, using a typical laser powder bed AM system. AM, additive manufacturing.
proprietary company information, and little is reported in published literature. However, some key points are listed over here. Since AM deposition is a layer-by-layer process, the microstructure of the material reflects this effect. Fig. 4.7 shows a typical microstructure observed with
76
Additive Manufacturing for the Aerospace Industry
Figure 4.8 EBAM Ti-64 macrostructure, as-deposited, section normal to X direction. EBAM, Electron beam additive manufacturing.
laser powder bed printed 15-5PH material. The process involved is the melting of multiple layers of powder as the laser passes through. In the image, layer height can be clearly differentiated. Fig. 4.8 shows a similar effect in electron beam AM (EBAM) printed Ti6Al4V material. This behavior typically has limited effects on tensile behavior but, depending on material type, may significantly affect fatigue performance. Microstructural features are usually much finer with AM processes when compared to cast parts. This is due to the smaller melt pool and resultant rapid solidification [1416]. The solidification rate is much higher than what is possible by traditional casting techniques with up to 104 K/s possible reported with Laser Engineered Net Shaping (LENS) technique [14]. This provides higher strength to material via Hall-Petch effect due to smaller grain size. However, there is a range of AM techniques possible with widely varying deposition rates (0.05 kg/h for powder bed systems to 7 kg/h for some EBAM techniques), leading to resultant changes in microstructure. This does provide opportunities to modulate the microstructure per performance needs; for example, finer grain sizes at an interface to improve tensile/fatigue properties with larger grains at other areas that are not critical, lowering the associated deposition costs. Orientation effects are also observable with the AM microstructure. Microstructural features may be refined in one orientation and elongated in another. Fig. 4.9 shows this effect for Ti6Al4V deposited via EBAM. Since the deposition substrates (build plates) are typically much cooler than the build media, a preferential heat transfer direction is setup during cooling of the molten metal. This establishes elongated grains in plane normal to the deposition substrate. Careful control of deposition parameters may alleviate this effect.
Design for metal additive manufacturing for aerospace applications
77
Figure 4.9 EBAM Ti6Al4V macrostructure, as-deposited at 200 3 magnification. (A) Section normal to X direction and (B) section normal to Z direction.
If the goal is to use AM as a near-net-shape technique for producing parts, there is very limited thermo-mechanical work that can be done in the material to refine the structure further. There are some techniques, like AmpliFORGE, which are using a hybrid approach by starting with an AM preform to reduce multiple forging costs, and then refining the structure using a one-step forging operation [17].
4.3.1.2 Defects Defects may be possible in AM parts. Lack-of-fusion defects and gas porosity or voids are some of the more commonly observed defects [14,18]. The occurrence and magnitude of these are dependent on the AM technique applied, and the corresponding process parameters. With appropriate part/process design and process controls, these can be significantly minimized or eliminated. A discussion on this is beyond the scope of this chapter. Additional details on defects may be found elsewhere in the book.
4.3.1.3 Mechanical properties Static mechanical properties of AM parts (tensile strength) are typically very close or slightly lower than their wrought counterparts. Elongation values may be lower depending on the occurrence of defects. Mechanical properties also show an orientation dependence, as expected, due to the microstructure changes. X/Y (directions parallel to the build surface) tensile values are typically higher than Z (direction perpendicular to the build surface) tensile values, and the degree depends on the AM technique applied and process parameters. Some specifications like AMS 4999 for Ti6Al4V show the typical directionality associated with AM process. Fatigue values are significantly affected by the AM process and tend to be lower than wrought counterparts. Fractography shows that fatigue cracks initiate faster in AM parts due to presence of voids/porosity near surface (see Fig. 4.10). Process parameters that reduce the occurrence of such porosity/voids are critical in improving fatigue life. Another way to address the issue is with the use of hot
78
Additive Manufacturing for the Aerospace Industry
Figure 4.10 Example of two 15-5PH H1025 fatigue samples tested at different stress levels showing the origin of fatigue cracks near surface.
isostatic pressing (HIP). This reduces the size of the pores, lowering the stress concentration factor, and improving fatigue life, but at the expense of adding cost to the final part. Lack-of-fusion defects, which typically have high aspect ratios, are not desired as they may lower fatigue life more so than spherical pores. Since AM technology is still at its early stages, other properties like fracture toughness and creep have not been evaluated as extensively. It is worth noting that AM part performance is tied directly to process parameters, which can be tuned per requirements.
4.3.2 Part quality Any AM parts producer would need to manufacture parts that meet their customers’ quality requirements. These requirements may be dimensional, surface finish, microstructure, or mechanical property requirements. The following are some aspects of the AM process that should be investigated when trying to ensure a specified level of part quality: G
G
G
Resolution: Feature resolution is dependent on the AM technique applied and on the deposition rate. Typically, with a higher deposition rate, the feature resolution goes down. For example, the feature resolution with a powder bed technique is in the range of 0.1 mm whereas with higher deposition techniques like EBAM this goes down to 3 mm. Therefore, selection of an AM technique to be applied is also dependent on what kind of feature resolution is desired, along with the post processing needs of the part. Surface finish: AM surfaces are typically rougher than casting counterparts and the degree depends on the process conditions used. Fig. 4.11 shows the typical surface roughness in a 15-5PH part produced using a laser powder bed process. Most customers have a surface finish requirement in their parts that the AM process would need to meet. Residual stress and build failure: Since molten metal is being deposited and solidified onto a substrate, there is potential for significant residual stresses in the AM build. Residual stresses can influence part performance, part geometry, and, specific to AM, can also affect the likelihood of a build failure. If a part becomes heavily distorted during a build, the AM system being used could prematurely stop resulting in a build failure. To mitigate the generation of residual stresses within a part, typically the substrate or “build
Design for metal additive manufacturing for aerospace applications
79
Figure 4.11 Typical surface roughness of a powder bed 15-5PH as-printed part.
G
G
G
G
plate,” on which an AM part is constructed, is heated to a specified temperature that can be process (laser or EBAM) or powder dependent. Build layout and support: The layout of parts on the build substrate as well as the allocation of support structure, if needed, are both highly critical to build success. Defining of these build inputs requires a thorough understanding of the specific AM process to be used, the desired part functionality, and the associated secondary operations that will be needed to achieve part quality requirements. Effect of powder reuse: In most builds with powder bed systems, powder usage per build is less than 10%. In blown powder AM systems, the usage is still around 30% or less. Therefore, a significant amount of powder that has been exposed is left for reuse after a build. The powder size distribution, contamination and hence characteristics of the built part may vary with the amount of reuse. Customers/manufactures typically put a limit on how many times a powder exposed to the AM build process can be reused. Open atmosphere systems: With efforts on reducing the costs of AM systems, there is a recent trend toward print-head systems that can add AM capability to a conventional Computer Numerical Control system. While lowering the costs significantly, they typically do not have a controlled atmosphere for builds, which is challenging for controlling contamination in reactive metals like titanium alloys. Geometric features: Certain features of a part can be more difficult to produce than others, and as such a build can become inadvertently more complicated because of it. A way of mitigating complexity of a build is choosing an AM build layout that simplifies cumbersome features. An example of this could be reorienting a part to reduce overhanging features, reducing the amount of necessary support needed for fabrication. As was mentioned before, AM system resolution also plays a substantial role in the complexity of shapes and geometries that can be produced, depending on the desired size and scale of the geometries.
4.3.3 Part evaluation: in-situ and after process nondestructive evaluation (NDE) To ensure a defect-free part, nondestructive techniques need to be applied both insitu and postbuild on an AM part. This is an important portion of quality control of
80
Additive Manufacturing for the Aerospace Industry
the part, and is often dictated by customer and part requirements. A complete description of this topic for AM is beyond the scope of the chapter, but the following are some key points of which to be aware: G
G
G
Incoming powder characterization: This is critical to ensure quality and consistency of input material. Parameters like particle size distribution, flow rate, chemistry, and morphology are important, and should be measured and tracked with different material lots. In-situ monitoring: This is critical for assessing parts while printing to detect and possibly mitigate real-time defects. This includes features like melt pool control, thermal imaging with each deposition layer, and closed loop controls. Postprocess NDE: Techniques like computed tomography (CT) scan, X-ray, and traditional microscopy are important to guarantee quality of a part. Fig. 4.12 shows an example of how CT scanning can be used to detect defects in an AM part.
4.3.4 Post processing AM is sometimes referred to as “3D printing.” This terminology often gives an impression that we can easily fabricate a finished metal part, basically making a 3D CAD digital model tangible merely with the push of a button. In most AM applications, though, this is not usually the case. What part manufacturers are realizing is that just like traditional manufacturing, the actual process of printing itself is but a small portion of tasks needed to make an AM part. The post processing of an AM part turns out to include a significant portion of the overall cost of the part, and thereby must be given equitable attention. Key points to keep in mind when evaluating the amount of post processing needed for an AM part: G
Ease of powder removal: This is an important consideration with powder bed AM parts. Parts need to be designed such that powder in not entrapped in the final part. Build supports also need to be designed in such a way that it is easier to remove powder after the build process has completed.
Figure 4.12 CT scan of an EBAM part showing linear lack-of-fusion defect in a suboptimized Ti6Al4V part. EBAM, Electron beam additive manufacturing.
Design for metal additive manufacturing for aerospace applications
G
G
G
G
81
Minimizing postprocess machining: Most powder bed AM parts would need some level of machining after printing. Part orientation and support structures should be designed such that machining operations and related time required are minimized. For example, orienting the part on the build plate such that the electrical discharge machining (EDM) operation required to remove the part from the plate also gives the cutting surface of the part the required specified finish. Minimizing postprocess surface treatments: Due to fatigue, wear, and other performance requirements, most parts have surface roughness requirements as mentioned in the previous section. Since the attractiveness of AM is dependent on it being a near-net-shape process, it is imperative that costly postprocess surface treatments are kept in mind to be minimized while designing the build layout. Heat treatment: Typically, most AM parts would need heat treatment either to reduce the residual stress from the build, improve microstructure uniformity, or enhance other microstructural/mechanical property characteristics. However, higher temperature heat treatments or rapid cooling rates post heat treatment may cause distortion in the parts and therefore should be addressed accordingly. HIP: Most AM parts after heat treatment show lower fatigue properties as compared to their wrought counterparts. One way to enhance fatigue properties in AM material is through HIP’ing. The inclusion of this process can add significant cost to the part, in which case the performance aspects of the application should be considered carefully prior to the design and selection of fabrication and secondary processes.
4.4
Cost considerations
To find a successful commercial application, AM parts need to be both technically and economically feasible. Most areas where AM currently brings value is with expensive materials and highly complex parts where a significant amount of machining is required. AM part cost breakdown for a typical powder bed AM process is shown in Fig. 4.13. These costs can be higher or lower based on the specific parts being printed, the material used, and what AM technology is employed. Equipment and material costs continue to be significant cost drivers for AM. As well, the labor involved with secondary operations having to do with part support removal and additional machining needed to produce final part geometries, can also have a substantial cost impact when evaluating the economic feasibility of producing a part using AM. However, with rapid advancements in the industry, the availability of newer faster machines, and the increasing customer demand for AM material stock, these costs are beginning to come down significantly. Some aspects about metal AM which should be considered when determining whether it is a viable economic process for an application are: G
If the application is for an industry that is heavily regulated, such as aircraft, what is the economic impact of qualifying an AM process. This could mean qualifying the powder or media used to fabricate the part, the specific AM machines that will be fabricating the actual parts, and/or developing procedures and specifications governing the process [19,20]. These are often the hidden costs of using/qualifying a new process for a part that
82
Additive Manufacturing for the Aerospace Industry
Figure 4.13 Typical cost breakdown of an AM part fabricated using a powder bed process and has undergone secondary machining operations to produce final part geometries. AM, additive manufacturing.
G
G
G
G
can go from a few thousand dollars to potentially millions of dollars for new materials, and can easily throw off the economics of producing the part via AM. Capital costs associated with purchasing and maintaining AM systems for manufacturing. At current levels, this typically dominates the AM part costs [21]. Cost of the AM media from which the parts will be fabricated compared to wrought cost of the same or comparable materials. Material costs can often be in the neighborhood of 20% of the total part costs. Determining necessary inspection processes, both destructive and nondestructive methods, required for evaluating part or product performance, and demonstrating that the product meets the desired part specifications. Secondary operations that will be needed after parts have been fabricated using an AM process. These operations may include support removal, surface treatments, heat treatment, or HIP.
These aspects as well as the inherent indirect costs associated with designing a part and build layout for the AM process should be strongly considered when determining the economic viability of implementing AM in a specific application. The current trend in the industry is to evaluate the economic feasibility of fabricating a part using AM, and comparing it to the existing incumbent fabrication processes such as a casting or a forging. However, what is often missed in this exercise is that the part being evaluated was designed for a casting or forging and not for AM. A part designed specifically for AM may add significantly higher value to the AM part. Also, just because a part can be fabricated using AM does not mean that it should be. To justify the use of AM, the part should make both technical and economic sense.
4.5
Product and process design tools
AM as an emerging technology is currently undergoing several paths of development in both process and technology. Current avenues of AM development include,
Design for metal additive manufacturing for aerospace applications
83
but are not limited to, additive machine development, process development, AM material characterization and production, inspection methods both during process and postprocess, design tools, and process synergy. In parallel with these direct focus areas of AM development is allied technology development through accompanying software tools. Discussed in this section are some of the current capabilities and areas of development for AM software tools.
4.5.1 Additive manufacturing design software The development of AM design software has primarily been focused in two areas: developing free-form design capabilities, and part shape and topography optimization. These avenues are currently being addressed by many of the industry computer-aided design (CAD) software producers as well as industry FE analysis software companies [2223]. In terms of free-form design, CAD programmers have begun to create tools and platforms allowing an AM part designer to be able to create 3D bodies that do not constrain or inhibit the capabilities of the AM fabrication process. Essentially this allows a CAD software package to be able easily to design a part that can be fabricated using the full breath of capabilities of the AM process. As well, some CAD packages are beginning to develop software features where the AM part designer can identify the AM process used to fabricate the part. The software will then apply specific constraints, such as geometrical or material constraints, on the 3D model that are representative of that process. Shape and topological optimization, as previously mentioned in this chapter, are design methodologies that are used heavily in AM part design because they allow for tailored part performance to specified applications with the objective of reducing part volume and weight. These methods of optimization, like free-form CAD modeling, are not new tools, but there has been a recent push in their development as applied to AM processes. These tools typically utilize existing numerical methods and algorithms based on FE analysis or computational fluid dynamics, for evaluating the performance of a part. Development in both these areas of AM design software applications, free-form design and part optimization, has been collaborative with many packages working toward having capabilities in both areas to work in tandem. An ultimate goal of development would be having the ability in a single environment to optimize and refine the design of a part to its final form, and transfer the digital file to an AM system for printing.
4.5.2 Additive manufacturing process software The AM process is a multiphysics process involving the transformation of starting powder or wire into a final product through heating, melting, and solidification operations resulting in a layer-by-layer construction of a part or feature. Understanding of this complex process is aided through the incorporation of
84
Additive Manufacturing for the Aerospace Industry
software tools, which use physics principles to model different aspects of the process. Current software tools being developed to aide with the understanding of the AM process have been focused on the following areas of interest, but are not limited only to these areas presented: G
G
G
G
G
Modeling of residual stresses in an AM part or feature during a build, and having a complete residual stress profile of the part at build completion. Being able accurately to represent the microstructure of a part or feature being produced using AM. Understanding how the melt pool and thermal gradients within a build, along with the raw material properties used to fabricate an AM part, effect the final part’s microstructure and chemistry. Build layout design, and understanding the effects of building multiple parts versus a single part in an AM build with respect to part functionality, induced residual stresses within parts, and part microstructure. Determining optimized part geometric orientation in a build layout both to ensure successful build completion and reduce build time. Modeling the effects of secondary operations performed on a part during the process of a build or after build. Secondary operations can include machining/material removal, HIP, surface treatment, and heat treatment.
Each of these focus areas entail their own complexity, but all have the general goal of developing the sophistication necessary to understand the AM process in its entirety, ability to predict a successful build, and predict part performance as a function of process parameters.
4.6
Conclusions
In the present chapter, a review of the techniques used in DFAM, where the techniques of topological optimization, part consolidation, and part integration were discussed in some detail. Process aspects of AM design were reviewed, illustrating the effects of material properties, part performance, and post processing operations in relevance to part design. Current industry design tools for both part and process design were summarized. The discussions emphasized that the traditional notions of part design need to be updated to include key process elements when evaluating design decisions for AM part fabrication.
Acknowledgments The authors would like to thank all the members of Arconic Fastening Systems-New Product Development Center, Arconic Engines, and the Arconic Technology Center, whose efforts were critical in designing of parts, manufacturing prototypes, and testing of samples used in the presented work. The authors would like to extend a special note of thanks to Luke Haylock, Dr. Hasim Mulazimoglu, and Rodrigo Pinheiro for their advice and support through different stages of the described work.
Design for metal additive manufacturing for aerospace applications
85
References [1] L. Yang, K. Hsu, B. Baughman, D. Godfrey, F. Medina, M. Menon, et al., Additive Manufacturing of Metals: The Technology, Materials, Design and Production, Springer, 2017, pp. 81152. [2] I. Gibson, D.W. Rosen, B. Stucker, Additive Manufacturing Technologies: Rapid Prototyping to Direct Digital Manufacturing, Springer, 2010, pp. 283314. [3] T. Wohlers, Wohler’s Report 2013, 2013. [4] Arconic Global, The Smart Ink of the Future, Arconic Global Website, 2017, ,https:// www.arconic.com/global/en/who-we-are/3d-printing-technology.asp.. [5] GE Additive, Additive Manufacturing Revolutionizes Design Possibilities for GE Engineers, GE Additive Website, 2016, ,https://www.ge.com/additive/press-releases/ additive-manufacturing-revolutionizes-design-possibilities-ge-engineers.. [6] Composite World, Airbus Expands Use of Additive Manufacturing, Composite World Website, 2014, ,https://www.compositesworld.com/news/airbus-expands-use-of-additive-manufacturing.. [7] M. Gorelik, Additive manufacturing in the context of structural integrity, Int. J. Fatigue. 94 (2017) 168177. [8] A. Uriondo, M. Esperon-Miguez, S. Perinpanayagam, The present and future of additive manufacturing in the aerospace sector: a review of important aspects, Proc. Inst. Mech. Eng., G: J. Aerospace Eng. 229 (2015). [9] M. Kamal, L. Haylock, G. Rizza, M. Travis, Component integration with hybrid assembly using additive manufacturing, in: SAE Aerotech Congress, Forth Worth, TX, 2017. [10] H.C. De Groh III, Development of laser fabricated Ti6Al4V, NASA/TM—2006214256, 2006. [11] A.M. Beese, B.E. Carroll, Review of mechanical properties of Ti6Al4V made by laser-based additive manufacturing using powder feedstock, JOM 68 (3) (2016) 724734. [12] L. Bian, S.M. Thompson, N. Shamsaei, Mechanical properties and microstructural features of direct laser-deposited Ti6Al4V, JOM 67 (3) (2015) 629638. [13] R. Grylls, LENS Process White Paper: Fatigue Testing of LENS Ti-6-4, Technical Brief, Optomec Inc., Albuquerque, NM, 2014. [14] W.E. Frazier, J. Mater. Eng. Perform. 2014 (1917) 23. Available from: https://doi.org/ 10.1007/s11665-014-0958-z. [15] ,https://manufacturing.llnl.gov/additive-manufacturing/metal-additive-manufacturing., 2017. [16] S. Gorssea, C. Hutchinson, M. Goune´a, R. Banerjee, Additive manufacturing of metals: a brief review of the characteristic microstructures and properties of steels, Ti6Al4V and high-entropy alloys, Sci. Technol. Adv. Mater. 18 (1) (2017) 584610. Available from: https://doi.org/10.1080/14686996.2017.1361305. [17] D. Bush, B. Bodily, H. Watson, M. Chastka, E. Colvin, G. Satoh, Arconic development of the ampliforge process, in: AeroMat 2017 Conference, South Carolina, April, 2017. [18] A. Debiccari, J. Moor, B. Lagow, S. Tewari, M. Kinsella, Additive manufacturing for superalloys—producibility and cost validation, in: Seventh International Symposium on Superalloy 718 and Derivatives, 2010. [19] MSFC-STD-3716, Standard for Additively Manufactured Spaceflight Hardware by Laser Powder Bed Fusion in Metals, 2017.
86
Additive Manufacturing for the Aerospace Industry
[20] MSFC-SPEC-3717, Specification for Control and Qualification of Laser Powder Bed Fusion Metallurgical Processes, 2017. [21] C. Lindemann, U. Jahnke, M. Habdank, R. Koch, Analyzing product lifecycle costs for a better understanding of cost drivers in additive manufacturing, in: 23rd Annual International Solid Freeform Fabrication Symposium—An Additive Manufacturing Conference, SFF 2012, 2012. [22] Autodesk Netfabb, Material Solutions for Additive Manufacturing Whitepaper, Autodesk Netfabb, 2016 ,https://www.autodesk.com/products/netfabb/overview.. [23] J. Fort, S. Sett, How simulation can help advance additive manufacturing technology, Simulia Community News, 2015 ,https://www.3ds.com/products-services/simulia/ resources/how-simulation-can-help-advance-additive-manufacturing-technology/..
Structure formation in A.M. processes of Titanium and Ni-base alloys
5
I.S. Polkin1, S.V. Skvortsova2, G.A. Turichin3 and M.B. Novikova1 1 All-Russia Institute of Light Alloys, JSC, Moscow, Russia, 2Moscow Aviation Institute (National Research University), Moscow, Russia, 3St. Petersburg State Maritime Technical University, St. Petersburg, Russia
Activities on the use of powder metallurgy methods for the manufacture of critical structural components have been in progress for more than 40 years. Sufficiently effective results were obtained for parts made of heat-resistant nickel and titanium alloys. During this time, many important issues were solved to prove the possibility of a significant reduction in the number of process stages and in the scope of machining of parts obtained by powder technology as compared with conventional deformation technologies. The achieved results speak for the possible decreasing of metal consumption by 23 times, and that for disk titanium materials by up to 35 times [1]. Further progress of research work on achieving high solidification rates of powder materials made it possible to start solving problems of increasing mechanical properties. An increase in cooling rate allowed us to increase the content of alloying elements in the solid solution and, due to this factor, to obtain higher mechanical properties than those of parts made by using conventional technologies. First, such advantages were obtained with the use of heat-resistant nickel alloys, when high rate of crystallization of fine powders facilitates formation of a large amount of the γ0 -phase during decomposition of the solid solution and refinement of initial intermetallic compounds due to changes in their chemical composition. In order to obtain such effects in titanium alloys it was necessary to achieve higher cooling rates than those of nickel alloys, at the same time an improvement in properties could be obtained only for special alloys of limited application. As for titanium alloys, therefore, it was possible to concentrate on solving the first problem—decreasing the metal consumption and reducing the cost of manufacturing parts. At the same time, the calculations have revealed that in case of production of titanium parts by conventional deformation technology, up to 30% of total costs of the entire process accrue to the machining, whereas in the case of powder technology, the cost of machining is much less (Fig. 5.1). The next evolutionary stage in the development of the structural materials production technology was the use of 3D printers to join individual particles (alloyed Additive Manufacturing for the Aerospace Industry. DOI: https://doi.org/10.1016/B978-0-12-814062-8.00006-6 © 2019 Elsevier Inc. All rights reserved.
88
Additive Manufacturing for the Aerospace Industry
Figure 5.1 Costs of production of Ti components by conventional technologies.
powders) into a finished part. This technology has “absorbed” all previous achievements in powder technology providing consolidation of particles in a hot isostatic press (HIPer) under the action of temperature and pressure, and a metal can was used as the shape-building element. As for the new additive manufacturing, the can, as a shape-building element, was replaced with a 3D printer, which controls the temperature beam in the 3D space and builds up the final part by joining the powder particles. The economic effect of application of additive technologies was caused by the shortening of processes to make, fill, seal, and remove cans after HIPing, and because of almost complete absence of machining. We can assume that the cost of manufacturing parts by this technology will be approximately 30% lower than that of conventional powder technology. It should be taken into account that this method makes it possible to manufacture parts with a shape that cannot be obtained by any other way, for example, hollow spaces or lattice structures that allow achieving a significant reduction in weight. Consideration of all methods of additive manufacturing, which have already stood the test of time, allows us to identify three main ways that have proved their usefulness: G
G
direct deposition of metal powders on the forming surface of a part [direct metal depositions (DMDs)]; an action of a heat beam on the surface of the powder platform; only that part of powder on the platform is consolidated, which is exposed to the beam [selective laser melting (SLM)];
Structure formation in A.M. processes of Titanium and Ni-base alloys
G
89
and, finally, when a beam controlled by a 3D printer does not act on the powder, as it is in the first two processes, but on a thin wire that is welded layer by layer and builds up a desired shape of the part so called wire arc additive manufacturing (WAAM).
Having the same end goal, that is, to obtain the final 3D model of a part set by the printer, these processes had different technological parameters: construction rate, beam power, size of the initial material, metal temperature at different stages of the process, the availability of additional equipment, and a number of other differences. In addition to provision of the final shape, overcoming these differences causes one and probably the most important task of obtaining sufficiently high mechanical properties that ensure reliability of the part working under operations conditions. The achievement of necessary properties of parts made by additive manufacturing is fully dependent on the absence of defects and on the microstructure characteristics. The solution of these tasks depends on the level of technology and the ability to manage its parameters. For example, Fig. 5.2 shows the liquid bath dimensions for two cases, that is, when the beam scanning speed changes from 50 up to 200 mm/s, and the beam power increases from 100 to 200 W. In each case, the liquid bath dimensions can vary approximately twofold, and the particles of 40 and 100 μm in size can be completely liquid or be in a partially solid/partially liquid state. The process of joining particles and layers formed from them occurs in both cases, but the microstructure of the interface will be different, and therefore the probability of formation of a defect on it will be different. Undoubtedly, the task of technologists to build up a model will become more complicated, since the temperature of the previous particle or layer will constantly change, depending on the size of the part and other parameters of the construction. The knowledge of the nature of changes in the structure during the production process of the part is extremely necessary to achieve stable and high mechanical properties. Let us consider the initial structure of powder particles and the nature of its changes in the process of manufacturing 3D parts.
5.1
Evaluation of the structure of powder particles of different sizes
Titanium and nickel alloy powders of 40120 μm in size produced by plasma rotate electrode process (PREP) were used in the present work. Electron microscope examination of the surface of powder particles of various sizes showed that all particles had a smooth surface, strongly pronounced round shape, and they were free of satellites, pores, and other defects that cause high fluidity and bulk density of such powders. Examination of the surface structure made it possible to reveal the dendritic structure of the surface for powder particles of all sizes, which was typical for both titanium and nickel alloys (Fig. 5.3).
90
Additive Manufacturing for the Aerospace Industry
Figure 5.2 Changes in liquid bath dimensions in case of increasing of scanning rate and laser power.
Investigation of the dendritic structure over the surface of powder particles of various sizes made it possible to reveal a regular decrease in size of the dendritic cells, accompanied by a refinement of powder particles. It was revealed most clearly in cases of measuring the specific length of the dendritic cell boundaries, which form the surface relief. No less than 25 powder particles of each size were studied by a scanning electron microscope with magnification of 20003000. It was also shown by studies that the more alloying elements in an alloy, for example, Grade 5 (Ti6Al4V, 10% alloying elements), VT22 (18% alloying
Structure formation in A.M. processes of Titanium and Ni-base alloys
91
Figure 5.3 Surface structure of titanium and nickel-based alloys.
elements) and EI698 (30% alloying elements), the smaller are the dendritic cells formed on the surface of powder particles of the same size, although the general pattern of their dependence on the particle size is kept (Fig. 5.3). Fig. 5.4 shows a dependence of the specific length of the dendritic cell boundaries on the powder particles surface on the size of powder particles (μm/μm2). As it follows from the Fig. 5.4, a decrease in the average size of powder particles from 140 to 30 μm leads to an almost twofold increase in the specific surface of boundaries, that is, from 0.55 to 1 μm/μm2. It should be noted that the extreme growth of
Additive Manufacturing for the Aerospace Industry
Specific length of dendritic cell boundaries, μm/μm2
92
1.6 1.4
y = 4.57 x–0.42 R² = 0.843
1.2 1 0.8 0.6 0.4 0.2 0
0
50
100
150
200
250
300
350
Diameter of powder particles, μm
Figure 5.4 Dendritic structure parameters of the powder particle surface as a dependence on the particle size by the example of TiAlMoZrSn alloy.
this value begins when the powder particle size is less than 40 μm, and, in case of refining the particles from 300 to 140 μm, the change in the specific surface area of the dendrite boundaries is rather small. Analysis of the obtained data allows us to describe the data by the following equation: y 5 Ax2B ; where y—specific length of the boundaries; x—diameter of the powder particles; A —coefficient (3.95.6); B—coefficient (0.390.40).
5.2
A dependence of the microstructure of powder particles in the initial state on their size
The microstructural study of Grade 5 alloy powder particles showed that the particles of less than 50 μm and within 150200 μm in size, being in the initial state, have the martensite structure and the length of martensitic plates is equal to the entire diameter of the particle (Fig. 5.5); almost exclusive presence of β-phase was observed in powder particles of alloys with higher content of alloying elements, such as VT22, which was confirmed by X-ray structure analysis. Thus, it was difficult to determine the difference in the internal structure of powder particles of different sizes in their initial state when they have been cooled at B1 3 10241025 С/s, because the single-phase state is developed due to the highrate cooling.
Structure formation in A.M. processes of Titanium and Ni-base alloys
93
Figure 5.5 Microstructure of Grade 5 alloy powder particles of ,50 and .150 μm in size in the initial state.
5.3
Determination of changes in the structure using samples produced by different additive technologies
Two methods of production of a compact preform were used for preparation of samples: high temperature DMD and consolidating of powder particles located on a platform under the action of a thermal beam, that is, SLM. The optimum temperature-time parameters of the process were chosen on the base of investigation results of the compact preform structure. Grade 5 alloy parts were obtained by the selective beam melting and by the direct deposition of powders onto the part to be built up. In the first case, the powder sizes were within 3040 μm, and in the second case—within 6080 μm. Microstructure of samples in the initial state produced by DMD and SLM processes is shown in Fig. 5.6A and B. The samples were cut out in the direction that was transversed to the direction of the layer build up. The presented photos demonstrate rather similar type of microstructure, which is characteristic for these production processes. In both cases, the boundaries of deposited layers are decorated with α-phase precipitations. Investigations of the longitudinal sections structure of the built-up layers have revealed more clearly elongated boundaries of layers with α-phase precipitations, while the structural directivity is not visible in the transverse direction Fig. 5.6C and D. In addition, metallographic studies showed that in the case of SLM, such defects as individual pores or other small defects occur much more often than in the case of direct deposition. Apparently, it can be explained by the increased energy of transfer of particles onto the part to be built up in the course of DMD process, which allows one to reach a higher degree of densification of the depositing metal particles. The analysis of samples made by additive manufacturing showed that after annealing (750 C, 2 hours in air) the microstructures in the initial state and after annealing are quite similar for two cases of buildup of a part (Fig. 5.7).
94
Additive Manufacturing for the Aerospace Industry
Figure 5.6 Microstructure of Grade 5 alloy samples in initial state produced by SLM and DMD ( 3 600): (A and B)—microstructure of samples cut out transversely; (C and D)— microstructure of samples cut out longitudinally. SLM, selective laser melting; DMD, direct metal deposition.
Figure 5.7 Microstructure of Grade 5 alloy samples ( 3 600) produced by (A) selective laser melting and (B) direct metal deposition after annealing at 750 С, 2 h.
The absence of significant differences in the microstructure of the samples in the initial state and after standard annealing can apparently be attributed to the time and temperature sufficient for recrystallization processes already in the process of building up the part, and the standard annealing temperature does not change noticeably the structure of the alloy.
Structure formation in A.M. processes of Titanium and Ni-base alloys
95
Figure 5.8 Microstructure of Grade 5 alloy samples ( 3 600) produced by SLM: (A) after annealing at 850 С, 1 h, in air (B) HIPing at 950 С, 1.5 h, Р 5 142 MPa. SLM, Selective laser melting.
An increase in the annealing temperature of samples up to 950 C leads to a significant increase in length and thickness of α-phase plates, both in longitudinal and transverse directions. HIP treatment of samples produced by SLM eliminates reliably pores and discontinuities, but increases additionally the dimensions of α-phase plates (Fig. 5.8). Investigation of microstructure of the samples produced by DMD and SLM methods showed an identical type of the structure both in the initial and in heattreated states; it is also important that Grade 5 alloy powder particles of 6080 and 3040 μm in size were used for DMD and SLM, respectively.
5.4
Testing of mechanical properties of samples of parts produced by direct metal deposition and selective laser melting
Mechanical tests, i.e. ultimate tensile strength (UTS), and yield strength (YS), were carried out using samples produced by DMD and SLM. Tests were carried out with the use of samples in the initial state, after heat treatment and HIPing. Table 5.1 shows properties of Ti6Al4V alloy samples produced by SLM. It can be seen in Table 5.1 that the initial material has a higher strength and reduced plasticity as compared to the samples tested after heat treatment and HIP. It is probably related to stress relieving and some increase in sizes of the α-plates, taking into account that the heat treatment and HIP operations were carried out in the upper part of the two-phase region (α 1 β). A similar pattern of changes in structure and properties after heat treatment can be observed with the use of samples obtained by DMD technology (Table 5.2). Table 5.2 shows that the heat treatment, in comparison with the initial state, causes a reduction in tensile strength by 40 MPa, in yield stress—by 50 MPa, and reduction in area increases by approximately 6%.
96
Additive Manufacturing for the Aerospace Industry
Table 5.1 Mechanical properties of Grade 5 alloy samples produced by SLM State
Cutting direction
UTS, МPа
0.2% YS, МPа
δ, %
ϕ, %
Initial state
Longitudinal Transversal Longitudinal Transversal
990 980 950 930
950 965 930 920
13 11 16 17
44 38 46 48
δ, %
Ψ, %
Annealing: 850 С, 2 h HIP: 950 С, Р-1400 atm, holding 1.5 h
SLM, Selective laser melting; UTS, ultimate tensile strength; YS, yield strength.
Table 5.2 Mechanical properties of Grade 5 alloy samples produced by DMD State
Sample cutting direction
Average value UTS, МPа
Initial state Heat treatment: 800 С, 1 h, in air Heat treatment: 800 С, 1 h, in air
0.2% YS, МPа
Longitudinal Longitudinal
1029 992
978 927
18 19
46 52.5
Transverse
1057
1017
14
34
DMD, direct metal deposition.
Tests of samples cut out in the plane transversed to the growth direction of layers produced by DMD showed higher strength values and a decrease in elongation as compared to samples cut out along the direction of the layers growth. The reduced plastic properties obtained at the testing of samples are evidently associated with a large number of grain boundaries presented in the cross section of the sample, possible defects and discontinuities occurring at the junction of layers, and coarse α-phase interlayers. In addition to the noted features in the structure and properties of samples cut out in different directions, it is necessary to point out a large number of discontinuities and pores formed in samples produced by SLM. That is indirectly confirmed by the fatigue tests of these samples, fatigue resistance values of which, after HIPing, are 1.31.5 times higher than those before HIPing, which is most probably caused by the healing up of pores and discontinuities during HIPing. Thus, consideration of mechanical properties of samples produced by DMD and SLM methods showed that in both cases higher strength values are obtained in the initial state as compared with properties of these samples after subsequent annealing or HIPing. Elongation characteristics after annealing are increased in comparison with the initial state. It can be explained by the processes of stress relief and equalization of the structure with increase in the size of α-phase plates by 12 μm; the plates are
Structure formation in A.M. processes of Titanium and Ni-base alloys
97
noticeably growing when annealing or HIPing is carried out at temperatures of the upper part of the (α 1 β) region. The second common property of the samples produced SLM and DMD is a difference in mechanical properties and structure of samples in different built-up directions. Unfortunately, this imperfection is inherent in the methods in question and further research to achieve higher homogeneity of the built-up structure is needed. The structure and properties obtained with the use of Grade 5 alloy samples exceed significantly those of cast samples and are close to the values specific for wrought material. There are some exception when defects in the form of cold shuts can be formed, which are not completely eliminated by HIPing.
5.5
Conclusions
1. Investigations of the structure and properties of powders and parts produced from them by additive technology have shown that the initial powder particles, as a rule, have a singlephase structure, and temperature effects arisen in the course of additive manufacturing, both by DMD and SLM, cause formation of a two-phase structure analogous to that of the alloy in the annealed state. 2. The annealing leads to a slight change in properties and homogenization of the structure and to a certain increase in the particle size of the α-phase. 3. The mechanical properties and structure of the samples manufactured by DMD and SLM methods are similar and did not show differences when using powder particles of 6080 and 3040 μm in size for both methods. 4. Depending on the direction in which the layers are grown in the samples, the anisotropy of properties necessitates the further improvement of the additive manufacturing technology.
Acknowledgement We would like to express our gratitude to the employees of Е.М. Golubeva, VILS OJSC, М.А. German, Moscow Aviation Institute (National Research University), and F. Shamray (St. Petersburg State Maritime Technical University) for their active participation and assistance in the work.
Reference [1] I.S. Polkin, V.N. Samarov, Advance in powder metallurgy of titanium, in: The 12th World Conference on Titanium, June, 2011, Beijing, China Ti-2011.
98
Additive Manufacturing for the Aerospace Industry
Further reading J.D. Cotton, R.D. Boyer, G.R. Weber, K.T. Slattery Technical Fellow, Titanium alloy development needs for commercial airframes, in: Metallurgy Seattle, Washington, 2008. I.S. Polkin, Additive manufacturing of titanium alloys, Technol. Light Alloys 3 (2015) 1116.
Measurement of powder characteristics and quality for additive manufacturing in aerospace alloys
6
Thomas F. Murphy and Christopher T. Schade Hoeganaes Specialty Metal Powders LLC, Cinnaminson, NJ, United States
6.1
Introduction
Titanium, superalloy, stainless steel, and tool steel powders used in the aerospace industry are produced by a variety of methods that influence the particle morphology and hence, their suitability for use in additive manufacturing (AM). These powders can be produced via gas atomization (GA), plasma spherodizing, plasma rotating electrode process (PREP), or electrode induction melting GA (EIGA) [1]. Each of the AM processes such as selective laser melting (SLM), electron beam melting (EBM), or direct energy disposition (DED) requires different powder attributes to optimize their performance [2]. Metal injection molding (MIM) is also used to produce parts for the aerospace industry and requires its own set of powder characteristics to maximize the final part properties. Characterization of metal powders for the aerospace industry (titanium, superalloys, stainless steel, and high strength low alloy steels) involves documenting the average particle size, particle size distribution, surface area, flowability, apparent density, tap density, moisture content, and porosity in a powder. These attributes are, in many cases, considered the minimum required information by the end user. Frequently, the manufacturer of these powders provides a certificate of analysis which contains this information. Additional characterization utilizing metallographic techniques is often provided to allow optimization of the powder for each application. This chapter will review basic powder testing used as quality control tools for powders as well as the advanced metallographic techniques.
6.2
Quality control measurements
6.2.1 Particle size and distribution Particle size and particle size distribution are important characteristics for each batch of powder used for an application. As an example, Fig. 6.1 shows the particle Additive Manufacturing for the Aerospace Industry. DOI: https://doi.org/10.1016/B978-0-12-814062-8.00007-8 © 2019 Elsevier Inc. All rights reserved.
100
Additive Manufacturing for the Aerospace Industry
Cumulative distribution Q3(x)/%
100
1.8
90
1.6
80
1.4
70
1.2
60
1.0
50 0.8
40
0.6
30 20
0.4
10
0.2
0 4
6
8
10
MIM cut 15 lb/h)
Sciaky
m
e nb
Low (S phase transf. microstructure Phase field Rules-based Data-based Part distortion Phase transformations
Figure 17.2 Linkages in a generalized ICME framework. ICME, Integrated computational materials engineering. indicates aspects of a model which are themselves dependent upon the instantaneous condition of the simulated material state (e.g., temperature).
Given that the principal utility of an ICME framework is to enable an engineering decision to be made, it is clear that not every element of this proposed ICME framework is necessary for every problem. For example, if the primary objective of executing an ICME framework is to predict the residual stress of a component, it is possible to arrive at a reasonably accurate prediction by considering the heattransfer and macroscale thermal fields. However, if the primary objective is to predict the end composition (One of the challenges associated with the insertion of additively manufactured aerospace materials into applications is the fact that the composition changes during melt processes. Interstitial content will change, and for some processes and alloys, the fraction of primary alloying elements will also change (e.g., Ti6Al4V processed under vacuum can lose a large amount of its Al content due to preferential vaporization). Thus, the end fabricator may be responsible for certifying part composition.) of a component, other models are required in an ICME framework. For the purposes of this chapter, we will focus on a slightly simpler ICME framework, shown in Fig. 17.3. This framework informs the structure of this chapter. Assuming an input composition, we will first describe how certain physics of the process can be understood and modeled. These physics include: heat flow in both the melt pool and the whole part; fluid dynamics effects; and macroscale effects from uneven heating and cooling on the whole part. Once the process effects are understood and simulated, it is then possible to predict both chemistry and microstructure (influenced by both chemistry and the complex thermal history) within the final part, including an understanding of the nature of solidification and
Table 17.1 Selected details of aspects that might be included in a fusion process model General category of physics in the fusion zone
Details and examples
Energymaterial interactions
Reflection, absorption, surface topographies where reflected energy at point X is absorbed at point Y, pressure/stress from income energy on liquid surfaces, ionization of elemental species in proximity to the interface Volatilization/absorption of certain elements and the attending changes in local stresses and temperatures at the liquidenvironment interface when species leave/ absorb Scattering of incoming energy due to plumes and ionized material All of the physics of melting processes, including mass and thermal driven convection (Marangoni flow), melt pool shape, formation and collapse of keyholes, gravity, buoyancy, and the thermophysical properties that influence these processes (e.g., surface tension, viscosity, density) GL: volatilization/absorption, details controlled by spatial distributions in elemental species and temperatures on both sides of interface, possible quantum effects. GS: gaseous species “plating” of atomic species as they cool and potential solid-state absorption of interstitial elements such as oxygen and nitrogen when the solid is at a very high temperature. LS: solidification/melting Primary heat flow modes (convection, conduction, radiation). Secondary heat flow modes due to surfacemediated processes, such as volatilization/absorption. Effects of heat flow on solidification and subsequent solidsolid phase transformations; part distortion; defect formation
Materialenvironment interactions
Energyenvironment Interactions Melting
Interfacial Interactions between states of matter
Heat flow
Input composition
Process model
Chemistry model Physics-based property model Constitutive equations
Microstructure model Probabilistic modeling (design allowables)
Figure 17.3 Simplified ICME model that has been developed and applied for Ti6Al4V. ICME, Integrated computational materials engineering.
Developing and applying ICME
379
reheating in AM builds. Once chemistry and microstructure are predicted, there have been models developed to predict properties, and subsequently performance, from microstructural and chemistry quantification as it varies within an AM part. We conclude by presenting some gaps in our current understanding, and thus our modeling capabilities.
17.2
Part 1: Process modeling
Our general understanding of the AM process, as it relates to final part properties, can be ascribed to two main regimes. The first relates to the overall thermal history of the final part, as well as uneven cooling rate and thermal histories resulting in residual stresses within the part. The complex localized thermal histories are also discussed, as a result of multiple reheats. This same regime, including the multiple reheats, is responsible for the solid-state phase transformations that lead to complex microstructures. The second regime includes the physics of the melt pool itself, which involves a complex multiphysical environment where heat flow (convection, conduction, and radiation) is coupled with fluid flow, mass transport, resulting in such complex phenomena as Marangoni convection and PlateauRayleigh instabilities that can lead to the formation of defects. The motion of the high-energy, highly localized heat source and relatively low heat conduction of some metals (including titanium, a critical alloy for aerospace components), results in large thermal gradients (G 5 |rT|) during the AM process. The motion of the highly localized heat source results in a complex thermal field that deviates from simple, Cartesian coordinate based thermal fields. Given that most AM processes involve multiple layers, with the heat source passing over previously solidified material, and it is clear that there is complex temporal nature to the thermal history. The theses of Kelly [1] and Ales [2] and the seminal works of the team of Denlinger, Martukanitz, and Michaleris [35] provide important modeling details to capture the thermal history of additively manufacturing, and some of the macrolevel effects (e.g., thermal cycling and thermal distortion). Due to expansion on heating and contraction on cooling, parts produced using AM often are affected by the creation of residual stress fields within each layer and across the entire part. While the exact nature and scale of these residual stresses depends on the part geometry, build rate, heat input, and thermophysical properties of the material, it is possible to draw some conclusions. For aerospace structural alloys, within each layer, cyclic expansion and contraction results in tensile stress at the top of the layer and compressive stress at the bottom. Collins et al. [6] compiled multiple modeling efforts to quantify the residual stresses present. The trend in terms of material type is higher stresses in Ni-based alloys (400800 MPa), lower in Ti alloys (100200 MPa), and even lower in Al alloys (25 MPa) [7]. Consistently, the tensile stresses within each layer were found to be greater than the corresponding compressive stresses. These macroscale residual stresses, which are fundamentally associated with gradients in strain-accommodating defects, such as
380
Additive Manufacturing for the Aerospace Industry
dislocations, result in the large-scale distortion found in finished parts. One observation that has been made is that the size of the melt pool, which corresponds directly to the energy density of the build, was found to correspond directly with the residual stresses. A larger (and therefore hotter) melt pool resulting from slower build rates and/or a larger input power leads to higher thermal gradient. Experimental and modeling confirms that large thermal gradients result in higher residual stress. In one example, Martukanitz et al. [3,8] used data from in situ temperature measurements made during electron beam deposition of a 107 layer part. Data from the melt pool measurements were fed into a sophisticated thermal model, resulting in data that agrees with 3D scan distortion measurements (Fig. 17.4) [3]. In addition to the macroscopic distortion, the high levels of residual stress can result in cracking, layer delamination, or hot tearing either during or after deposition. There are at least two possible origins for these high residual stresses. First, the spatially varying cyclic thermal histories will result in local thermal distortion 60 Distortion (mm)
(A)
Experiment
40
Simulation
20 0
–20 –40 –60 3500
3000
2500
2000
1500
1000
500
0
X-coordinate (mm)
5.00+01
(B)
4.50+01 4.00+01 3.50+01 3.00+01 2.50+01 2.00+01 1.50+01 1.00+01
Z Y X
5.00+01 0
Figure 17.4 (A) Experimental versus computational comparison of part distortion in a large AM build and (B) graphical representation of computational results [3]. AM, Additive manufacturing.
Developing and applying ICME
381
due to the coefficients of thermal expansion. Second, for materials with solidsolid phase transformations, the repeated excursions through the solid phase transformations may lead to crystallographic strains that cannot be rapidly (and repeatedly) accommodated by the matrix crystal phases, resulting in the formation of defects. Interestingly, the formation of large residual stresses in a part may also have further effects on phase transformations not only in systems with strain-induced phase transformations (e.g., some types of martensites), but also in variant selection in alloys such as Ti6Al4V. For the purposes of the simplified example ICME outline shown in Fig. 17.3, we focus on the thermal history and its influence on the resulting chemistry and microstructure. As noted previously, other articles and theses deal very effectively with thermal modeling (e.g., [15,913]). Regarding the effects of the various physics of the molten pool and solidification, some very exciting modeling results have emerged over the past few years, and we present the generalities of this past research briefly. There has been a large body of work focused on the complex physics within the melt pool, especially in recent years [1417]. Regarding these physics, the first consideration is the distribution of the heat within the molten pool, which is often (but not necessarily) modeled as a Gaussian distribution [1820]. This ideal input distribution is likely rarely accurate, owing to scattering of incident photons or electrons by the vapor clouds and both “static” and dynamic powder in both powderblown and powder bed systems [15] (Fig. 17.5A). Further complicating these energy inputs is the fact that the absorptivity (often assumed to be an extrinsic variable) does change with surface temperature and surface topography, and is another physical factor that is considered in the modeling of this process. Knapp et al. [18] used a model to study the effect on the final part of the input power (see Fig. 17.5B), where the model included heat flow via radiation and conduction, mass input, fluid flow, and a moving heat source. In this work (and reassuringly), even excluding loss of mass via vaporization, the geometry of a single pass agrees with experimental results of nominally identical parameters, as seen in Fig. 17.5B. Another important physical process present in all AM systems is the Marangoni Effect. This is a mass flow fluid mechanics effect, where a gradient in surface temperature results in a force. In AM systems, this gradient in surface temperature is due to the presence of a steep thermal gradient. Often, the Marangoni Effect is responsible for the large amount of convection present in the melt pool and thus the primary means of ensuring chemical homogeneity in the melt pool. Models of powder bed fusion process have shown the effect of Marangoni convection on the geometry and motion within the melt pool [16] (see Fig. 17.5C). When the models are executed with a temperature independent, constant surface tension, a nonphysical bulbous melt pool is generated. While this melt pool shape can give some insight into the process [18], accounting for Marangoni allows the modelers to move closer to the reality. Others have demonstrated that accounting for the jet of metal vapor further changes the shape and motion of the pool. The study of this jet of vapor can lead to interesting insights as to otherwise unpredicted effects of the state of the build chamber environment. Matthews et al. [17] studied
0.8
(A)
(B) 14.0
0.4
Height (mm)
0.6 α
Temperature (K) 1000 1693 1733
Total Spheres Substrate
0.2
13.5
12.5 12.0
0.0 500 Distance (μm)
0
50 (mm/s)
13.0
1000 14.0
2.0
1.5 1.0 0.5 Width (mm)
0
Temperature (K)
Height (mm)
1000 1693 1733
13.5 50 (mm/s) 13.0 12.5 12.0
(C)
Homogeneous laser deposition
Laser ray tracing
Liquid
(i) Flat contact
Solid Point contact shadow
40 30 20 10 0 –10 –20 –30 –40
2.0
1.5 1.0 0.5 Width (mm)
Constant surface tension
(ii)
0 Recoil + Marangoni
(iii)
Marangoni effect
(iv)
60 80 100 120 140 160 180 200 220 240 60 80 100 120 140 160 180 200 220 240 60 80 100 120 140 160 180 200 220 240
Figure 17.5 (A) calculated absorptivity (a) for a bimodal powder bed, with the incident beam size given by the circles on the insets [15]; (B) calculated deposit shape, size, with information regarding temperature and fluid flow also provided immediately adjacent to transverse cross sections of as-deposited 316 L stainless steels at different powers (1500 W top and 2500 W bottom) [19]; (C) incremental inclusion of additional physics, showing the increase in model fidelity as heat transfer, melt pool depth, and fluid flow are all mediated by the additional physics that are noted in each sub-image [18].
Developing and applying ICME
383
Figure 17.6 (A) Micrographs and (B) confocal height maps with varying argon pressure [17].
the effect of changing the argon pressure on a powder bed AM system, and found that this had a profound effect on the melt pool and powder available to be melted (Fig. 17.6). Larger positive argon pressure suppressed the negative pressure of the vapor jet and contributed to more metal powder being pulled into the melt track. As the argon pressure was taken to low values, the vapor jet was able to blow metal powder away from the melt track, and resulted in piled-up powder along the track. This vapor jet within the melt pool also leads to another important phenomenom in the AM process, which is the cyclic formation and collapse of a keyhole. With sufficient power input, the metal vapor within the melt pool can form a deep, narrow hole, which cyclically collapses and reforms, and can trap porosity in the build. This is most pronounced in the extremely high input power in electron beam processes, where the keyhole can even result in plasma jets that further complicate the process and any associated modeling effort [21,22]. Modeling of the keyhole has also been useful for predicting porosity in builds, with high speed camera imaging of the keyhole collapse and reform serving to verify the process models further, in both electron beam and laser systems [23]. Thus, porosity can result from both incomplete melting of the feedstock (i.e., when the energy density is too low, and insufficient for to achieve fusion) and the cyclic keyhole formation and collapse (i.e., which the energy is too high and causes elemental vaporization/volatilization). With respect to the subsequent solidification, generally speaking, solidification models can be ascribed to one of four main approaches: (1) process map models; (2) phase field models; (3) cellular automata models; and (4) kinetic Monte Carlo. The process map models typically start with concepts of heat transport and then
384
Additive Manufacturing for the Aerospace Industry
60,000 Δt (0.0153 s) (A)
100,000 Δt (0.0255 s) (B)
135,000 Δt (0.0344 s) (C) 15
Model including fluid flow
10 20 µm
(D)
(E)
(F) 15
Control runs without fluid flow
0
Figure 17.7 Solute profiles in a TixW system for models (A-C) including and (D-F) excluding fluid flow under different time steps in the model (real times in parentheses) [33].
calculate values of the thermal gradient (G) and the velocity of the solidliquid interface (R), enabling the formulation of the so-called GR plots that provide maps of the predominate types of microstructures based upon the operating solidification mechanisms. These approaches have been adopted for both titanium and nickel-based alloys [12,13,2426]. Researchers pursuing phase field modeling for solidification have often coupled the phase field method with finite element modeling to calculate the macroscopic temperature fields [27,28]. Cellular automata has seen increased use (see an example showing the influence of calculated fluid flow in Fig. 17.7), as it is based upon a relatively simple framework where rules evolve over time, and can also capture some of the fundamental aspects of thermodynamics and diffusion [2934]. The regular grid of cellular automata models is also beneficial, as it enables it to be linked with finite element methods that give the thermal gradients. Kinetic Monte Carlo [35,36] is the least well-explored method, but seems to show some promise for accurately predicting grain boundaries in multilayered AM builds.
17.3
Part 2: Predicting chemistry
The composition of structural alloys has a strong influence on not only the microstructure, but also directly on the mechanical properties through mechanisms such as solid-solution strengthening. Indeed, in previous work [37,38], it has been shown that the composition can be attributed to between 70% and 90% of the strength
Developing and applying ICME
385
displayed by Ti6Al4V. As noted previously, the fact that fusion is the basis for current AM processes, it will be necessary to understand and predict chemistry of AM products. The final composition will likely be different from the starting composition, due either to interstitial pickup in atmosphere (including inert atmosphere) or solute loss due to preferential vaporization under vacuum. In addition to these macroscale effects, it is also necessary to consider the partitioning of solute species during solidification. Both composition effects will be considered below.
17.3.1 Solute loss (vaporization) or pickup (gettering) Previously, we have reviewed compositional variations found in many AM parts [6]. Since the AM process is a melting and solidification process, certification of the input material does not guarantee the final part will retain that chemistry. The presence of interstitial elements (e.g., oxygen, nitrogen, hydrogen) in the atmosphere, even in trace amounts (e.g., partial pressures in inert atmosphere), can have a pronounced effect on microstructure and mechanical properties. The use of metal powder feed, with its high surface-to-volume ratio, can exacerbate this problem as interstitial elements often lead to nanometer scaled surface layers, such as oxides or nitrides. Titanium, for example, is extremely susceptible to oxygen pickup, where oxygen serves to increase strength and lower ductility of titanium alloys. The mitigation of these effects can manage in the process environment, where a clean build chamber and careful control of the partial pressures of certain elements within the atmosphere can reduce any chemistry changes due to interstitial pickup. With regard to elemental loss during the process, it has already been established that the melt pool sees extremely high, localized temperatures. For AM techniques that occur under vacuum (e.g., electron-beam-based AM techniques), such high temperatures combine with the low pressures of the build environment and results in preferential vaporization of certain elements [6]. This evaporation of elemental species is related to the partial vapor pressure of the individual elements. The functional form of the partial vapor pressures is nonlinear with respect to temperature (e.g., in Ti6Al4V, Ales [2] has calculated the loss of Al under vacuum using the equation pAl(T) 5 10.91716211/T, where the partial pressure is given in Bayres [39]). Elements with low vapor pressures and melting points relative to the other constituent elements are even more susceptible. While the problem is exacerbated for vacuum-based processes, similar vaporization does occur for all AM processes. One example is electron beam melted Ti64, where aluminum levels can be reduced up to 15% by weight [4042]. Laser systems, such as the Optomec LENS, have also seen reduction in Al content, and feedstock must be adjusted to compensate for this expected loss [43]. Thus, in addition to controlling the composition of the input material (i.e., powder, wire), it is also necessary carefully to control the temperature of the molten pool to achieve desired compositions of as-built parts. Often, the loss or pickup of elements throughout the AM process is simply reported with respect to the starting material, and few studies have tried to model the evaporation or condensation of elements. Given that AM is a nonequilibrium process, more complex modeling than simple partial pressure equations is needed.
386
Additive Manufacturing for the Aerospace Industry
Semiatin [44] used the Langmuir equation to predict the flux at free surface of the melt (Js) for each element i, expressed as Js 5 Xi P0i γ i
rffiffiffiffiffiffiffiffiffiffiffiffi Mi 2πRT
where X is the mole fraction, P is the vapor pressure at absolute temperature T, γ is the activity coefficient for liquid melt, M is the molar mass, and R is the gas constant. Based on Semiatin’s approach, Collins [6] and Ales [2] have also applied the original Langmuir equation [45] to elemental absorption, where the mass flux towards the surface, m, is defined as rffiffiffiffiffiffiffiffiffiffiffiffi Mi m5 pi 2πRT where p is the partial pressure of element i. While the equations for interstitial flux seems to indicated that increasing temperature would lead to less interstitial pickup, increasing temperature means increasing melt pool size, and thus a greater area of flux for the elements, and conducted a simple analysis that shows this approach can be used to predict the gettering of interstitial elements from argon environments of varying purity. At the time of writing this chapter, the authors do note there are other issues that remain to be understood, including composition gradients adjacent to the molten pool. Thus, while there are quite possibly other factors at play, the rate of pickup and absorption of elemental species during the AM process is determined by surfacemediated flux. This also means that the melt pool itself is being sufficiently mixed as to be homogenous, and solute redistribution occurs primarily during solidification. The temperature and size of the melt pool is also an important parameter, and efforts should be made to monitor better and control the melt pool, so that the superheating of the melt pool is minimized [6].
17.3.2 Solidification partitioning The high cooling rates and high velocity of the solidliquid interface in the AM process is consistent with definitions of rapid solidification. As this is a far-fromequilibrium process, there is the potential for significant supersaturation in the solid solution after the liquid-to-solid phase transformation. This can result in solute trapping within the final part, leading to distinctive features unique to the AM process. Aziz [46] presents a basis to understand the conditions under which solute trapping can occur, which can be rewritten for AM and is presented below, with respect to the velocity of the solidification front V, equivalent to the rate of solidification R. V 5R5
1 δT Dliquid c i jrTj δt a0
Developing and applying ICME
387
Peak intensity of AI (a.u.)
60 (A)
(B)
55 50 45 40 35
0
50 100 150 200 250 300 Distance from top surface (µm)
Figure 17.8 Solute partitioning that leads to the so-called fish scaling and the subsequent local ordering (A) micrograph showing composition-driven fish-scaling and (B) composition profile across such bands [47].
In the above equation, T is the temperature, t is time, a0 is the interatomic spacing during growth, and Dliquid is the diffusion coefficient of solute in liquid. When this condition is satisfied, the liquid phase experiences solute trapping, and the solid solubility is extended. As the primary heat loss from the melt pool is via conduction into the as-built part, the highest thermal gradient is usually in the z-direction. This is directly related to the strong texture seen in AM parts, such as h0 0 1i fiber || z-direction. In addition to promoting a strong texture in AM parts, the high thermal gradient promotes solidification that initiates at the solidliquid interface and proceeds along the maximum gradient. The solidification modeling activities, including the cellular automata models developed by Rolchigo [33,34] for various binary titanium alloys, show cellular structures with partitioning between the domains. The solidification also promotes other compositional partitioning, giving rise to features including the so-called fish scaling microstructural features [47] that correspond with geometric aspects of the molten pool, and which are visible in cross section. This solute trapping that can occur in additively manufactured metallic materials has various effects on microstructure and properties, including the formation of nonequilibrium phases. Tomus et al. [48] observed a supersaturation of the Al2Sc system deposited with electron-beam AM process and noted that it corresponded with much higher properties than conventionally manufactured alloys. Thijs et al. [47] studied supersaturation and fish scaling in Ti6Al4V, and found some areas with Al content of up to 25 wt%, resulting in the formation of Ti3Al (Fig. 17.8).
17.4
Part 3: Predicting microstructure
As with any alloy system, chemistry strongly influences the microstructures of alloys produced using AM. However, given that one of the main benefits of AM is
388
Additive Manufacturing for the Aerospace Industry
the ability to produce net- or near-net-shape components (i.e., no subsequent deformation processes to modify the microstructure), AM requires that the fundamentals of solidification be considered to engineer and control the microstructure. As noted previously, additively manufactured materials can exhibit a strong texture and spatial variation in the composition. One such alloy where texture exhibits a pronounced influence on the mechanical properties is Ti6Al4V. It has been shown that the texture can result in differences in the yield strength by at least 5% [38]. Ti6Al4V is one of the most widely used alloys, but is particularly susceptible to forming texture, given its very narrow freezing range. In principal, it should be possible to change chemistry and thus modify the texture by controlling the physics of nucleation and solidification. One such method is the growth restriction factor Q that was introduced by Maxwell et al. [49] and which has been used to describe the effect of solute concentration on the solidification and grain nucleation in cast alloys. Q is effectively a thermodynamic metric given as Q 5 mc0 ðk 2 1Þ 5
dΔTc dfs
where m is the slope of the liquidus line, k is the partition coefficient, and c0 is the solute concentration. In addition, the growth restriction factor is equivalent to the rate of development of constitutional undercooling (Tc) relative to the rate of development of solid (i.e., fraction solid fs). The grain refinement of cast titanium-based alloys with addition of boron was described, using this model, by Tamirisakandala et al. [50]. The observed effect of boron has also been exploited to refine grains and eliminate texture in β-Ti alloys by Mantri et al. [51]. Fig. 17.9 shows electron backscattered diffraction (EBSD) maps and pole figures of the four systems studied, where the addition of trace boron to binary TiV and TiMo systems markedly decreased the grain size and eliminated the texture and large columnar grains. For the Ti12Mo wt%, the addition of 0.5 wt% B resulted in a 100-fold reduction in grain size. In addition to the effects of the growth restriction factor, Mantri et al. attribute the insolubility of boron in titanium as having an effect where the rejection of boron to the solidification front resulted in a constitutionally supercooled front, and a larger frequency of grain nucleation. The effect of solute concentration, while a factor in some systems, does not always provide the full picture. A model was developed by Easton and St. John to describe, semiempirically, the effect of nucleant particles, combined with undercooling and the growth restriction factor, on the final average grain diameter in cast Mg and Al alloys [52,53]. The average grain diameter d is given as d5a1
b Q
Ti–20V
(A) (a)
Build direction
Ti–12Mo
(b)
(B)
(a)
001
101
TD
TD
RD
(b) Titanium-Beta 111
500 μm
(c)
001
Ti–20V–0.5B
101
Ti–20V
RD
Build direction
RD
Ti–12Mo–0.5B
(d)
(c)
Ti–20V–0.5B
001
RD
(d)
Ti–12Mo
001
RD
101
111
TD
RD
RD
111
TD
100 μm
RD
Ti–12Mo–0.5B
RD
1.601 1.457 1.326 1.207 1.099 1.000 0.910 max = 34.541
TD
101
TD
100 μm
TD
RD
TD
max = 1.759
111
TD
TD
500 μm
TD
19.940 10.959 6.023 3.310 1.819 1.000 0.550
RD
101
001
max = 36.280
111
19.140 10.606 5.877 3.257 1.805 1.000 0.554 max = 1.935
TD
1.733 1.553 1.391 1.246 1.116 1.000 0.896
RD
Figure 17.9 (A) EBSD maps and (B) the corresponding pole figures showing considerable grain refinement and reduction of peak textures of beta stabilized titanium alloys upon the incorporation of small amounts of boron [51].
390
Additive Manufacturing for the Aerospace Industry
where a and b are defined in terms of the density of nucleant particles ρ, fraction of activated particles f, a constant b1, and the undercooling required for nucleation ΔTn: a5
1 ðρf Þ1=3
b 5 b1 ΔTn The Easton and St. John model proved useful for describing the effect of grain refinement of silicon on as-cast pure Ti [54]. Mendoza [19] applied the Easton and St. John model to explore the effects of tungsten on the grain refinement of binary titaniumtungsten alloys deposited using a powder-blown, laser-based AM platform. Fig. 17.10 shows the results of these studies, where increasing concentrations of W resulted in a measurable refinement in the resulting grains. While the singular effect of the growth restriction factor had a predominate effect on the grain refinement, the full EastonSt. John model was required to describe fully the effect by considering nucleation effects. Specifically, partially unmelted tungsten particles served as nucleation sites for grains during the rapid cooling of the AM process.
(A) (a)
Ti–5W
(b)
Ti–13W
10
15 % wt Tungsten
(c)
Ti–25W
100 µm
(B) 175 Grain size (µm)
150 125 100 75 50 25 5
20
25
Figure 17.10 (A) SEM micrographs and (B) stereo logically measured values showing the grain refinement of a binary Ti-xW system confirming the Easton-St. John model describing grain refinement during solidification [19].
Developing and applying ICME
391
While these are just a few ways of controlling and modeling microstructure in AM parts via chemical additions, the application of older models developed for cast materials to new AM parts should continue to be explored. As with any new process, new alloy chemistries are needed fully to exploit the benefits of AM. Beyond these solidification microstructures, it is necessary to predict the solidstate phase transformations as well [3]. The types of modeling approaches that might be considered in predicting the solid-state phase transformations include: (1) classical methods, including the JohnsonMehlAvramiKolmogorov and Sestak and Berggren equations [5565]; (2) the phase field method [6669]; (3) and rules-based or database governed predictions that are based upon either modeling or experimental data.
17.5
Part 4: Predicting properties and performance
One of the main the goals of an ICME framework to understanding the AM process is to predict the mechanical properties, such as yield strength, toughness, or fatigue of a component based upon the previous elements in the ICME framework. We have previously accomplished this by rigorously quantifying [70] microstructural features, and using a hybrid modeling approach that integrated two machine learning tools, namely artificial neural networks (ANNs) with genetic algorithms (GAs) [37,38,41,71]. The ANN uses highly flexible functions (e.g., the hyperbolic tangent function) to establish the interrelationships that exist among a set of rigorously quantified input variables (e.g., composition, microstructure) and a set of output variables (e.g., yield strength, fracture toughness, fatigue [7275]). Virtual experiments allow the trained ANNs to be explored in systematic ways. For example, by holding all of the input parameters except one at a single value, such as their average value for a given dataset, and then allowing that single input to vary over its range, it is possible to see the dependency of a property on that single input. This type of experiment is likely to be possible in a laboratory, given the complex and interdependent ways in which composition and microstructural features influence properties. The GA can then be exercised on the same dataset with a postulated equation where physically relevant terms are present. The GA then optimizes the equations, solving unknown weights and exponential powers, and once the equation is optimized, it can be compared to the ANN through the use of the same virtual experiments. The use of this approach has led to the deduction of a constitutive equation to predict strength of the material. A model that was previously developed for wrought structures was applied to additively manufactured Ti6Al4V. Over the course of our most recent work on large-scale AM of Ti6Al4V structures [38], we have made a few important discoveries. First, texture has a pronounced influence on the mechanical properties. Second, once texture is accounted for, the equation for wrought Ti6Al4V structures is nominally identical to additively manufactured Ti6Al4V. Third, the equations can be applied to any AM component, independent of subsequent heat
Predicted yield strength (MPa)
(A)
(B)
Yield strength-EBAM Ti-6AI–4v predicted vs experimental
1000 950
Pred. YS - α+β stress relief Pred. YS - α+β HIP Pred. YS - β anneal 0% +5% –5%
Fvα·89 + Fvβ·45+ 0.667 0.5 0.7 0.5 2 Fαv ·(149·xAI + 759·xo0.667) + Fvβ·((22·x0.7 v ) + (235·xFe ) ) +
900
σys =
850
–0.5 Fcol ·(tβ–nb)0.5+ v ·150·(tα–lath) –0.5 Fcol + v ·125·(tcolony)
800
(–1)·(AxisDebit) +
750
FBW v ·αMGb ρ
700 700 750 800 850 900 950 1000
Measured yield strength (MPa)
Probability
(C)
0.999 0.990 0.900 0.750 0.500 0.250 0.100 0.050
σYS, yield stress Phenomenological model Monte Carlo simulation
0.010 0.005
+/–σMSE +/–2σMSE
0.001 650
700
750
800
850 900 950 1000
Stress σ(MPa)
Figure 17.11 (A) Prediction of the properties of as-deposited Ti6Al4V subjected to three heat treatments [38]; (B) the equation corresponding equation [38]; and (C) a representation of a similar equation for the cumulative probability distribution of Ti6Al4V [76].
Developing and applying ICME
393
treatments, as long as the composition and microstructure can be measured or predicted accurately. Finally (and importantly), there is an unexpectedly large dislocation density in certain heat treatments (i.e., those that keep most of the as-deposited material state). The demonstration of a single equation to predict the strength of additively manufactured Ti6Al4V subjected to three different heat treatments is shown in Fig. 17.11A and B, with the predicted versus experimental data shown in Fig. 17.11A, and the equation presented in Fig. 17.11B, where the volume fractions of phases, size of features, and chemistry are included [38]. Continuing this demonstration of the ICME framework for additively manufactured Ti6Al4V, it is necessary to transition the prediction of properties to the prediction of performance. One way to interpret performance is through the use of cumulative probability distribution functions (pdfs), where the probability of achieving any given value of a property (e.g., yield strength) is plotted (see Fig. 17.11C; [76]). If the models described previously were to describe perfectly the properties of the material, the cumulative pdfs of the predicted data would overlap perfectly with experimentally measured data. However, even for good models, when looking at the data on a probability plot in which the tails of the data are emphasized, it may be that the model does not accurately predict the experimentally observed data. Thus, we have successfully turned to using a “distributiontranslationrotation” approach to shift and skew the models as necessary to represent the physical data. This statistical refinement of the model can occur in statistical space, and does not change the model. This approach has been previously reported in the literature [76], and additional publications will be forthcoming in the near future.
17.6
Limitations
Throughout the remainder of this chapter, we have referred to successful demonstrations of modeling activities, and/or have provided some important fundamental details to enable others to develop modules to an ICME approach. However, despite the successful outcomes that these activities have demonstrated, there are still limitations that need to be considered. Arguably, the most important limitations are associated with what we either do not know or have difficulty measuring/computing. Three limitations will be discussed briefly. The first limitation is associated with the fact that the process is quite complex, and multiple physics are active, potentially “erased” in the previous layer, and reactivated. In addition, many of the important physics associated with phase transformations, defect formation, chemistry, or texture formation occurs at different regions (e.g., solid-state phase transformations occur below the molten pool; nucleation events, convection, conduction, and instabilities occur within or at the surface of the molten pool; chemistry changes at the surface and above the molten pool; and solidification/solid-state phase transformations occur behind the molten pool) are characterized by details that are at either the nanosecond or nanometer scales. Currently, the AM community lacks the tools that may permit the investigation of
–60
3400 –100 (A)
(B)
(C) –55 –50
3300
–45
–80
–40
–60
–35
3200
–30 –25
–40
–20
3100
–15 –10
–20 3000
0
10
20
30
40
50
60
(D)
20 2900 40 2800
60 80
2700
100 –30 –20 –10 0 10 20 30
2600
Figure 17.12 (A) One direction SAW velocity map of electron-beam as-deposited Ti6Al4V; (B) the orientation map deduced from multiple SAW velocity maps; (C) the high resolution inset from (B); and (D) a tiled, mosaic optical image from the same region as (C). All units in (A) and (C) are in millimeters.
Developing and applying ICME
395
these physics at the appropriate length and time scales to understand what is happening. However, very recent investments in programs to develop in situ AM cells in synchrotron beamlines should allow the community to discover new science to understand better AM processes. The second limitation is associated exclusively with defects. In some systems, including titanium-based alloys, inspectability of additively manufactured components is a challenge. Spatial variation in the anisotropy of the as-deposited microstructures can interact with nondestructive evaluation techniques, providing new challenges when identifying defects, and potentially influencing the probability of detection (i.e., pod). The companion chapter in this volume speaks specifically to the challenges associated with nondestructive evaluation of additively manufactured articles. Interestingly, while there are challenges, the line-by-line, layer-by-layer of AM may permit the measurement of local microstructural state (including defects), and allow the so-called digital twins to be created for each component built. The third limitation that we will discuss briefly is the variation in scale of microstructural inhomogeneity. In traditional processing, the multistep forging sequences can chemically homogenize the material, and predictably produce texture that is relatively spatially consistent throughout a part. While texture can be controlled in AM [77], most AM processes have relatively small melt pools (e.g., ,1 mm). There are large-scale AM processes where the size of the molten pool permits heattransport mechanisms to compete and produce spatially varying microstructures, including texture (Fig. 17.12). As is apparent in the figure, the scales of these microstructural domains are much greater than what the materials scientist would typically measure. To measure the texture of these domains, the authors have turned to adopting a new technique (spatially resolved acoustic spectroscopy, or SRAS) to analyze the local orientation. During SRAS, a laser passes through a grating and sets up a surface acoustic wave (SAW), the velocity of which can be accurately determined. The velocity of the SAW is related to the elastic stiffness tensor (cij). If the SAW velocity and elastic stiffness tensor are known, the orientation can be calculated [7881]. This method is especially exciting as it permits the measurement of texture and local orientation over areas that far exceed what is typically measured using other techniques. The limitation is the spatial resolution, which is currently B25 μm.
17.7
Summary
It is possible to develop and execute an ICME framework for structural metallic materials for aerospace applications. Any ICME framework is based upon a series of decisions that depend upon the overall objective. There are some exciting modeling activities that can be integrated into an ICME framework, including activities that predict distortion and solidification microstructures based upon a thermal history. One of the important aspects of any ICME framework is the ability to predict the composition of the as-deposited material, as it will invariably deviate from the
396
Additive Manufacturing for the Aerospace Industry
precursor powder or wire. The Langmuir equation has been used to predict elemental loss under vacuum and solute pickup under atmosphere. However, there are still unknown details of both the molten pool shape and the material transfer physics that will influence the successful implementation of the Langmuir (or similar) approach. Once chemistry is known, it is possible to couple chemistry with cooling rate and predict microstructure. Given a specific microstructure and composition, it is currently possible to predict the yield strength of the widely used aerospace alloy Ti6Al4V. Given the knowledge base that exists for certain aluminum-based alloys and nickel-based superalloys, it should be possible to integrate that knowledge into an ICME framework and make preliminary predications regarding their properties. Once a constitutive equation for properties is known, it is possible to predict the performance of the material, as represented by design allowables and the cumulative probability distribution function. While such modeling capabilities have been demonstrated, there are still gaps in the AM knowledge base. These gaps will be reflected in any ICME framework. However, there is extensive work underway to fill these knowledge gaps, and over the next decade, significant progress is expected.
References [1] S.M. Kelly, Thermal and Microstructure Modeling of Metal Deposition Processes With Application to Ti6Al4V (Doctoral dissertation), Virginia Tech, 2004. [2] T.K. Ales, An Integrated Model for the Probabilistic Prediction of Yield Strength in Electron-Beam Additively Manufactured Ti6Al4V (Masters dissertation), Iowa State University, 2018. [3] R. Martukanitz, P. Michaleris, T. Palmer, T. DebRoy, Z.-K. Liu, R. Otis, et al., Toward an integrated computational system for describing the additive manufacturing process for metallic materials, Addit. Manuf. 1 (2014) 5263. [4] E.R. Denlinger, J. Irwin, P. Michaleris, Thermomechanical modeling of additive manufacturing large parts, J. Manuf. Sci. Eng. 136 (6) (2014) 061007. [5] P. Witherell, S. Feng, T.W. Simpson, D.B. Saint John, P. Michaleris, Z.-K. Liu, et al., Toward metamodels for composable and reusable additive manufacturing process models, J. Manuf. Sci. Eng. 136 (6) (2014) 061025. [6] P.C. Collins, D.A. Brice, P. Samimi, I. Ghamarian, H.L. Fraser, Microstructural control of additively manufactured metallic materials, Annu. Rev. Mater. Res. 46 (2016) 6391. [7] L. Sochalski-Kolbus, E. Payzant, P. Cornwell, T. Watkins, S. Babu, et al., Comparison of residual stresses in Inconel 718 simple parts made by electron beam melting and direct laser metal sintering, Metall. Mater. Trans. A 46 (2015) 14191432. [8] E.R. Denlinger, C.H. Jarred, P. Michaleris, Residual stress and distortion modeling of electron beam direct manufacturing Ti6Al4V, Proc. Inst. Mech. Eng., B: J. Eng. Manuf. 229 (10) (2015) 18031813. [9] S.M. Kelly, S.L. Kampe, Microstructural evolution in laser-deposited multilayer Ti6Al4V builds: Part II. Thermal modeling, Metall. Mater. Trans. A 35 (6) (2004) 18691879.
Developing and applying ICME
397
[10] S. Bontha, W.K. Nathan, P.A. Kobryn, H.L. Fraser, Effects of process variables and size-scale on solidification microstructure in beam-based fabrication of bulky 3D structures, Mater. Sci. Eng., A 513 (2009) 311318. [11] S. Bontha, W.K. Nathan, P.A. Kobryn, H.L. Fraser, Thermal process maps for predicting solidification microstructure in laser fabrication of thin-wall structures, J. Mater. Process. Technol. 178 (13) (2006) 135142. [12] P.A. Kobryn, S.L. Semiatin, Microstructure and texture evolution during solidification processing of Ti6Al4V, J. Mater. Process. Technol. 135 (23) (2003) 330339. [13] N.W. Klingbeil, S. Bontha, C.J. Brown, D.R. Gaddam, P.A. Kobryn, H.L. Fraser, et al., Effects of Process Variables and Size Scale on Solidification Microstructure in LaserBased Solid Freeform Fabrication of Ti6Al4V, Wright State University, Dayton, OH, 2004. [14] W. King, A.T. Anderson, R.M. Ferencz, N.E. Hodge, C. Kamath, S.A. Khairallah, Overview of modelling and simulation of metal powder bed fusion process at Lawrence Livermore National Laboratory, Mater. Sci. Technol. 31 (8) (2015) 957968. [15] W.E. King, A.T. Anderson, R.M. Ferencz, N.E. Hodge, C. Kamath, S.A. Khairallah, et al., Laser powder bed fusion additive manufacturing of metals; physics, computational, and materials challenges, Appl. Phys. Rev. 2 (4) (2015) 041304. [16] S.A. Khairallah, T.A. Andrew, A. Rubenchik, W.E. King, Laser powder-bed fusion additive manufacturing: physics of complex melt flow and formation mechanisms of pores, spatter, and denudation zones, Acta Mater. 108 (2016) 3645. [17] M.J. Matthews, G. Guss, S.A. Khairallah, A.M. Rubenchik, P.J. Depond, W.E. King, Denudation of metal powder layers in laser powder bed fusion processes, Acta Mater. 114 (2016) 3342. [18] G.L. Knapp, T. Mukherjee, J.S. Zuback, H.L. Wei, T.A. Palmer, A. De, et al., Building blocks for a digital twin of additive manufacturing, Acta Mater. 135 (2017) 390399. [19] M.Y. Mendoza, P. Samimi, D.A. Brice, B.W. Martin, M.R. Rolchigo, R. LeSar, et al., Microstructures and grain refinement of additive-manufactured TixW alloys, Metall. Mater. Trans. A 48 (7) (2017) 35943605. [20] A. Raghavan, H.L. Wei, T.A. Palmer, T. DebRoy, Heat transfer and fluid flow in additive manufacturing, J. Laser Appl. 25 (5) (2013) 052006. [21] Y.H. Tian, C.Q. Wang, D.Y. Zhu, Modeling of the temperature field during electron beam welding of aluminum alloy by a pre-defined keyhole, Key Engineering Materials, vol. 353, Trans Tech Publications, Zurich, Switzerland, 2007, pp. 20112014. [22] D.N. Trushnikov, G.M. Mladenov, Numerical model of the plasma formation at electron beam welding, J. Appl. Phys. 117 (1) (2015) 013301. [23] R. Rai, J.W. Elmer, T.A. Palmer, T. DebRoy, Heat transfer and fluid flow during keyhole mode laser welding of tantalum, Ti6Al4V, 304L stainless steel and vanadium, J. Phys. D: Appl. Phys. 40 (18) (2007) 5753. [24] J. Gockel, L. Sheridan, S.P. Narra, N.W. Klingbeil, J. Beuth, Trends in solidification grain size and morphology for additive manufacturing of Ti6Al4V, JOM 69 (12) (2017) 27062710. [25] N. Raghavan, R. Dehoff, S. Pannala, S. Simunovic, M. Kirka, J. Turner, et al., Numerical modeling of heat-transfer and the influence of process parameters on tailoring the grain morphology of IN718 in electron beam additive manufacturing, Acta Mater. 112 (2016) 303314. [26] N. Raghavan, S. Simunovic, R. Dehoff, A. Plotkowski, J. Turner, M. Kirka, et al., Localized melt-scan strategy for site specific control of grain size and primary dendrite arm spacing in electron beam additive manufacturing, Acta Mater. 140 (2017) 375387.
398
Additive Manufacturing for the Aerospace Industry
[27] T. Keller, G. Lindwall, S. Ghosh, L. Ma, B.M. Lane, F. Zhang, et al., Application of finite element, phase-field, and CALPHAD-based methods to additive manufacturing of Ni-based superalloys, Acta Mater. 139 (2017) 244253. [28] L.-X. Lu, N. Sridhar, Y.-W. Zhang, Phase field simulation of powder bed-based additive manufacturing, Acta Mater. 144 (2018) 801809. [29] A. Rai, M. Markl, C. Ko¨rner, A coupled cellular automatonlattice Boltzmann model for grain structure simulation during additive manufacturing, Comput. Mater. Sci. 124 (2016) 3748. [30] A. Zinoviev, O. Zinovieva, V. Ploshikhin, V. Romanova, R. Balokhonov, Evolution of grain structure during laser additive manufacturing. Simulation by a cellular automata method, Mater. Des. 106 (2016) 321329. [31] O. Zinovieva, A. Zinoviev, V. Ploshikhin, Three-dimensional modeling of the microstructure evolution during metal additive manufacturing, Comput. Mater. Sci. 141 (2018) 207220. [32] W. Tan, X. Li, Numerical modeling of grain growth in laser engineered net shaping (LENS) of AISI 316 stainless steel, ASME 2017 12th International Manufacturing Science and Engineering Conference Collocated With the JSME/ASME 2017 6th International Conference on Materials and Processing, American Society of Mechanical Engineers, New York, NY, 2017, pp. V002T01A004. [33] M.R. Rolchigo, Y.M. Michael, P. Samimi, D.A. Brice, B. Martin, P.C. Collins, et al., Modeling of TiW solidification microstructures under additive manufacturing conditions, Metall. Mater. Trans. A 48 (7) (2017) 36063622. [34] M.R. Rolchigo, R. LeSar, Modeling of binary alloy solidification under conditions representative of Additive Manufacturing, Comput. Mater. Sci. 150 (2018) 535545. [35] T.M. Rodgers, A.M. John, V. Tikare, A Monte Carlo model for 3D grain evolution during welding, Modell. Simul. Mater. Sci. Eng. 25 (6) (2017) 064006. [36] H.L. Wei, J.W. Elmer, T. DebRoy, Three-dimensional modeling of grain structure evolution during welding of an aluminum alloy, Acta Mater. 126 (2017) 413425. [37] I. Ghamarian, P., Samimi, V. Dixit, P.C. Collins, A constitutive equation relating composition and microstructure to properties in Ti6Al4V: as derived using a novel integrated computational approach, Metall. Mater. Trans. A 46 (11) (2015) 50215037. [38] B.J. Hayes, W.M. Brian, B. Welk, S.J. Kuhr, T.K. Ales, D.A. Brice, et al., Predicting tensile properties of Ti6Al4V produced via directed energy deposition, Acta Mater. 133 (2017) 120133. [39] C.B. Alcock, V.P. Itkin, M.P. Horrigan, Vapour pressure of the metallic elements, Can. Metall. Q. 23 (3) (1984) 309313. [40] R. Banerjee, S. Nag, H.L. Fraser, A novel combinatorial approach to the development of beta titanium alloys for orthopaedic implants, Mater. Sci. Eng. C25 (3) (2005) 282289. [41] P.C. Collins, C.V. Haden, I. Ghamarian, B.J. Hayes, T. Ales, G. Penso, et al., Progress toward an integration of processstructurepropertyperformance models for “threedimensional (3-D) printing” of titanium alloys, JOM 66 (7) (2014) 12991309. [42] S.M. Gaytan, L.E. Murr, F. Medina, E. Martinez, M.I. Lopez, R.B. Wicker, Advanced metal powder based manufacturing of complex components by electron beam melting, Mater. Technol. 24 (3) (2009) 180190. [43] B. Peterson, A Combinatorial Approach to the Development of a Creep Resistant Beta Titanium Alloy, The Ohio State University, Columbus, OH, 2008. [44] S.L. Semiatin, V.G. Ivanchenko, O.M. Ivasishin, Diffusion models for evaporation losses during electron-beam melting of alpha/beta-titanium alloys, Metall. Mater. Trans. B 35 (2) (2004) 235245.
Developing and applying ICME
399
[45] Irving Langmuir, The vapor pressure of metallic tungsten, Phys. Rev. 2 (5) (1913) 329. [46] M. Aziz, Model for solute redistribution during rapid solidification, J. Appl. Phys. 53 (1982) 11581168. [47] L. Thijs, F. Verhaeghe, T. Craeghs, J. Van Humbeeck, J.-P. Kruth, A study of the microstructural evolution during selective laser melting of Ti6Al4V, Acta Mater. 58 (9) (2010) 33033312. [48] D. Tomus, M. Qian, C.A. Brice, B.C. Muddle, Electron beam processing of Al2Sc alloy for enhanced precipitation hardening, Scr. Mater. 63 (2) (2010) 151154. [49] I. Maxwell, A. Hellawell, A simple model for grain refinement during solidification, Acta Metall. 23 (2) (1975) 229237. [50] S. Tamirisakandala, R.B. Bhat, J.S. Tiley, D.B. Miracle, Grain refinement of cast titanium alloys via trace boron addition, Scr. Mater. 53 (12) (2005) 14211426. [51] S.A. Mantri, T. Alam, D. Choudhuri, C.J. Yannetta, C.V. Mikler, P.C. Collins, et al., The effect of boron on the grain size and texture in additively manufactured β-Ti alloys, J. Mater. Sci. 52 (20) (2017) 1245512466. [52] D.H. StJohn, M.A. Qian, M.A. Easton, P. Cao, Z. Hildebrand, Grain refinement of magnesium alloys, Metall. Mater. Trans. A 36 (7) (2005) 16691679. [53] D.H. StJohn, M.A. Easton, M. Qian, J.A. Taylor, Grain refinement of magnesium alloys: a review of recent research, theoretical developments, and their application, Metall. Mater. Trans. A 44 (7) (2013) 29352949. [54] M.J. Bermingham, S.D. McDonald, K. Nogita, D.H. St John, M.S. Dargusch, Effects of boron on microstructure in cast titanium alloys, Scr. Mater. 59 (5) (2008) 538541. [55] M. Avrami, Kinetics of phase change. I General theory, J. Chem. Phys. 7 (12) (1939) 11031112. [56] M. Avrami, Kinetics of phase change. II transformation-time relations for random distribution of nuclei, J. Chem. Phys. 8 (2) (1940) 212224. [57] M. Avrami, Granulation, phase change, and microstructure kinetics of phase change. III, J. Chem. Phys. 9 (2) (1941) 177184. [58] A.N. Kolmogorov, On the statistical theory of the crystallization of metals, Bull. Acad. Sci. USSR, Math. Ser. 1 (1937) 355359. [59] J. William, R. Mehl, Reaction kinetics in processes of nucleation and growth, Trans. Metall. Soc. AIME 135 (1939) 416442. [60] E.G. Prout, Fo. C. Tompkins, The thermal decomposition of potassium permanganate, Trans. Faraday Soc. 40 (1944) 488498. [61] A. Celli, D.Z. Edgar, I. Avramov, Primary crystal nucleation and growth regime transition in isotactic polypropylene, J. Macromol. Sci., B 42 (2) (2003) 387401. [62] J. Sesta´k, Modeling of reaction mechanism: use of Euclidian and fractal geometry, Chapt. 10, Science of Heat and Thermos Physical Studies: A Generalized Approach to Thermal Analysis., Elsevier, Amsterdam, 2005, pp. 276314. ˇ ˇ ´ kBerggren equation, Thermochim. Acta 520 (12) [63] P. Simon, Forty years of the Sesta (2011) 156157. ˇ ´ kBerggren equation a general expression of kinetic [64] J. Ma´lek, M.C. Josee´, Is the Sesta models? Thermochim. Acta 175 (2) (1991) 305309. ˇ ´ k, Generalized kinetics of overall phase transition in terms of logis[65] I. Avramov, J. Sesta tic equation. arXiv preprint arXiv:1510.02250 (2015). [66] T. Li, M. Ahmed, G. Sha, R. Shi, G. Casillas, H.-W. Yen, et al., The influence of partitioning on the growth of intragranular α in near-β Ti alloys, J. Alloys Compd. 643 (2015) 212222.
400
Additive Manufacturing for the Aerospace Industry
[67] R. Shi, Y. Wang, Variant selection during α precipitation in Ti6Al4V under the influence of local stress—a simulation study, Acta Mater. 61 (16) (2013) 60066024. [68] S. Balachandran, A. Kashiwar, A. Choudhury, D. Banerjee, R. Shi, Y. Wang, On variant distribution and coarsening behavior of the α phase in a metastable β titanium alloy, Acta Mater. 106 (2016) 374387. [69] C. Shen, N. Zhou, Y. Wang, Phase field modeling of microstructural evolution in solids: effect of coupling among different extended defects, Metall. Mater. Trans. A 39 (7) (2008) 16301637. [70] P.C. Collins, B. Welk, T. Searles, J. Tiley, J.C. Russ, H.L. Fraser, Development of methods for the quantification of microstructural features in α 1 β-processed α/β titanium alloys, Mater. Sci. Eng., A 508 (12) (2009) 174182. [71] I. Ghamarian, B. Hayes, P. Samimi, B.A. Welk, H.L. Fraser, P.C. Collins, Developing a phenomenological equation to predict yield strength from composition and microstructure in β processed Ti6Al4V, Mater. Sci. Eng., A 660 (2016) 172180. [72] P.C. Collins, S. Koduri, B. Welk, J. Tiley, H.L. Fraser, Neural networks relating alloy composition, microstructure, and tensile properties of α/β-processed TIMETAL6-4, Metall. Mater. Trans. A 44 (3) (2013) 14411453. [73] S. Kar, T. Searles, E. Lee, G.B. Viswanathan, H.L. Fraser, J. Tiley, et al., Modeling the tensile properties in β-processed α/β Ti alloys, Metall. Mater. Trans. A 37 (3) (2006) 559566. [74] P.C. Collins, S. Koduri, V. Dixit, H.L. Fraser, Understanding the interdependencies between composition, microstructure, and continuum variables and their influence on the fracture toughness of α/β-processed Ti6Al4V, Metall. Mater. Trans. A 49 (3) (2018) 848863. [75] J.W. Foltz, B. Welk, P.C. Collins, H.L. Fraser, J.C. Williams, Formation of grain boundary α in β Ti alloys: its role in deformation and fracture behavior of these alloys, Metall. Mater. Trans. A 42 (3) (2011) 645650. [76] C.V. Haden, P.C. Collins, D.G. Harlow, Yield strength prediction of titanium alloys, JOM 67 (6) (2015) 13571361. [77] R.R. Dehoff, M.M. Kirka, W.J. Sames, H. Bilheux, A.S. Tremsin, L.E. Lowe, et al., Site specific control of crystallographic grain orientation through electron beam additive manufacturing, Mater. Sci. Technol. 31 (8) (2015) 931938. [78] R. Patel, W. Li, R.J. Smith, S.D. Sharples, M. Clark, Orientation imaging of macrosized polysilicon grains on wafers using spatially resolved acoustic spectroscopy, Scr. Mater. 140 (2017) 6770. [79] W. Li, J. Coulson, P. Marrow, R.J. Smith, S.J. Laine´, M. Clark, et al., Crystallographic orientation determination of hexagonal structure crystals by laser ultrasonic technique, J. Phys. Conf. Ser. 684 (1) (2016) 012001. [80] R.J. Smith, W. Li, J. Coulson, M. Clark, M.G. Somekh, S.D. Sharples, Spatially resolved acoustic spectroscopy for rapid imaging of material microstructure and grain orientation, Meas. Sci. Technol. 25 (5) (2014) 055902. [81] W. Li, J. Coulson, J.W. Aveson, R.J. Smith, M. Clark, M.G. Somekh, et al., Orientation characterisation of aerospace materials by spatially resolved acoustic spectroscopy, J. Phys. Conf. Ser. 520 (1) (2014) 012017.
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition
18
Lucas W. Koester1, Leonard J. Bond1,2,3, Hossein Taheri1,2 and Peter C. Collins4 1 Center for Nondestructive Evaluation, Applied Sciences Complex II, Iowa State University, Ames, IA, United States, 2Department of Mechanical Engineering, Iowa State University, Ames, IA, United States, 3Department of Aerospace Engineering, Iowa State University, Ames, IA, United States, 4Department of Materials Science and Engineering, Iowa State University, Ames, IA, United States
18.1
Introduction
Additive manufacturing (AM) is a rapidly emerging technology consisting of the joining of materials to make parts from 3D model data, usually layer upon layer, as opposed to subtractive and formative manufacturing methodologies [1]. The ability to construct complex components rapidly with little material waste is poised to have transformative effects on many industries. However, qualification of novel manufacturing processes and methodologies is a common slowing point before wider application can be realized. AM methods in the aerospace industry have been used for production of less-than-critical components, including cockpit and fuselage interiors for a number of years. The flexibility and enabling design capability are, however, driving wider application to critical structures for repair, replacement, and new part production.
18.1.1 Additive manufacturing components in service A number of components have now been produced by AM and flown by various aerospace entities. The first Federal Aviation Administration (FAA)-approved engine component to fly in commercial jets was in 2015 with the T25 housing for a sensor to monitor compressor inlet pressure and temperature [2]. The FAA approved the use of the housing in the GE90 engine after identification of the opportunity to utilize AM based on value. A process from proper material and machine identification to component and subsequently engine level certification was then formulated. Since then, the framework has been followed for the use of various AM components in a number of engines, including a novel fuel nozzle that enabled weight savings of 25% and reduced the number of brazes and welds from 25 to 5 in the Leap engine [3]. Additive Manufacturing for the Aerospace Industry. DOI: https://doi.org/10.1016/B978-0-12-814062-8.00020-0 © 2019 Elsevier Inc. All rights reserved.
402
Additive Manufacturing for the Aerospace Industry
Predating the applications for aviation systems, several space applications had driven the development of AM for some critical applications. Lockheed Martin utilized AM for fabrication of titanium waveguide brackets that have now traveled billions of miles and currently reside on the Juno spacecraft as they orbit Jupiter. Since the maiden spaceflight of an AM component, Lockheed Martin has announced its redesigned A2100 geosynchronous satellite launched in 2017 utilized AM for 10% of its components. The National Aeronautics and Space Administration (NASA) has also identified the potential for a diverse range of AM parts for use in space systems. In looking further to the future, the capability to produce reliable, complex components in space on demand has obvious implications for manned space travel. NASA recently produced the first fused deposition modeling 3D-printed part in the microgravity environment of the International Space Station as a proof of concept [4]. While the applications and design freedom enabled by AM are exciting, particularly for the aerospace community, companies and agencies are proceeding with caution in deploying AM components for mission-critical applications. In looking back at the history of the development of traditional manufacturing methods, the building of a sound experience and knowledge base developed with time. Such a process of gradual incremental change and maturation is being accelerated with AM and presents some significant challenges. That said, the development and deployment of AM-fabricated parts is continuing to proceed at an unprecedented pace. Furthermore, the complexity of AM processes and the dependence of the performance of the resulting material on process conditions have yet to be fully understood, which introduces the potential for significant variation in performance for AM components. Quantitative characterization techniques are required for finished components to comply with existing inspection standards and procedures for identifying defects, verifying microstructure, determining final part geometry, and assessing material property variation in critical components. The potential variability and flexibility in process conditions motivates development and deployment of rapid, quantitative process monitoring and characterization in situ to fully realize the capability of AM systems and the parts that they can produce. Whether post-production or during fabrication, AM is a challenging area for both process monitoring and nondestructive testing.
18.1.2 Regulatory actions and standardization The FAA has submitted an AM Strategic Roadmap to establish plans and practices for businesses dealing with the development and application of AM technologies as of September 2017 [5]. The document will likely follow the examples set by other organizations, such as the Department of Defense (DOD) and the Additive Manufacturing Standardization Committee (AMSC), in which the American Society for Testing and Materials (ASTM) has a lead role [6,7]. Both have established roadmaps to accelerate standards and specifications for the rapid development in the AM industry. The roadmaps examine the state of the industry, identify existing standards in development, and make recommendations for priority areas
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition
403
where gaps have been identified. A general structure has been produced that acknowledges commonality among AM methods, but realizes the need for specific standards and test methods based on raw materials, process, and end applications considered (Fig. 18.1). ASTM established Committee F42 on AM Technologies in 2009 [8]. The International Organization for Standardization (ISO) formed ISO/TC 261 for similar purposes in 2011 [9]. After a meeting in 2013, the two groups formed a Joint Plan for Additive Manufacturing Standards Development to expedite standards development across the two organizations and avoid duplication of effort. The group has since produced and adopted several joint standards, including AM terminology and standard guides. Numerous Work Items are underway to address specific AM methods and materials including metals and polymers [8]. Areas of concern identified by these working groups and agencies include process monitoring to enable closed-loop feedback and control, reliable nondestructive evaluation (NDE), and quality control, among others. The design of the standardization structure surrounding a process or material can take several forms. For example, Metallic Material Properties Development and Standardization, often used to
Additive manufacturing standards structure Terminology
General AM standards
Design guides
Test methods
Test artifacts
Feedstock materials
Process / Equipment
Material category-specific
Process category-specific
Metal powders Ceramic powders Photopolymer resins Metal rods
Polymer powders
Polymer filaments
etc.
System performance and reliability
Qualification guidance
Data formats
Safety
Material jetting
Powder bed fusion
Binder jetting
Directed energy deposition
Material extrusion
Sheet lamination
Vat photopolymerization
Nylon powder ABS filament
Steel rods Nickel-based alloy powders etc.
Application-material-specific Aerospace Automotive
Inspection methods
Medical etc.
Powder bed fusion with nylon
Material extrusion with ABS
Directed energy deposition with titanium alloy
Powder bed fusion with steel
Automotive
Medical etc.
General top-level AM standards • General concepts • Common requirements • Generally applicable
All finished parts Mechanical test methods NDE/NDT methods
Post-processing methods
Bio-compatibility test methods Chemical test methods
etc.
Category AM standards Specific to material category or process category
Material-specific Titanium alloy Nylon
ABS
Paper
Sand
Aluminum alloy
Nickel-based alloy
etc.
etc.
Application-processmaterial-specific Aerospace
etc.
Finished parts
Process-material-specific
Material-specific Titanium alloy powders
Round robin test protocols
Specialized AM standards Specific to material, process, or application
Application-material-specific Aerospace Automotive
Medical etc.
Figure 18.1 Standards Structure approved by ASTM F42 and ISO TC261. [2017 American National Standards Institute/National Center for Defense Manufacturing and Machining, operating America Makes—the National Additive Manufacturing Innovation Institute (America Makes)]. ASTM, American Society for Testing and Materials; ISO, International Organization for Standardization.
404
Additive Manufacturing for the Aerospace Industry
derive design allowables statistically from experimentally determined statistical distributions of material properties, is not straightforward to apply to AM materials due to the breadth of processes and materials that AM encompasses. Thus, the focus is placed on characterizing and “freezing” a process in an acceptable state, followed by validation using “witness” specimens produced under identical conditions for validation. This is evidenced by recent guidance produced by NASA for metallic spaceflight hardware produced by laser powder bed fusion (LPBF) [10]. A key component of this quality control strategy is quantitative nondestructive evaluation of critical components for build errors that may induce defects detrimental to material performance. Thus, application of existing nondestructive evaluation techniques or modification to address issues specific to AM components is a necessary and likely part specific for critical applications. Part geometric complexity, microstructure, and surface condition all impact nondestructive evaluation techniques and need to be addressed to ensure reliable defect detection.
18.1.3 Material properties and defects in additive manufacturing Applications of AM that seek to supplant traditional manufacturing methods due to advantages in cost, performance, or lead-time must still meet material property and characterization requirements for the end application, regardless of the manufacturing method. For aerospace components, attempts at weight reduction can lead to complex internal and external geometries that complicate inspection. Defects tend to form between layers and generally do not propagate far in the build direction leading to crack-like defects perpendicular to the build direction. Furthermore, surface roughness tends to be high (on the order of or larger than casting processes) due to a weld like material consolidation process. A coverage map of NDE methods over the spatial resolutions and target depths within the material is given in Fig. 18.2. Complex external/internal geometry and roughness have compelled the use of penetrating radiation as the most likely method for successful inspection. However, internal defects may or may not be void, which negatively impact contrast in x-ray and ultrasonic methods [11]. For large components, penetration issues can limit the inspectable volume of the material quickly with radiographic and ultrasonic techniques. Surface roughness affects the reconstruction process in x-ray computed tomography, particularly for near-surface defects that are of particular concern for fatigue considerations. Roughness also negatively impacts defect detection capabilities in ultrasonic inspection in most cases [12]. Multiple layers of internal structures or internal lattice structures are not inspectable with ultrasonic techniques. Postproduction surface finishing can help mitigate the effects of roughness on inspectability, but can induce compressive surface residual stresses that may make inspection for surface breaking defects more difficult [13]. The resulting, as-built microstructure in metal AM also tends to follow the melt pool solidification path and thermal gradients [14]. Seeding from powder particles and previous melted layers can cause complex microstructure that complicates ultrasonic inspections for defects. Post-processing of materials by heat treatment
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition
405
St
an
da
X-
Location of interest
in
ith
w e ep um De vol
ra y
rd
,m
CT
ini
,m
icr
o-
Co
m
p m lex Ult ple ge ras ou ge om nd om etr y et Th ry erm al Volume/surface e th Eddy defects r e a c current Ne urfa s Si
At ce rfa su
Non-interferometric optics surface breaking
,n
an
o-
fo c
us
Volumetric defects
White light interferometry very near surface
Optical μm
mm
nm
Resolution Figure 18.2 Resolution coverage map for various NDE techniques with applicability to AM for geometric defects. NDE penetration limits the application of techniques that are diffusive and optical methods are limited to surface interrogation. AM, Additive manufacturing; CT, Computed tomography.
can form microstructures more amenable to inspection, but removes the capability for tailoring of local material properties during building that may be achieved by additives to feedstock powder that has been demonstrated for certain materials [15,16]. Such post-processing also adds cost to components that may or may not meet the criteria for deployment. A number of geometric internal defects may also form during fabrication. Microporosity is linked primarily with porosity entrapped in starting-powders. These defects are generally higher contrast, but can be small (necessarily smaller than starting powder diameters), and thus difficult to detect. Larger scale porosity tends to be formed by improper heat source characteristics that lead to entrapped gasses due to melt pool dynamics or keyhole collapse. Post-processing with hot isostatic pressing has shown some capability to consolidate these defects below a certain size, though the defects may reemerge after additional heat treatment, particularly for materials deposited in inert as opposed to vacuum environments [17]. Lack of fusion (LOF) type defects are unique to AM and are primarily caused by poor processing parameter selection. The defects may be filled with unconsolidated powder that reduces x-ray CT contrast in post-production inspections with large voxel sizes. The presence of unmelted powders may be resolved if sufficient resolution can be achieved by using μ-CT techniques [18]. Resonance type testing can be a method of detecting LOF defects, but is binary in its decision and
406
Additive Manufacturing for the Aerospace Industry
nonspecific, relying on a statistical basis for rejection of a component based on shifts in resonant frequencies [19]. Balling is a phenomenon that causes powder particles to become entrapped within the melt pool, leading to the formation of small spheres on the order of the heat source spatial extent. These unstable melt pool dynamics can lead to higher roughness, which negatively impacts fatigue and inspectability. These balled regions may also be removed during powder spreading, increasing the chance of producing LOF and layer defects [20].
18.2
State of the art
Nondestructive evaluation plays a key role in AM for the aerospace industry. The production of components with AM is often most attractive for low-production runs and high-value alloy materials. Qualification strategies primarily consist of determining an operating window, “freezing” the process to produce components and witness test pieces, and post-production inspection. Establishing process windows for new AM methods or materials could prove costly and time consuming, however. Thus, numerous agencies have identified the need for enhanced in situ sensing and characterization techniques. The DOD AM Roadmap identified the need for development in NDE and process control technologies to enable consistent processing and verification of part quality [6]. Monitoring and in situ characterization for closed-loop control is predicated on the existence of nondestructive techniques that are reliable, rapid, and quantitative. Similar views were expressed by the AMSC, acknowledging the low technology readiness level of in situ process monitoring for machine state and material condition [7].
18.2.1 Optical and thermal monitoring Optical and thermal emissions monitoring have received much attention due to the relative ease with which they can be applied. Line-of-sight access is available in most commercially available machines on which most research has been performed. Methods to monitor optical emissions can include the use of photodiodes, highresolution cameras, and high-speed cameras. Point-based measurement systems, such as photodiodes and pyrometers, often require system integration to track the melt pool during deposition. In contrast, full-field techniques monitor optical and thermal emissions for a large area, potentially continuously during a build. The accumulation of this data in a tomographic assembly over the entirety of the build after processing can then be likened to a quality map, as shown in Fig. 18.3. Optical monitoring of melt pool spatter has shown to be linked to different energy density conditions of the heat source for 18Ni (300) maraging steel powder [22]. Monitoring of balling in the melt pool has also been linked with energy density in fabrication of stainless steel [23]. Photodiode and camera optical signatures can also be spatially correlated to establish process maps of thermal accumulation
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition
407
Figure 18.3 Beginning with (A) CAD geometry, the model is (B) discretized according to the spatial resolution of the inspection method. Layer-by-layer inspection data is (C) postprocessed and stored according to position and (D) assembled upon build completion to produce a tomographic image [21], CC BY 4.0. CAD, Comparison to design.
due to geometry effects [24]. A photodiode-high speed camera hybrid system that monitored melt pool dimensions and average thermal emission from the melt pool has been shown to be effective in reducing over-melting and gas porosity. The method was patented and licensed to Concept Laser, but requires integration with the system optics [25]. In summary, the majority of optically based techniques infer part quality from optical emissions and melt pool stability and dynamics. Direct measurement of defects using optical methods must be at or very near the surface [using optical coherence tomography (OCT)]. Imaging of surface breaking defects is difficult due to low contrast between consolidated and unconsolidated regions, but has shown some promise with multiple viewing angles and lighting conditions [26]. Optical determination of geometric accuracy would involve high-resolution optical imaging at each layer and comparison to design files (or other) for geometric accuracy. Thermal monitoring allows for a means to monitor heat accumulation within the component, potentially over the entire build area when using thermal cameras. However, obtaining absolute temperatures is difficult due to varying emissivity values at the surface, depending upon the state of the material (e.g., molten, solidified, or powder). Thus, early investigations examined variation in thermal characteristics qualitatively. Pavlov et al. varied hatch spacing (distance between subsequent material deposition passes) and layer height, utilizing a two-wavelength pyrometer to examine variations in pyrometer signal levels on in a selective lasermelting process. Small hatch spacing that allowed sufficient contact between laser passes showed heat accumulation with subsequent passes and higher pyrometer signal levels. As hatch spacing increased, a transition region was observed between complete contact and no contact between deposited lines. Varying the height of the deposition also showed increasing pyrometer signal levels caused by eventual loss of contact with the substrate. They also observed drops in pyrometer signal level where powder spreading was insufficient and proposed the instrument as a process monitoring tool. In a similar manner, balling was monitored optically and linked with pyrometer signal analysis for variations of energy density in two laser-based AM systems, and an operating window that minimized balling was identified [23]. In an effort to quantify thermographic measurements, Rodriguez et al. [27] proposed additional calibration and instrumentation to account for variables that affect
408
Additive Manufacturing for the Aerospace Industry
the thermographic imaging process, including the effect of the viewing window, mean radiant temperature within the environment, and surface emissivity. Their results showed that thermocouples and algorithms to predict surface temperature (a parameter used to adjust beam current to mitigate thermal accumulation) likely underestimate surface temperature when compared with corrected thermal imaging data. As mentioned previously, quantitative data such as temperature evolution during manufacturing can be related to microstructural evolution and resulting part material properties, eventually leading to microstructural control. This potential was partially demonstrated by Raplee et al. with thermographic data corrected for surface emissivity. The resulting thermal gradients and estimated solidliquid interface velocity could aid in predicting areas with equiaxed or columnar type grain regions and (Fig. 18.4), varying melting strategies [28]. Given the influence of melt pool dynamics, thermal history, and dimensions on resulting part geometry and microstructure, melt pool monitoring in fine detail has also been investigated as a monitor and potential control on the resulting part properties. Fox et al. investigated coaxially aligned near infrared imaging of a novel, (A) Layer temperature 1200
1100
100
1050
Y (pixels)
20 μm
150 1000 200
950
250
900
300
850 800
350
750
50 100 150 200 250 300 350 400 450
X (pixels)
20 μm
New temperature data @ 70 mm build height pixel x = 200, y = 150 point melt
(B)
1150
1100
1050
1000
950
1200
1150
1100
1050
1000
950 0
5
10
15
20
25
30
Time (s)
35
40
45
50
New temperature data @ 70 mm build height pixel x = 200, y = 200 line melt
(C) Apparent temperature (ºC)
Apparent temperature (ºC)
1200
Temperature (ºC)
1150
50
0
5
10
15
20
25
30
35
40
45
Time (s)
Figure 18.4 The relationship of (A) corrected thermal imaging and (B and C) thermal history and resulting microstructure in Inconel 718 fabricated with electron beam additive manufacturing with different melt strategies [28], CC BY 4.0.
50
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition
409
instrumented laser-based AM test bed (Additive Manufacturing Metrology Testbed). Infrared image collection at 30,000 frames per second can be used to extract melt pool dimensions with image analysis [29]. However, even with such acquisition rates, blurring, high dynamic temperature ranges, and relatively low resolution of the detector (12 μm/pixel) remain obstacles and significant work remains in this promising area.
18.2.2 Post-production inspection Geometric and material aspects of AM components pose challenges for reliable NDE. These challenges include roughness, anisotropy of material properties, complex internal and external geometry, and the nature of the defects encountered. The surface condition of as-built components most closely resembles castings, Surfacebased inspection techniques, primarily optical and penetrant methods, are negatively impacted by surface roughness [13]. Existing standards for inspection of rough casting components will likely be applicable with some modification to AM components [30]. Similar concerns exist for surface condition improvement and the effects on defect detection, including inducing compressive residual stresses caused by peening (commonly used to remove loosely attached surface powder particles) that may close surface cracks, or the masking of defects by smeared material after subtractive machining. Many of the aspects of inspection have been addressed by other works and a complete discussion is outside the scope of this article [13,31].
18.2.3 Emerging methods Due to the limitations of surface-based inspection techniques, volumetric techniques are being investigated for in situ characterization and inspection. Volumetric techniques may afford the capability to inspect at less frequent intervals by interrogating multiple layers at a time. Furthermore, defects that may not be surface breaking may also be indicated. These techniques include primarily ultrasonic and acoustic methods as well as novel, system-integrated melt pool imaging, as discussed previously [32]. Laser ultrasonics has been identified as a potential method to monitor AM builds for surface breaking defects, sub-surface defects, and microstructure by a number of authors [33]. The method has been shown to be capable of identifying larger artificial defects (produced via electrical discharge machining) in a direct energy deposition (DED)-produced Inconel alloy [34,35]. The investigation of porosity in situ is also under investigation by incorporating laser ultrasonic excitation and detection into LPBF optics [36]. The mapping of surfacewave acoustic velocity can also be used for determining information about microstructure and defects [37]. Spatially resolved acoustic spectroscopy (SRAS) has been demonstrated ex situ for AM metals, mapping acoustic wave speeds with resolutions as low as 20 μm can infer grain structure by abrupt changes in wave speed associated with neighboring crystallites with different crystallographic orientations [38]. The experimental configuration and an example result are shown in Fig. 18.5, comparing optical metallography with the wave-speed mapping result. Wave-speed values for pores
410
Additive Manufacturing for the Aerospace Industry
Detection laser (532 nm)
Generation laser (1064 nm)
(A)
1 0.8 0.6 0.4
Optical train
Detector
0.2 0
Optical spot
Generation patch
Surface acoustic wave
(B)
ms–1 2950
2900
∅p οs
2850
2800
Figure 18.5 SRAS experimental configuration (left) showing the generation patch and resulting surface acoustic wave. Correlation between images (right) produced by (A) optical imaging and (B) the acoustic wave speed map can been seen for pores [dark in (A)] and additional wave-speed variations caused by microstructural crystallite orientations [39], CC BY 4.0, assembled from two figures. SRAS, Spatially resolved acoustic spectroscopy.
are missing due to the absence of waves for detection in the voids [39]. Scanning in multiple directions can then determine crystallographic orientation. The detection capability for defects has also been recently demonstrated ex situ, to inform a rework or repair strategy demonstrated on polished samples [40]. A major impediment to laser ultrasonic investigation of AM materials in situ is surface roughness and condition. Surface roughness highly attenuates surface waves and complicates reliable detection of surface displacements. Methods to ameliorate this restriction include development of new sensors and detectors that can correct for the effects of surface roughness with relative ease. Speckled knife edge detectors have been developed that utilize an array of sensors to determine corrections to speckle introduced by surface roughness using a sensor array and comparing the data from adjacent sensors [41]. While capable, laser ultrasonics hardware tends to be relatively expensive when compared with full-field visualization techniques. Thus, incorporation of such equipment into commercial machines will need to demonstrate value added, potentially through in situ repair or enabling customizable spatial microstructures.
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition
411
Acoustic monitoring has also garnered consideration given that stress waves propagate through the bulk of the material and can be used to infer or measure directly material properties and defects [42]. Gaja and Liou [43] observed good correlations between high energy acoustic events collected with an acoustic emissions transducer attached directly to the build plate and defect formation including cracks and porosity. Acoustic monitoring can also detect signatures that are potentially characteristic to the fabrication process, such as powder impacts during DED, laser generated ultrasound near/inside the melt pool, and machine vibrations and noises. Wasmer et al. [44] monitored acoustic noise in a noncontact manner with a fiber Bragg grating and, based on wavelet analysis and a trained convolution neural network, were able to classify build condition based on acoustic noise in powder bed fusion AM. Acoustic methods collect information nonselectively provided there is sufficient acoustic power in the bandwidth of the transducers used. Acoustic monitoring metrics can be separated to allow independent examination of indications from crack-like events and characteristic process noise, given the relatively short duration of such high-energy events. An example of this method is given in Fig. 18.6, in which acoustic metrics can be used as indicators of damage and process state in directed energy deposition of titanium 6Al-4V on tool steel. Root-mean-squared (RMS) noise levels vary dramatically during the build as well as a sharp decrease in noise levels after build completion seen in Fig. 18.6A. Also depicted is the total count of high amplitude events (referred to as “Hits”) that may be used as a defect density indicator. After eliminating high-amplitude events often associated with defect formation, including cracks and porosity, isolation of low-amplitude noise results in a noise level that has been shown to be unique for this process under various build conditions (Fig. 18.7). Contact acoustic emissions sensors were attached
Figure 18.6 Acoustic monitoring of a directed energy deposition system can be (A) an indicator of material damage from tracking high-amplitude events associated with defects and (B) a passive process monitor by measuring RMS noise levels and comparing with known “good” levels [45]. RMS, Root-mean-squared.
412
Additive Manufacturing for the Aerospace Industry
(A)
× 10–5
6
Baseline Powder only Normal Low laser power Low powder feed
Normal standard dev. (σ)
5
4
3
2
1
0 0.5
0
2
1.5
1
2.5
Normal mean (μ)
× 10–4
(B) Clustering all data and silhouette mean values using spectral features Cf, CA, and PA in one experiment for each build condition–high frequency band
20 15
1150
fo ur ie H (k
qu en
cy
of
0
C
0.5 nsform (CA)
fre
1200
1 rier tra
ro id
1.5 litude of Fou
id amp
z)
1250
2 Centro
rt
1300
0 2.5
sf or
5
ra n
1350
m
(C
f)
1400
10
en t
Peak amplitude of Fourier transform (PA)
BL: Base Line C1: Optimum settings C2: Low power C3: Low powder feed CO: Powder feed only
Figure 18.7 Clustering of acoustic metrics derived from temporal samples of process noise can be used to classify machine states (varied around normal with baseline tests) in (A) the temporal metrics including RMS noise level central tendencies and deviations and (B) spectral characteristics, including centroid locations and amplitudes [4547]. RMS, Rootmean-squared.
to the build plate and noise levels (RMS) were recorded during deposition under different build conditions, including Normal (100% Laser Power, 100% Powder Feed), Low Laser Power (78% Laser Power), Low Powder feed (50% Powder Feed), and Powder Only (No Laser Power). Baseline conditions were recorded when the machine was prepared for deposition, but otherwise at rest to determine a lower noise threshold.
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition
413
Apart from the potential capability as a process monitor for AM equipment including pumps, motors, heat sources, and material condition, others have raised concern that such a monitoring technique is a bridge between physical and cyber systems that may enable bad actors to do harm by intellectual property theft [48]. The potential of acoustic monitoring has been identified by General Electric evidenced by patents of acoustic monitoring methods [42,49]. Work remains to be done to provide quantitative, physics-based predictions of noise levels from machine/material sources and the direct relationship between noise levels or high amplitude events and the resulting material state need to be further elucidated.
18.3
Practical considerations
The drive to monitor and characterize AM materials in situ has led to great technical achievements toward this end. However, once the measurement science has matured to identify and characterize the material properties and defects of interest, the implication on added cost and inspection time must also be addressed. This added cost is illustrated in Fig. 18.8 for two schemes of NDE data collection during building for LPBF [21]. Hirsch et al. investigated the applicability of two NDE techniques: SRAS and OCT, a method requiring translucency/transparency primarily applicable to polymers, based on implications of added build time and potential spatial resolution of the methods. The method provided a penalty factor on the build time that quantifies the estimated added time required for inspection at every layer. SRAS was found to have a temporal penalty of 59 and 1442 for coarse and fine scans (50 and 20 μm minimum detectable defect), respectively. Thus, optimization, inspection at intervals greater than one layer, or other means likely need to be formulated to make the inspections palatable to end-users. OCT systems simulated had a temporal penalty of 2.4 and 5.4 for a 30 μm minimum detectable defect [21]. (A) In-situ Energy beam interaction time, tbuild
NDE scan time, tscan
NDE latency time, tlatency
Recoating/Resetting for next ALM layer, treset
Total time per layer
(B) Online Energy beam interaction time, tbuild
NDE scan time, tscan
Recoating/Resetting for next ALM layer, treset
NDE latency time, tlatency
Total time per layer
Figure 18.8 Potential inspection schemes for layer additive manufacturing (A) in situ and (B) online, suggesting that parallel material consolidation and NDE inspection can enable inspection at every layer [21], CC BY 4.0.
414
Additive Manufacturing for the Aerospace Industry
The added cost of these inspections is reflected in machine productivity, additional hardware associated with the inspections, and calibration, upkeep, and validation of the inspection systems. For well controlled and calibrated AM processes and materials, it is anticipated that the approach of “freezing” a process and the production of “witness” coupons on validated systems will remain the favorable method. However, the production of prototype components and multiple design iteration processes, particularly with experimental alloys or processes that are not thoroughly well characterized, could benefit greatly from layer-by-layer inspection, but will be slow to be adopted until inspection times are reduced to a minimum.
Acknowledgments This article was funded as part of an Industry-University Core Project by the Center for NDE (CNDE), Iowa State University. Thanks go to Quad City Manufacturing Lab (QCML) which has provided access to and operation of the Direct Energy Deposition (DED) system for generation of the experimental data used in Figs. 18.6 and 18.7.
References [1] ASTM ISO/ASTM52900-15 Standard Terminology for Additive Manufacturing— General Principles—Terminology. ASTM International, West Conshohocken, PA, 2015, https://doi.org/10.1520/ISOASTM52900-15. [2] T. Kellner, The FAA cleared the first 3D Printed part to fly in a commercial jet engine from GE, 2015. GE reports. Available from: ,https://www.ge.com/reports/post/ 116402870270/the-faa-cleared-the-first-3d-printed-part-to-fly-2/.. [3] T. Kellner, Fit to print: new plant will assemble world’s first passenger jet engine with 3D Printed fuel nozzles, next-gen materials. GE reports. (Jun 23, 2014). Available from: ,https://www.ge.com/reports/post/80701924024/fit-to-print/.. [4] T.J. Prater et al., Summary report on phase I results from the 3D printing in zero-G technology demonstration mission, vol. I, NASA/TP-2016-219101, A.L. Hunstville (Ed.), National Aeronautics and Space Administration—Marshall Space Flight Center, Huntsville, AL, 2016. [5] D. Werner, FAA prepares guidance for wave of 3D-printed aerospace parts, Space News, October 20, 2017. Available from: ,http://spacenews.com/faa-prepares-guidancefor-wave-of-3d-printed-aerospace-parts/.. [6] J. Fielding, A. Davis, B. Bouffard, M. Kinsella, T. Delgado, J. Wilczynski, Department of defense additive manufacturing roadmap, America Makes, #88ABW-2016-5841, 2016. [7] America Makes & ANSI Additive Manufacturing Standardization Collaborative (AMSC), Standardization Roadmap for Additive Manufacturing, Version 1.0, USA, ANSI and NDCMM/America Makes, Print, 2017. Available from: , https://share.ansi.org/Shared% 20Documents/Standards%20Activities/AMSC/AMSC_Roadmap_February_2017.pdf . . [8] ASTM, Committee F42 on Additive Manufacturing Technologies, Am. Soc. Testing & Materials (ASTM), 2017. Available from: ,https://www.astm.org/COMMITTEE/F42. htm..
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition
415
[9] International Organization for Standards, ISO/TC 261 Additive Manufacturing, International Organization for Standards, 2011. Available from: ,https://www.iso.org/ committee/629086.html.. [10] NASA Technical Standards System, MFSC-STD-3716—Standard for Additively Manufactured Spaceflight Hardware by Laser Powder Bed Fusion in Metals, NASA Technical Standards System, 2017. Available from: ,https://standards.nasa.gov/standard/msfc/msfc-std-3716.. [11] H. Taheri, M.R. Mohammad Shoaib, L. Koester, T.A. Bigelow, P.C. Collins, L.J. Bond, Powder based additive manufacturing—a review of types of defects, generation mechanisms, detection, property evaluation and metrology, Int. J. Addit. Subtractive Mater. Manuf. 1 (2) (2017) 172209. [12] P.B. Nagy, L. Adler, J.H. Rose, Effects of acoustic scattering at rough surfaces on the sensitivity of ultrasonic inspection, in: D.O. Thompson, D.E. Chimenti (Eds.), Review of Progress in Quantitative Nondestructive. Evaluation, Vol. 12B, Plenum, New York, 1993. [13] M. Seifi, A. Salem, J. Beuth, O. Harrysson, J.J. Lewandowski, Overview of materials qualification needs for metal additive manufacturing, JOM 68 (3) (2016) 747764. [14] A.A. Antonysamy, J. Meyer, P.B. Prangnell, Effect of build geometry on the β-grain structure and texture in additive manufacture of Ti6Al4V by selective electron beam melting, Mater. Charact. 84 (Supplement C) (2013) 153168. [15] P.C. Collins, D.A. Brice, P. Samimi, I. Ghamarian, H.L. Fraser, Microstructural control of additively manufactured metallic materials, Annu. Rev. Mater. Res. 46 (1) (2016) 6391. [16] S.A. Mantri, et al., The effect of boron on the grain size and texture in additively manufactured β-Ti alloys, J. Mater. Sci. 52 (20) (2017) 1245512466. [17] S. Tammas-Williams, P.J. Withers, I. Todd, P.B. Prangnell, The effectiveness of hot isostatic pressing for closing porosity in titanium parts manufactured by selective electron beam melting, Metall. Mater. Trans. A 47 (5) (2016) 18. [18] J.A. Slotwinski, E.J. Garboczi, Porosity of additive manufacturing parts for process monitoring, AIP Conf. Proc. 1581 (2014) 11971204. [19] G.R. Stultz, R.W. Bono, M.I. Schiefer, Fundamentals of resonant acoustic method NDT, Adv. Powder. Metall. Part. Mater. 3 (2005) 111. Available from: ,http://www. modalshop.com/techlibrary/FundamentalsofResonantAcousticMethodNDT.pdf.. [20] K. Mumtaz, N. Hopkinson, Top surface and side roughness of Inconel 625 parts processed using selective laser melting, Rapid Prototyping J. 15 (2) (2009) 96103. [21] M. Hirsch, et al., Assessing the capability of in-situ nondestructive analysis during layer based additive manufacture, Addit. Manuf. 13 (2017) 135142. [22] G. Repossini, V. Laguzza, M. Grasso, B.M. Colosimo, On the use of spatter signature for in-situ monitoring of Laser Powder Bed Fusion, Addit. Manuf 16 (2017) 3548. [23] M. Islam, T. Purtonen, H. Piili, A. Salminen, O. Nyrhila, O. Nyrhil¨a, Temperature profile and imaging analysis of laser additive manufacturing of stainless steel, Lasers Manuf. 41 (2013) 835842. [24] T. Craeghs, S. Clijsters, J.-P. Kruth, F. Bechmann, M.-C. Ebert, Detection of process failures in Layerwise Laser Melting with optical process monitoring, Phys. Procedia 39 (2012) 753759. [25] S. Berumen, F. Bechmann, S. Lindner, J.-P. Kruth, T. Craeghs, Quality control of laserand powder bed-based Additive Manufacturing (AM) technologies, Phys. Procedia 5 (2010) 617622. [26] O. Holzmond, X. Li, In situ real time defect detection of 3D printed parts, Addit. Manuf. 17 (2017) 135142.
416
Additive Manufacturing for the Aerospace Industry
[27] E. Rodriguez, J. Mireles, C.A. Terrazas, D. Espalin, M.A. Perez, R.B. Wicker, Approximation of absolute surface temperature measurements of powder bed fusion additive manufacturing technology using in situ infrared thermography, Addit. Manuf. 5 (2015) 3139. [28] J. Raplee, et al., Thermographic microstructure monitoring in electron beam additive manufacturing, Sci. Rep. 7 (2017) 43554. [29] J.C. Fox, B.M. Lane, H. Yeung, Measurement of process dynamics through coaxially aligned high speed near-infrared imaging in laser powder bed fusion additive manufacturing, Thermosense: Thermal Infrared Applications XXXIX, vol. 10214, International Society for Optics and Photonics, 2017. [30] J. Slotwinski, S. Moylan, Applicability of existing materials testing standards for additive manufacturing materials, in: Report # NISTIP 8005, US Department of Commerce, National Institute of Standards and Technology, 2014. [31] L.W. Koester, L.J. Bond, P.C. Collins, H. Taheri, T.A. Bigelow, Metals handbook, Volume 17: Nondestructive evaluation and quality control, Metals Handbook, Volume 17: Nondestructive Evaluation of Materials, ASM International, Metals Park, OH, 2018. [32] L.W. Koester, H. Taheri, T.A. Bigelow, P.C. Collins, L.J. Bond, Nondestructive testing for metal parts fabricated using powder-based additive manufacturing, Mater. Eval. 76 (4) (2018) 386396. [33] S. Everton, P. Dickens, C. Tuck, B. Dutton, Evaluation of laser ultrasonic testing for inspection of metal additive manufacturing, in: Proc. SPIE Laser 3D Manuf. II, 9353, 2015, 935316. [34] SP Santospirito, R. Łopatka, D. Cerniglia, K. Słyk, B. Luo, D. Panggabean, J. Rudlin, Defect detection in laser powder deposition components by laser thermography and laser ultrasonic inspections, Proc. SPIE 8611, Frontiers in Ultrafast Optics: Biomedical, Scientific, and Industrial Applications XIII, 86111N, 15 March 2013. [35] D. Cerniglia, N. Montinaro, Defect detection in additively manufactured components: laser ultrasound and laser thermography comparison, Procedia Struct. Integr. 8 (2018) 154162. [36] T.A. Bigelow, National Science Foundation—Standard Grant, Award Number 1661146. Ultrasound Based In-line Assessment of Porosity for Laser-Sintered Parts, Iowa State University, 2017. Available from: , https://www.nsf.gov/awardsearch/ showAward?AWD_ID 5 1661146 . . [37] S.D. Sharples, M. Clark, M.G. Somekh, Spatially resolved acoustic spectroscopy for fast noncontact imaging of material microstructure, Opt. Express 14 (22) (2006) 1043510440. [38] R.J. Smith, W. Li, J. Coulson, M. Clark, M.G. Somekh, S.D. Sharples, Spatially resolved acoustic spectroscopy for rapid imaging of material microstructure and grain orientation, Meas. Sci. Technol. 25 (5) (2014) 55902. [39] R.J. Smith, M. Hirsch, R. Patel, W. Li, A.T. Clare, S.D. Sharples, Spatially resolved acoustic spectroscopy for selective laser melting, J. Mater. Process. Technol. 236 (2016) 93102. [40] M. Hirsch, et al., Targeted rework strategies for powder bed additive manufacture, Addit. Manuf. 19 (2018) 127133. [41] S.D. Sharples, R.A. Light, S.O. Achamfuo-Yeboah, M. Clark, M.G. Somekh, The SKED: speckle knife edge detector, J. Phys. Conf. Ser. 520 (1) (2014) 12004. [42] T.G. Spears, S.A. Gold, In-process sensing in selective laser melting (SLM) additive manufacturing, Integr. Mater. Manuf. Innovations 5 (1) (2016) 2.
Nondestructive evaluation of additively manufactured metallic parts: in situ and post deposition
417
[43] H. Gaja, F. Liou, Defects monitoring of laser metal deposition using acoustic emission sensor, Int. J. Adv. Manuf. Technol. 90 (14) (2017) 561574. [44] K. Wasmer, C. Kenel, C. Leinenbach, S.A. Shevchik, In situ and real-time monitoring of powder-bed AM by combining acoustic emission and artificial intelligence, in: Industrializing Additive Manufacturing—Proceedings of Additive Manufacturing in Products and Applications—AMPA2017, Springer International Publishing, Cham, 2018, pp. 200209. [45] L.W. Koester, H. Taheri, L.J. Bond, T.A. Bigelow, E. Faierson, Acoustic emissions monitoring of directed energy deposition additive manufacturing: temporal characteristics and data clustering for process monitoring, Addit. Manuf. (2018). submitted. [46] H. Taheri, L.W. Koester, T.A. Bigelow, E. Faierson, L.J. Bond, In-situ process monitoring of additive manufacturing using clustering of spectral features for acoustic signals, J. Manuf. Sci. Eng (2018). Submitted. [47] H. Taheri, L.W. Koester, T.A. Bigelow, E. Faierson, L.J. Bond, In-situ additive manufacturing process monitoring with an acoustic technique: clustering performance evaluation using K-means algorithm, J. Manuf. Sci. Eng (2018). Submitted. [48] M.A. Al Faruque, S.R. Chhetri, A. Canedo, J. Wan, Acoustic side-channel attacks on additive manufacturing systems, in: Proceedings of the 7th International Conference on Cyber-Physical Systems, 2016, pp. 19:119:10. [49] S.A. Gold, T.G. Spears, Acoustic monitoring method for additive manufacturing processes. US Patent Application US20170146488, 2017.
This page intentionally left blank
Combining additive manufacturing with conventional casting and reduced density materials to greatly reduce the weight of airplane components such as passenger seat frames
19
Francis Froes Light Metals Industry, Tacoma, WA, United States
Autodesk research scientist Andreas Bastian, at the company’s Pier 9 technology center in San Francisco has “bridged the gap” between 3D printing and conventional metal casting [1], he has created a lightweight frame for an airplane seat (Fig. 19.1A), which could not only reduce carbon emissions, but also save airlines a lot of component weight and money through the associated fuel saving (Fig. 19.1B). The mass distribution in Seats is shown in Fig. 19.2 [2]. Autodesk has used additive manufacturing (AM) and generative design to help airliners reduce carbon emissions, fuel consumption, and weight before: In 2015, the company collaborated with Airbus to create a 3D printed airplane cabin structural component. Once the dividing bionic partition, currently undergoing Federal Aviation Administration (FAA) testing, is deployed across the production backlog of Airbus A320 jets, it’s estimated to reduce carbon emissions drastically: equal to removing 96,000 cars from the road. An algorithm in Autodesk’s Netfabb software was used to produce the geometry for Bastian’s aircraft seat frame, which would work in any standard commercial jet. The goal was to keep the frame as strong as the original, but make it much more lightweight; using lattice and surface optimization, the software was able to design a complex structure that will make the aircraft seats so lightweight that the need for jet fuel is significantly reduced. However, conventional manufacturing methods would not be able to create the complex geometries that are often used with 3D printing technology, and the cost of 3D printing at scale is not yet at a point where it is competitive with traditional production methods, such as casting. While there are several metals used for 3D printing, the casting process can be completed with thousands of metals and composites. Also, even using Autodesk Project Escher technology [1], 3D printing volumes are typically just a few cubic feet, while casting can work with huge objects. So Bastian combined the two Additive Manufacturing for the Aerospace Industry. DOI: https://doi.org/10.1016/B978-0-12-814062-8.00021-2 © 2019 Elsevier Inc. All rights reserved.
420
Additive Manufacturing for the Aerospace Industry
Figure 19.1 (A) Airline passenger seats. (B) Andreas Bastian with airplane seat frame.
technologies: positive molds for the seat frames, containing the lattice geometry, were 3D printed in plastic in order to save money and time, and were next used to make affordable, ceramic casting molds (Fig. 19.3) by basically using the “lost wax” process (Fig. 19.4). Examples of the final complex seat assembles are shown in Figs. 19.1, 19.5 and 19.6. “While additive manufacturing holds great promise for the future of manufacturing, it’s still very new for many product developers,” said Bastian [1]. “Casting, by contrast, has been around for millennia and is incredibly well understood. There are hundreds of thousands of engineers, foundries, and factories with deep expertise in it. That’s one of the reasons I am looking for a bridge between the two.”
Combining additive manufacturing with conventional casting
421
Actuation system; 0,860 kg; 2% PU-foam; 0,823 kg; 1% Fabric/Leather; 2,027 kg; 4%
Steel; 5,808 kg; 10%
Foam; 4,626 kg; 8% Cable; 1,954 kg; 3%
LRU (IFE); 6,792 kg; 12%
Aluminium; 24,776 kg; 43% Seat Belts; 0,832 kg; 1% Plastics; 9,248 kg; 16%
Figure 19.2 Mass distribution in conventional airline seats.
Figure 19.3 Since Pier 9 is not equipped with molten metal investment casting capabilities, Bastian worked with Michigan-based foundry Aristo Cast.
422
Additive Manufacturing for the Aerospace Industry
Figure 19.4 The basic steps in the conventional investment casting “lost wax” Process. In the fabrication process used in the present work, the wax injection is replaced by the construction of a complex plastic assembly using AM. This assembly is coated with a ceramic shell and the plastic is melted out in the dewaxing step. The molten metal is then poured into the complex ceramic cavity and the final metal (aluminum or magnesium) part is formed as one monolithic piece (rather than the separate castings shown above) as shown in Figs. 19.1, 19.5 and 19.6. AM, Additive manufacturing.
Figure 19.5 Close-up view of the complex, almost organic, lattice structure resulting from the design optimization software.
Andy Harris, part of Autodesk’s advanced consulting group, worked with Bastian on the seat frame project. “We can generate these incredible high-performance designs, but we had to look beyond direct metal additive manufacturing for this project,” Harris explained. “The size and cost just wouldn’t work for fabricating this part.”
Combining additive manufacturing with conventional casting
423
Figure 19.6 Bastian shows off how lightweight the new seat frame really is. “We’ve seen a lot foundries in our region shutter their doors in recent years as manufacturing moves overseas,” said Aristo Cast CEO Jack Ziemba. “We see adopting new techniques, like additive manufacturing, even when blended with our expertise in casting, as a way forward—not just for our company but for lots of other foundries in the Midwest” [1].
Aristo Cast realized that the weight of the airplane seat frame could be reduced even more if it were cast in magnesium, which is 35% lighter than the typical aluminum. “We leapt at the opportunity to work with Andreas and Autodesk,” said Aristo Cast Vice President Paul Leonard. “It’s an exciting project and allowed us to pioneer some new techniques for magnesium casting. It also gave us a chance to learn more about advanced design and optimization techniques. That’s still quite new in our industry” [1].
Harris re-ran the part simulations in Netfabb for magnesium to confirm its properties, and Bastian sent the updated 3D model to Aristo Cast. It was 3D printed in plastic resin first, and then coated in ceramic to make a negative mold; the plastic was later heated and vaporized off after the ceramic shell had hardened. Using the mold, the foundry cast small quantities of the parts, but was able to prove that the process could actually be used to scale up to 160 seats every 2 days (Table 19.1). Bastian and Pier 9 resident Rhet McNeal determined that each seat frame, weighing in at 766 g, is 56% lighter than the aluminum seats currently in use: the
424
Additive Manufacturing for the Aerospace Industry
Table 19.1 Characteristics of airbus 320 and 380 with modified seat frames
Weight savings Annual fuel savings per aircraft Annual carbon emission reduction per aircraft Annual fleet savings (assuming fleet of 100 aircraft) Lifetime fleet savings (100 aircraft over 20 years) Fleet lifetime reduction in carbon (100 aircraft over 20 years)
A321 236 seats
A380 615 seats
214 kg 9.6 tons 28.9 tons $1,569,365
557 kg 63 tons 190.1 tons $10,332,446
$31,387,300
$206,648,920
57,800 tons/ 12,298 cars
126,000 tons/ 80,894 cars
magnesium accounts for 24% of this weight reduction, while the design optimization is responsible for the other 32%. So, if Airbus, for example, replaced the 615 seats on 100 A380 jets, which have a typical 20-year lifespan, with Bastian’s lightweight frames, the airline could save over $205,000 (this is based on 2015 average jet fuel costs). Going back to eco-friendly matters, this translates into a reduction of 126,000 tons of carbon emissions. Autodesk and Aristo Cast were recently honored by the American Foundry Society with its Casting of the Year award for the lightweight seat frame, which Bastian is quick to note is still just a research project, but one with “clear commercial applicability.” Bastian said, “The purpose of this project was never to sell seat frames. The intent is to show the power of combining Autodesk’s advanced technologies in generative design and AM with a much more widely-used fabrication process: casting. Yes, there are great applications for aerospace, but this combination can also be used in automotive, medical devices, industrial equipment, and many other fields [1].
A concern with magnesium alloys is their flammability. However, Magnesium Elektron has recently developed rare earth containing alloys that are not susceptible to this problem [2]. Both Elektron 21 (also known as ASTM EV31A), a sand casting, and Elektron 43 (Mg-4.0Y-2.9RE-0.2Zr) (also known as ASTM WE43C), a wrought plate or extrusion alloy, have undergone extensive flammability testing by the FAA, which has shown that the use of these Elektron 21 and Elektron 43 alloys in aircraft interior components (such as seat frames) does not reduce the level of safety of the aircraft when compared to heavier aluminum seat components. This is a very significant and major breakthrough for Mg alloys. The technology being used produces a cast structure and so the integrity of the investment casting is
Combining additive manufacturing with conventional casting
425
critical, also the ability to scale up to the necessary volumes while maintaining the quality. There are at least 400,000 seats being made annually. All the work Magnesium Elektron has done to date involves both high-strain-rate impact testing to meet the various loading criteria, most extreme of which is the 16 g forward load requirement, and fatigue; although 20,000 cycles is the minimum specified the seat producers and airlines generally work to a much higher standard [2].
Conclusions Key to the technology described in this paper is the utilization of AM to fabricate a complex plastic precursor that is then used to produce the mold for a metal casting. As there are a lot of areas in an aircraft seat where moderate tensile strength in conjunction with moderate ductility is best practice, magnesium now has a chance for introduction into the aircraft seat business. The aircraft seat industry has been eager to get clearance for takeoff and landing with magnesium alloys for a long time, and now has this approval.
References [1] Autodesk web site, accessed 5-12-17. [2] M. Alderman, Magnesium Elektron, Private Communication, May 15, 2017.
This page intentionally left blank
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
20
Petr A. Vityaz1,2, Mikhail L. Kheifetz1,3 and Sergei A. Chizhik1,4 1 Presidium of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus, 2 Joint Institute of Mechanical Engineering of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus, 3State Scientific and Production Association hhCenterii of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus, 4 A.V. Lykov Heat and Mass Transfer Institute of the National Academy of Sciences of Belarus, Minsk, Republic of Belarus
20.1
Introduction
The aerospace industrial products are typically characterized by a complex shape and, very so often, are made from materials that feature poor machinability. Surface engineering of parts formed from such materials necessitates the use of concentrated energy fluxes to make processing more intense and ensure high product quality in additive manufacturing [13]. Therefore, when designing aerospace industrial processes, of primary importance is the system analysis of processing methods that use the concentrated energy fluxes in additive manufacturing in order to deposit functional layers and to shape surfaces and edges of articles [46]. The energy fluxes cannot only shape the product, but also create a composite material with a gradient of properties. Therefore, layer-by-layer synthesis of the product shape should be considered as closely related to the synthesis of composite materials in additive technologies [79]. Additive synergetic technologies of the layer-by-layer synthesis, which are based on the surface self-organization phenomena, can stabilize the material’s properties and the thickness of a directly deposited layer, smooth out the topography of complex surfaces, and ultimately fuse into a gradient composite material formed as a result of the layers interpenetration [10,11].
Additive Manufacturing for the Aerospace Industry. DOI: https://doi.org/10.1016/B978-0-12-814062-8.00022-4 © 2019 Elsevier Inc. All rights reserved.
428
20.2
Additive Manufacturing for the Aerospace Industry
System analysis of the processing methods using the concentrated energy fluxes
At present, the scientific and technological requirements toward the surface strength, hardness, toughness, and wear resistance turn out to be much more stringent than those imposed upon traditional processing methods, which, in some cases, appear inefficient in ensuring the necessary quality parameters of the product functional layers. So, the aerospace enterprises often apply processing methods connected with the use of plasma, electric arc, laser, electron beam, ion, and other high-energy-density sources. Due to the processing complexities that are connected with the use of various energy fluxes, either sequentially or in parallel order, for forming various surface layers, a rational layer design for the composite-material product should be considered. Classification of technological processes (Table 20.1) includes the following generalized groups: (1) separation (e.g., by cutting) of a blank material into workpieces with volume V1 and manufacturing of a part with volume V2; (2) deposition of a surface layer, i.e., coating; in this case V2 . V1, (3) thermal or other treatment by energy fluxes (V2 V1); (4) removal of excess material by cutting (V2 , V1); (5) deformation, compaction, and smoothing of the material (V2 V1) [10,12]. This classification enables one to formulate appropriate boundary conditions to an open technological system (here term “open” is used in the synergetic meaning). Additional degrees of freedom of boundaries in the open system, such as movements or changes of external influences, enable one to control the nonequilibrium state of the system, while additional energy fluxes at the initial moment may provide stabilization of a nonequilibrium technological process [5,11]. Fig. 20.1 summarizes the analysis of efficient processing technologies that employ concentrated energy fluxes [4,5], which combine several methods: (1) those that change the boundary conditions by introducing additional degrees of freedom for the motion of a working medium (e.g., tool, technological medium, machining allowances, formed surface, etc.), and (2) those that change the initial conditions through additional energy sources having different concentration levels. The latter are conventionally subdivided according to the mode of energy distribution, or localization, over the surface of a part: uniformly surface-distributed (I), multiply localized (II), and focused (III). In fact, most of the combined methods can be implemented in manufacturing and only some of them are virtually impossible at present. But, at the same time, such synergetic effects may also occur as side effects and incidental phenomena in surfacing. The performance analysis demonstrates the feasibility of processing methods that combine both mechanical and thermal effects [10,11]. Distribution analysis of the processing methods, with regard to the accuracy of surface formation, was performed according to the classification scheme shown in Table 20.1 depending on the energy concentration level (I, II and III) for a variety of typical energy sources (Fig. 20.1). For deformation and machining, the accuracy of the surface treatment was assessed by deviations in size and shape, waviness, and roughness; for thermal treatment, it was judged by the irregularities in the depth of thermal hardening or softening zones and by the thickness of a defective surface
Table 20.1 Combined physicochemical methods of surface formation: effective in industry ( 3 ), of low effectiveness (2) and unfeasible (0)
Cladding
Amorphization/ melting
Spray coating
Surface alloying
Thermal shock
Quenching
Annealing
Tempering
Edge tool/cutter
Tool with forced displacement
Self-moving tool
Abrasive
Striker
Plate
Roller
Ball
V0 . Deformation using different tools
Evaporation
IV0 . Machining
Melt blowing
III. Single focused
III0 . Heat treatments
Melt outflow
II. Multiple localized
1. IH (102). . . 103. . .104 2. GF 102. . .103. . . (3 3 103) 3. PA 5 3 102. . . 3 3 104 4. EH 103. . . 5 3 104 5. WA 103. . .105. . . (106) 6. SD 5 3 106. . . 8 3 108 7. EB (103). . . 106. . . 8 3 108 8. CW laser (5 3 103). . . 106. . .109 9. PL (107). . . 1010. . .1014
II0 . Surface engineering
Thermal splitting
I. Bulk
Typical sources of energy and their power q, W/cm2
Heat release zones
I0 . Material removal
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
0
0
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
IH, Induction heating; GF, Gas flame; PA, Plasma arc; EH, Electrocontact heating; WA, Welding arc; SD, Spark discharge; EB, Electron or ion beam; CW, Continuous wave; PL, Pulse laser.
430
Additive Manufacturing for the Aerospace Industry
Figure 20.1 The distribution of the components’ surface forming operations according to the surface treatment accuracy and quality δ (μm), depending on the power density q (W/cm2) of standard sources with various levels of energy concentration (The legend is in Table 20.1).
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
431
layer; for cutting into pieces and coating deposition, all the aforesaid indicators were taken into account. Analysis of the surface accuracy formation shows (Fig. 20.1) that the power density increases from the first to the third level of energy concentration (I ! II ! III), which, in turn, results in a reduced size of the energy distribution zones. On the other hand, reducing the size of heat distribution zones (from I0 to V0 ) increases stress concentration, which affects the shape and accuracy of the formed surface. It is obvious that on level I, the accuracy does not increase. This is connected not with a higher energy concentration, but primarily with a wider heat distribution zone. At level II, the accuracy is minimal. This is because of formation of a large number of stress concentrators due to the existence of numerous localized heat concentration zones. At level III, the accuracy does not decrease and, then, with raising the energy density, it substantially increases due to heat focusing into narrow zones, which is accompanied with a rapid increase in stress concentration. From the above analysis, it follows that optimal initial conditions for different technological operations of surface formation can be provided by a proper choice of a concentrated energy source under given boundary conditions. Typical sources of level I can be most efficiently used for deformation of largesize parts, machining in the conditions when a large volume of material has to be removed, for coating deposition and bulk heat treatment/hardening. Sources of level II significantly decrease the surfacing accuracy, so they should be combined with cutting and deforming tools, in particular in the coating deposition and heat treatment processes. The best results in all technological surfacing operations are obtained with the sources of level III. The data presented in Table 20.1 support this conclusion, and the surfacing accuracy analysis proves that the proposed classification of processing methods, which employ concentrated energy fluxes, is valid [4,5]. The above analysis outlines promising ways to advance surface engineering that will provide specified accuracy and other quality factors. A technological system for high-energy-density processing is open (in the synergetic sense) primarily to thermal and mechanical energy fluxes, which determine the change of initial and boundary conditions for surfacing with an accuracy of the order of microns. The energy that exceeds a certain level, below which the system maintains dynamic equilibrium, should have conditions to dissipate or be adsorbed by additional degrees of freedom of the system. In a thermomechanical system, additional degrees of freedom are provided by the movements of a working medium (tool, technological media, and environments) as well as by additional structures, phases, and increased number of interfaces that adsorb excess energy and maintain the shaping process in one or more particular states. Thus, increased surfacing efficiency is ensured both by additional energy fluxes and the degrees of freedom of the system’s elements. With increasing energy concentration, the zone, where the incoming energy flux interacts with surface, undergoes spontaneous evolution: from a wide, surface-distributed zone it disintegrates into multiply localized zones and then they focus into a single spot. Additional degrees of freedom of the system’s elements allow modification
432
Additive Manufacturing for the Aerospace Industry
of the surfaces, phases, and structures and thus permit maintaining the energy fluxes and interaction processes in certain states through certain surface selforganization phenomena.
20.3
Additive synergetic technologies of layer by layer synthesis
Advanced technologies, along with the new hardware and software, equipment, and outfit, are based on layer-by layer growing of surfaces and self-organization of structures of the composite material [7,9]. Thus, when defining the basics of advanced nanotechnologies, Alferov [13], in addition to atom probe microscopy, mentioned the epitaxial growth of films on a surface and processes of self-assembly of the material heterostructures. Moreover, according to the synergetic concept, there is a limited number of states and transition rules in a technological system [4]. The dominant processes of structure formation under intensive influences can be defined with the notion of mode, which is used when describing distributions of continuous random variations of a control parameter [10]. The notion of mode signifies such a parameter value when its distribution density has a maximum. According to the synergetic concept, stable modes adjust to the dominant unstable ones and, thus, can be ignored. This results in a sharp reduction of the number of control parameters, and the remaining unstable modes can serve as order parameters determining the processes of structure formation [7]. So, particularly promising is an approach that considers additive methods as synergetic technologies that enable surface self-organization phenomena to occur in the layer-by-layer synthesis of different materials and provide control of their properties under various physical influences [14,15]. The surface self-organization brings about stable layer formation of a certain thickness when the distance from the energy source or a feed material to the forming surface changes significantly and thus allows to join successive layers as a result of interpenetration [4]. The distance to the surface is a very sensitive factor in the layer formation mechanisms that operate in the direct deposition (DD) technique while a change in the distance is especially important for the layer-thickness stabilization in the bed deposition (BD) technologies. From the above, it can be concluded that the choice of an energy source and/or feed material in additive manufacturing determines not only the technological route but also surface phenomena that provide the self-organization processes to occur in synergetic technologies [4]. Direct growth of parts during the layer-by-layer synthesis is possible in different phase states, such as solid, liquid, or gaseous, and can be implemented in a variety of high-energy processes [6,7] with surface-distributed, multiply localized, and single-focused energy adsorption zones [16,17] depending on the power density in the energy flux (Table 20.2).
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
433
Table 20.2 High-energy processes of layer-by-layer synthesis of parts Cutting and deposition processes
Phase state of the deposited material
Energy adsorption zones and energy flux density, W/cm2
Thickness of the formed or removed layer, mm
Plasma cutting, spray powder deposition, and hard-facing Electroerosion treatment and electromagnetic hard-facing of powder Ion implantation and deposition Electron-beam and laser cutting, surface melting, and alloying
Solid (powder)
Surface-distributed, 5 3 102105
0.110.0 [18]
Hardsurfacing (powder) Gaseous
Multiply localized, 103105
0.050.50 [19]
Multiply localized, 103105 Single focused, 5 3 103107
0.0020.200 [5]
Liquid (melt)
110 (in thermoelectric convection 0.11.0) [20]
Surface-distributed energy adsorption zone is formed in plasma-base processes, such as plasma cutting, deposition, and hard-facing using metal powders [18]. The thickness of the formed layer is determined jointly by kinetic and thermodynamic factors, such as the velocity of powder particles and the thermodynamic potential of the plasma flux, while the layer formation process is characterized by a ratio between the kinetic and Joule energy of the flux. Multiply localized energy adsorption zones are formed in the course of surface electroerosion processing and electromagnetic powder deposition. In electromagnetic deposition, the thickness of a quality coating is restricted: above a certain thickness the layer loses stability, surface peaks appear, and grow. Then the apexes of these peaks melt and evaporate so that craters are formed on their summits [19]. The electromagnetic flux allows controlling the surfacing process. Thus, motion of the ferromagnetic powder particles and fixation of the latter to the surface is determined by the induction of the magnetic field while the heat release intensity in the formed surface joints and powder melting are determined by the electric field strength. The synergetic effect of the aforementioned factors brings about stabilization of the layer thickness [4]. The process of electromagnetic surfacing is determined by electromagnetic and inertial forces and proceeds through electromagnetic interaction of particles with the electric field in the working zone. In the ion implantation and gas-phase deposition, the unfocused ion beam is distributed over the surface of the part, forming multiply localized energy adsorption zones. The thickness of the produced coating is determined mainly by the thickness of the layer where the electric potential applied to the part effectively influences the ions [19]. This layer is determined by a relation among the electric field potential, the plasma ion flux density, and the charge and mass of ions. Due to a joint effect of the potential energy of ions in the electric field and thermal energy of the plasma
434
Additive Manufacturing for the Aerospace Industry
flux, the ions become spatially distributed according to the exponential law thereby determining the thickness of deposited coating [5]. In the electron-beam and laser cutting, surface melting, and alloying, a singlefocused energy adsorption zone scanning over the entire formed surface should be considered [4]. In electron beam processing, dissipative vortex structures are formed as a result of convective instability in a thin surface layer of a melt. At rapid cooling, a cellular structure is formed along the crystallization front. The thickness of the modified layer is described by the material properties such as the surface tension and volume expansion coefficients and the density in molten state [20]. The formation of dissipative structure in the melt is determined by thermocapillary phenomena and is related to the buoyancy force and energy dissipation in the melt [5,20]. The considered high-energy processes permit performing layer-by-layer deposition of materials with special properties on a complex-shape surface. This, in turn, enables engineers purposefully to modify the physical and mechanical properties of a surface, tailoring them to the working requirements imposed on a particular machine part [4,5,21]. The layer deposition conditions in high-intensity processes should be related to the design features of a target product. The plasma spraying and surfacing processes involve edge rounding [18]. The thickness of a deposited layer decreases on the most protruding sections in electromagnetic surfacing [19]. In ion deposition of coatings, the layer grows most intensively on the peaks with a small radius at the vertex due to the increase of the electric potential [5]. In electron-beam or laser melting, the thickness of the modified layer is nonuniform and depends on the alloying elements (if any) because of the segregation of elements in the melt due to the formation of vortices [20]. Finally, the conditions of layer deposition in high-energy processes should be related to the design features of the target product. For layers of a prescribed thickness, which are formed in physical fields, the stability of a technological system should be ensured in the induction, plasma, electromagnetic, laser, electron, and ion beam surface treatment. The system analysis of the concentrated energy fluxes used in additive manufacturing of aerospace products from poor machinable materials and the surface self-organization phenomena that occur in layer-by-layer synthesis reveal that ion and electron-beam treatment of aviation materials in a vacuum should be considered as very promising.
20.4
Ion implantation and ion deposition of coatings
Combined ion modification or ion implantation followed by ion deposition of coatings is a method for improving the quality parameters both in the process of ion modification of a surface and in the subsequent service of the latter [6,11]. As a result of high-energy surface processing, alloying atoms can be implanted at a rather high concentration into a material to form a modified transition layer. The implanted ions experience numerous elastic collisions with atoms of the crystal
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
435
lattice, and radiation defects, whose number by two or three orders of magnitude exceeds the number of implanted atoms, are produced in this layer. These two processes have a substantial effect on physical, mechanical, and physicochemical properties of the surface, which sometimes brings about structural and phase transformations. When such a composite layer works in the condition of friction, a specific self-organization process occurs within the layer, which involves interaction of dislocations with grain and phase boundaries and slows down the development of microplasticity [6,11]. Initial conditions for the formation and reorganization of ion-modified layers are determined by chemical, phase, and structural composition of the substrate, type of the ions and base atoms, and the implantation energy. The structural self-organization of such a layer during friction depends on the energy of frictional interaction of the surfaces. The boundary conditions in this system depend on design of modified layers and relative displacement of the surfaces during service. To reveal the mechanism of energy dissipation and the behavior of atoms in an ion-modified layer, molecular dynamic simulation was performed for the case of chromium-implanted heat treated R6M5 high-speed tool steel (0.82%0.9% C, 5.5%6.5% W, 4.8%5.3% Mo, 3.8%4.4% Cr, 1.7%2.1% V and up to 0.6% Ni according to Russian standard GOST 19265-73) [6,11]. In the as-implanted state, Cr atoms were considered as either located in interstitial sites (tetrahedral or octahedral) of the martensite lattice or forming asymmetric dumbbell-shaped pairs with host atoms (the so-called interstitialcy defects). Such configurations of chromium atoms are nonequilibrium due to a substantial elastic distortion of the lattice around them. Meanwhile, atoms in these configurations are known to be highly mobile. As the implanted atoms migrate, they may 1. interact with vacancies and form stable substitutional configurations, 2. interact with each other or with other impurities in solid solution (for example, with carbon) to form clusters, and 3. escape into sinks (dislocations, grain, and phase boundaries).
As mentioned above, during implantation, the vacancy concentration in the modified surface layer substantially increases because of the radiation damage. Therefore, one of the most probable diffusion-controlled reactions is the relocation of implanted chromium atoms into substitutional positions. Analysis of the Auger spectra confirmed the presence of a modified transition surface layer with a thickness of about 1 μm (Fig. 20.2A). In this domain, the microhardness, which was measured using a 10 g load on the indenter, was found to increase from 5000 MPa in the initial state to 20,000 MPa after modification (Fig. 20.2B). X-ray diffraction analysis has revealed a substantial increase in the martensite lattice parameter in the modified layer, from 0.2879 to 0.2883 nm. At the same time, a shift of the (1 1 0) and (2 0 0) martensite reflections toward smaller diffraction angles was observed or a slight decrease in the half-widths of the martensite reflection lines. All these observations confirm our previous calculations performed for the layer modified with chromium ions and testify to an increase of compressive stresses. Moreover, the obtained results reject the idea that the martensite lattice parameter changes due to distortion and different kinds of lattice strain and suggest
436
Additive Manufacturing for the Aerospace Industry
Figure 20.2 Distribution of chemical elements over depth H (μm) in R6M5 high-speed steel (6% W, 5% Mo) after combined vacuum-ion surface modification (A); plot of surface microhardness Hμ (MPa) versus load P (N) on the indenter (B) in the initial state (1) and after modification (2) of R6M5 steel.
that only one explanation appears to be valid: iron atoms are substituted by chromium atoms on the lattice sites, i.e., solid solution hardening occurs. Microstructural analysis has shown that after vacuum-ion modification, the number of microcracks in the surface layer decreases without a notable change in the phase distribution pattern. In general, the type of microstructure does not change. The transition zone becomes a barrier for microdefects emerging on the surface, thus increasing wear resistance of the high-speed steel. So, combined vacuum-ion modification of the surface of the high-speed tool steel with chromium atoms brings about substantial strengthening, which was predicted by the energy-dissipation simulation of ion-beam processing and subsequent operation of the material. The predicted formation and growth of chromium clusters in the martensite lattice is energetically favorable, and they can serve as sites for
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
437
Figure 20.3 The model of the formed multicomponent coating on the peaks (A) and macropeaks (B) of the surface asperity: 1—heavy ions; 2—light ions; 3—self-spraying zones.
further precipitation of intermetallic phases that improve the service tribological properties of the surface. The deposition of coatings or structure modification of surface layers by energy fluxes results in layered structure of the processed material. Transition zones are formed between the layers and their dimensions have a significant effect on physical and mechanical properties. The study of surfacing mechanisms combined with modeling of the coating structure and, particularly, formation of boundaries between the deposited layers, enables one to select optimal modes of deposition and control the quality of surface treatment and hence working properties of the material [3,7,8]. When considering the mechanism of deposition of multicomponent ion-vacuum coatings, a smooth surface with peaks was taken as a basis (Fig. 20.3). A negative accelerating potential is applied to the surface. The ion beam directed to the substrate consists of singly charged ions of several elements, which significantly differ by mass. Far from the solid material, the ion velocity vector is directed perpendicular to the surface. When ions approach the peaks, the direction of their flight changes because of a significant change in the electric field strength. Due to a difference in ion masses, deviations will be different for unlike species. The trajectory of ions with a smaller mass has a larger curvature, and they are deposited mostly near the peak’s top. Ions with a larger mass, as well as an electrically neutral phase (vapor, liquid droplets, fragments), are deposited more uniformly [7,8]. Due to inhomogeneity of the electric field strength, separation of the ion beam by mass is observed near the surface of the material, while, in reality, the same is observed with respect to the charge as well. Such separation of the ion flux at the initial stage of coating formation will result in rapid growth of the coating at the top of the peak (Fig. 20.3A). However, as it grows, the corner radius will decrease.
438
Additive Manufacturing for the Aerospace Industry
As a result, at a certain instant of time, the corner radius and thus the electric field intensity will reach the values when the sputtering rate and the condensation rate become equal, and the growth of the surface peak terminates. At the same time, the process of coating formation in between the asperities will still continue, mostly due to deposition of heavy ions and the neutral phase. As a result, the space between them becomes so filled that it itself will become a site for the growth of a new peak (see Fig. 20.3A). In the considered model, separation of the ion flux by the mass of atoms occurs. However, in reality, the beam includes differently charged ions, which will lead to ion separation by charge. In ion-vacuum processing, condensation is accompanied by ion sputtering (Fig. 20.3B). Moreover, if the condensation surface features macropeaks that increase the electric field strength, a more intense self-sputtering process can be expected. As a result of the multicomponent ion condensation, zones with increased concentration of light and multiply charged ions will appear, where the self-sputtering process is more intense than in other regions (see Fig. 20.3B). Consequently, the concentration of light elements and those forming multiply charged ions in the coating will decrease due to self-sputtering. Thus, modeling of the multicomponent ion-vacuum coating formation should take into account ion separation by mass and charge in asperity zones, which occurs as a result of increased electric field strength. Also, it is necessary to consider the scale factor of surface irregularities and their shape [3,7,8].
20.5
Electron-beam heating of a coated surface
Electron-beam heating (EBH) of a surface with a preliminary deposited galvanic, chemical. or detonation coating is a technological process that combines melting of the coating and the substrate, with subsequent formation of a transition layer with a large thickness [6,11]. Depending on the conditions of thermal processing and the coating thickness and composition, the melted and then crystallized layer may have either equiaxial finegrained or cellular structure [4,22]. Structure formation is determined by the initial conditions, i.e., melting and rapid chilling of a molten layer, as well as by the boundary conditions, such as trajectory and rate of the electron beam scanning. Billets of titanium alloys VT6 (Ti 1 3.5%5.3% Al 1 0.3% Zr) and VT9 (Ti 1 5.8%7.0% Al 1 2.8%3.8% Mo 1 1.0%2.0% Zr) were subjected to case hardening in order to obtain a hardened layer with a thickness of up to 1.3 mm and then processed by EBH [6,11]. It has been revealed that after EBH, the size of β-converted grains near the surface increases substantially, up to 150 μm, from the initial size of about 15 μm. In this case, the size of martensite plates is determined not only by the grain diameter but also by structural microinhomogeneity caused by enhancement of diffusion processes during EBH and arresting of the formed concentration distribution by rapid chilling. Near the surface of the specimen, the phase
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
439
composition includes two types of martensite, α0 and αv, with different concentration of β stabilizing alloying elements. This is connected with the fact that α phase, which experiences the α-to-β transformation at rapid heating, features nonuniform composition. The hardness gradually decreases over the cross-section from HRC 4345 on the surface to HRC 2437 in the bulk. Studying the distribution of aluminum and titanium over a cross section enables one to differentiate two surface layers. The first (outer) layer with the thickness of 140160 μm features almost constant content of these elements throughout the depth, aluminum concentration being almost two times lower than in the initial alloy. In the second layer, increased concentration of aluminum and decreased concentration of titanium are observed. At a larger distance from the surface, the alloy structure features a large amount of retained α phase and grains of β solid solution with a lower concentration of molybdenum, which were formed during the dissolution of α phase particles at rapid heating. So, the concentration drop of alloying elements on a microlevel increases when moving away from the surface. Taking into account that α phase, as well as grains of β phase with reduced content of molybdenum, have an increased concentration of aluminum, the process of EBH causes directed motion of aluminum atoms into zones with the lowest concentration. They correlate with the domains of β phase that border undissolved particles and molybdenum-depleted zones, which are located at a larger distance from the surface. With increasing the heating temperature, the surface areas of β solid solution tend to an equilibrium state while the domains of undissolved α phase and the zone with a noticeable concentration drop displace farther off the surface, and so the concentration distribution changes. When the titanium surface experiences melting during EBH, thermocapillary convection develops in the molten layer. This phenomenon is caused by the temperature dependence of surface tension. In such situation, the so-called Benard convection cells are formed in the melt layer [6,22]. The microstructure of titanium alloys VT6 and VT20 (Ti 1 5.5%7.0% Al 1 0.8%2.5% V 1 0.5%2.0% Mo 1 1.5%2.5% Zr), which forms after quenching from a liquid state due to fast heat removal to the interior, features closely packed hexagonal-cylindrical cells with a diameter of 57 μm. Microvoids that are observed at grain boundary junctions have a size of up to 1 μm. In the transition zone from cellular to β-transformed structure, the amount and size of cells steadily decrease, the grain size being 100200 μm. In the interior, the maximum size of martensite-type needles corresponds to the cell diameter (Fig. 20.4). In the process of dissipative structure formation, intensive redistribution of alloying elements occurs in liquid phase; the elements that reduce surface tension are accumulated near the walls and at corners of the cells. Substantial segregation of elements in β phase results in the formation of alloyed αv martensite in the VT20 single-phase pseudo-α alloy in zones with cellular structure. The αv phase lattice parameters change from a 5 0.2922 nm and с 5 0.4667 nm in the initial state to a 5 0.2923 nm and c 5 0.4729 nm for cellular structure, i.e., the tetragonality of α0 martensite increases. The α0 martensite lattice parameters are the following: a 5 0.2952, b 5 0.5294, с 5 0.4691 nm. As the martensite-type αv phase is formed in the structure of VT20 alloy, the thermostability of the latter may increase.
440
Additive Manufacturing for the Aerospace Industry
Figure 20.4 Cellular structures ( 3 3000) of titanium alloys VT20 (a) and VT6 (b) with unetched surfaces after electron-beam heating for 1 s with power density of 3 kW/cm2.
EBH of titanium alloys VT6 and VT23 (Ti 1 4.4%5.7% Al 1 4.0%5.5% V 1 4.0%5.5% Mo 1 0.5%1.5% Cr 1 about 0.3% Zr) with a preliminary oxidized surface permits increasing the microhardness both in the surface layer and beneath it, which testifies to the diffusion of oxygen to deeper layers. The α-phase layer becomes four times thicker. EBH of VT6 and VT23 alloys after preliminary siliconizing causes redistribution of aluminum farther from the surface. The thickness of the α-phase layer increases twice, the layer becomes more porous and partially undergoes martensitic transformation. The total thickness and porosity of the silicide layer, which was formed on the surface by the preliminary siliconizing treatment, also increase [6,22]. When VT20 alloy with a galvanic chromium coating is heated by an electron beam, the structure with grain diameter of 1020 μm is formed. The thickness of diffusion layer and that of residual chromium coating are 1012 and 56 μm, respectively [6,22]. As the heating temperature increases, the continuity of the outer chromium coating is destroyed. Due to a higher rate of grain boundary diffusion, chromium diffuses along the β titanium grain boundaries into the titanium matrix of VT20 alloy. In this case, the grain size in the bulk is 80120 μm. At further heating to a higher temperature, TiCr2 intermetallic starts melting. This brings about a characteristic fine-grained structure that is formed after crystallization. The grain size is about 210 μm and the thickness of the diffusion layer is about 30 μm. When for EBH processing is performed with the goal of diffusion saturation of the aforesaid titanium alloys with chromium, the heating regime is adjusted so as to avoid surface melting. Then, solid solution of alloying elements in the surface layers of α-titanium is formed along with compound TiCr2. The presence of α phase throughout the entire diffusion layer indicates that the content of titanium is larger than of chromium in the layer because Cr reacts with Ti to form intermetallic compound TiCr2. Deeper into the material, where the concentration of chromium is insufficient to produce TiCr2, chromium dissolves in titanium to form β solid solution. After EBH of alloy VT20 with a preliminarily deposited Ni coating, the grain structure of the nickel layer becomes distorted. With increasing the heating
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
441
temperature, recrystallization of the surface zone occurs and the resulting structure consists of equiaxed grains having 1030 μm in diameter with a pronounced martensitic intragranular structure. Solid solutions with martensitic structure, as well as β solid solution of nickel in titanium, are formed in more distant layers. The Nienriched grain boundaries are formed due to dominating nickel diffusion along the boundaries of β-grains in VT20 alloy, which is accompanied by the intermetallic compound formation [6,22]. At the EBH of chromium coating with a nickel sublayer, which were preliminarily deposited on the surface of titanium alloy VT20, interdiffusion the TiNiCr system occurs, which results in the formation of multicomponent solid solutions and phase Ti2Ni (see Fig. 20.5AG). At the boundary between the nickel sublayer and titanium substrate, eutectic reaction between Ti and compound Ti2Ni occurs. Then the eutectic-melt layer crystallizes in the form of dendrites with a length of 812 μm in the direction normal to the outer surface. With a further rise in temperature, the diffusion layer thickness reaches 300 μm. As the eutectic reaction proceeds, convection flows appear in the melt pool, thus enhancing the exchange of elements between the coating and substrate, which results in the growth of strengthening layers and homogenization of the diffusion zone.
Figure 20.5 The microstructures of the surface layer ( 3 300) of the VT20 alloy with NiCr coating after electron-beam heating to 700 C (A), 800 C (B), 900 C (C), 1000 C (D), 1100 C (E), 1200 C (F), and 1300 C (G).
442
Additive Manufacturing for the Aerospace Industry
Figure 20.6 The microstructures of the surface layer ( 3 500) of the VT9 alloy coated with WC-15% Ni in the initial state (A) and after electron-beam heating to 1100 C (B), 1300 C (C), and 1400 C (D).
Experimental study of surface engineering by EBH was performed for titanium alloy VT9 with a hard-alloy WC-15% Ni coating, which was earlier deposited by detonation spraying. EBH results in partitioning of the coating and underlying subsurface layers into a number of characteristic zones (Fig. 20.6AD). Due to diffusion of nickel from the coating into titanium substrate, a zone of Ti-base solid solution is formed at the former Ni/VT9 interface. Closer to the outer surface, a zone of the β 1 Ti2Ni eutectic appears, and intermetallic compound Ti2Ni is formed on the outside. The total thickness of these zones is about 20 μm. Together with nickel, up to 5 wt.% of the tungsten initially contained in WC diffuses into the titanium substrate, thus forming a Ti-base solid solution. Titanium concentration decreases smoothly in the direction to the surface while the content of nickel remains approximately the same throughout the formed solid solution layer. In the course of heating, aluminum actively diffuses across the coating from the VT9 substrate to the outer surface [6,22]. An increase in heating temperature causes a transformation of the complex structure of the whole coating into a eutectic mixture containing round-shaped particles of tungsten carbide. The coating processed by EBH acquires a fine-grained structure with a grain size not exceeding 1 μm. Also, experimental study of EBH of the VT9 alloy with a hard-alloy WC-25% Co coating, which, as in the previous system, was preliminarily deposited by detonation spraying. In this case, intermetallic compounds of titanium with Co (the binding metal in the initial hard-alloy coating) are formed at the coating/substrate interface (Fig. 20.7AD). Liquid phase that forms during heating substantially intensifies the diffusion processes. The thickness of the modified layer as a whole exceeds the thickness of intermetallic-compound layers in the titanium substrate by a factor of 15. This is associated with a higher rate of titanium diffusion in the metal binder, and may partly be attributed to the enhancement of diffusion in the field of temperature gradient [6,22]. In the outer layer, the TiCo2-base eutectic acts as a binder that provides strong adhesion of hard tungsten carbide particles to the alloy thus imparting high wear resistance to the article.
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
443
Figure 20.7 The microstructures of the surface layer ( 3 500) of the VT9 alloy coated with WC-25% Co in the initial state (A) and after electron-beam heating at 1100 C (B), 1300 C (C), and 1400 C (D).
The microhardness distribution over the cross-section of the resulting coating features a gradual decrease from the initial hard alloy coating to the titanium substrate. After EBH, microhardness of the 0.10.15 mm thick outer layer is close to that of the initial tungsten carbide-cobalt hard alloy coating. Deeper into the material, microhardness decreases and in the bulk it corresponds to a value typical of the titanium alloy [6,22]. Thus, during EBH of a titanium alloy with a hard-alloy detonation coating, the nature of phase and structural transformations and the resulting composition are mainly determined by diffusion processes at the initial coating/substrate interface. Unlike EBH of electrolytic coatings on titanium alloy, in the case of detonation coatings, the lamellar structure, which features different content of elements in the layers, is preserved to temperatures up to 15001600 C. The refractory tungsten carbide particles, which are present in the outer layer, hinder the development of convective flows in the melt that forms at heating, and hence prevent homogenization of the microstructure and composition in the outer layers. In electron-beam surface engineering, the value of a quality parameter, e.g., hardness (Fig. 20.8A) should be compared with a structural parameter of the surface. The latter can be characterized by a relative area of a modified surface, i.e., a ratio of the surface area with modified structure to the total surface area (Fig. 20.8B) [4,5,11]. Thus, formation of cellular structure on the EBH-modified surface of a singlephase pseudo α titanium alloy (Fig. 20.4) occurs through the formation of dissipative vortex structures in the molten state. In this situation, a change in the state of the thermodynamic system under consideration is determined by a prevailing convection mechanism. At the onset of melting, narrow cells that are formed due to thermocapillary force, which is described by Marangoni number, first originate and then displace to the periphery of the heated spot. These cells consist of dissipative vortices. With increasing the power density of an electron beam, natural convection, which is characterized by Grashof number, develops in the central part of the heated zone and blurs the aforesaid vortex dissipative structures. Wide toroidal
444
Additive Manufacturing for the Aerospace Industry
Figure 20.8 The distribution of the microhardness Hμ (GPa) over the depth H (μm) of the surface layer of a titanium alloy with a chromium-nickel coating (A) with a temperature change T(K) and the relative surface area S(%) with a modified structure as a function of the specific power q (kW/cm2) and the duration τ (C) of electron-beam heating (B), the visible boundary of the modified layer (I) and the melting point (II) are indicated by dashed lines.
vortices generated by the buoyancy force, which is described by Rayleigh number, in the conditions of EBH are observed only when deep melting occurs [4,5,11]. Studying the ratio of the surface area with a cellular structure to the total surface area (parameter S), which depends on power density q and heating time τ, has shown that the largest area with regular structure, S 5 40%, is formed in a narrow range of heating intensity (see Fig. 20.8B) [4,5]. From the viewpoint of the optimal control theory and nonequilibrium thermodynamics, the above described situation can be considered as a controlled twoparameter system. Then the cellular structure formed over a maximum surface area (with S 40%) is characterized by a special state of the system, which is named the unstable node, and with time the latter evolves to a limit cycle. This cycle is limited by the solid-to-liquid phase transformation of the material. Within the frame of these theories, formation of the solid/liquid interface during melting is described as the unstable saddle mode. Heat/mass transfer from/to the solid/liquid interface occurs by thermal conduction and convection, which stabilizes the state of the system [6,7].
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
445
The microhardness distribution (see Fig. 20.8A) depends on structural changes in the system, which are connected with complex physicochemical transformations that occur in the surface layer of titanium alloys with preliminarily deposited coatings under the action of EBH. This distribution may be correlated to the pattern of cellular structure formation (see Fig. 20.8B). Fig. 20.8A refers to electron-beam processing of a titanium alloy with a chromium-nickel coating [4,5]. The aforementioned transformations include chemical reactions, solid-state diffusion of elements form the coating into substrate, melting and eutectic reactions at the coating/substrate interface, propagation of the melting front into the solid substrate, development of convection in the melt pool, and subsequent solidification of the multicomponent melt in the field of temperature gradient. In general, the pattern of these complex processes strongly depends on the nonsteady-state temperature field in the material, which, in turn, is eventually connected with the electron-beam processing regime. Thus, in order to attain a high coating-to-substrate bonding strength and improve physicochemical parameters of the obtained surface structures, it appears necessary rigorously to control the range of electron-beam power density and scanning regime, which should be linked to the chemical composition of the system and the coating thickness.
20.6
Conclusion
Additive processes of direct growth, or layer-by layer synthesis, which are selected in accordance with the design features of the formed layers (BD-technologies) and shells (DD-technologies) offer new opportunities for customized design and rapid prototyping of machine parts. The surface self-organization phenomena that occur in metals/alloys under the action of high energy density fluxes enable an engineer to form outer layers of required thickness over a complex-shape surface. Moreover, these technologies open up novel, very so often unique opportunities for synthesizing surface layers with phase composition, structure, and hence properties tailored to the target function of an article. In other words, synergetic processes that are intrinsic in the additive manufacturing technologies employing high-energy density fluxes, e.g., ion and electron beams, present substantial interest not only for mechanical engineering but for materials science as well. The synergetic processes that occur in these technologies include, first of all, self-organization of the energy adsorption zones on the material’s surface: from surface-distributed to multiply localized and to single-focused. Second, selforganization reveals itself in the formation of dynamic dissipative vortex structures in the melt pool that forms on the metal surface at fast heating by an energy flux. This has a profound effect on the spatial distribution of alloying elements, microstructure, and phase composition of the final solid product. In particular, formation of regular cellular structures that are observed in the as-solidified state occurs namely due to the aforesaid synergetic phenomena.
446
Additive Manufacturing for the Aerospace Industry
A deeper insight into the underlying physical and physicochemical mechanisms that, in their synergism, produce the final structure of materials/coatings, necessitates complex cross-disciplinary research that should employ experts in mechanical engineering and in materials science. Such research will contribute to further development of additive manufacturing using high-energy density fluxes and to the creation of both novel advanced materials and coatings for complex-shape articles.
Acknowledgments The authors wish to thank Professor Anatoliy I. Gordienko for a helpful discussion and Professor Boris B. Khina for numerous discussions and invaluable help in improving the English language and presentation style of the paper.
References [1] I. Gibson, D. Rosen, B. Stuker, Additive Manufacturing Technologies: 3D Printing, Rapid Prototyping, and Direct Digital Manufacturing, Springer, NY, 2015. 498 pp. [2] M.A. Zlenko, A.A. Popovich, I.N. Mutylina, Additivnye tekhnologii v mashinostroenii [Additive Technologies in Mechanical Engineering], Publ. House of Polytechnical University, St. Petersburg, 2013. 222pp. (In Russian). [3] O.P. Golubev, S.V. Kuchta, Zh.A. Mrochek, D.N. Svirsky, B.N. Soukhinenko, M.L. Kheifetz, in: Zh.A. Mrochek, M.L. Kheifetz (Eds.), Perspektivnye tekhnologii mashinostroitel’nogo proizvodstva [Advanced Technologies in Engineering Production], The Polotsk State University Publ., Novopolotsk, 2007. 204 pp. (In Russian). [4] M.L. Kheifetz, Proektirovanie protsessov kombinirovannoy obrabotki [Design of Combined Processing], Mashinostroenie Publ., Moscow, 2005. 272 pp. (In Russian). [5] M.L. Kheifetz, L.M. Kozhuro, Zh.A. Mrochek, Protsessy samoorganizatsii pri formirovanii poverkhnostey [Self-Organization Processes in the Formation of Surfaces], The V. A. Belyi Metal-Polymer Research Institute of the National Academy of Sciences of Belarus Publ., Gomel, 1999. 276 pp. (In Russian). [6] P.A. Vitiaz, M.L. Kheifetz, S.V. Koukhta, Laser-Plasma Techniques in ComputerControlled Manufacturing., Belorusskaya Nauka Publ., Minsk, 2011. 164pp. [7] M.L. Kheifetz, Formirovanie svoystv materialov pri posloynom sinteze detaley [Formation of Materials Properties at the Layer-by-Layer Synthesis of Machine Parts], The Polotsk State University Publ., Novopolotsk, Belarus, 2001. 156 pp. (In Russian). [8] P.A. Vityaz, A.F. Ilyushchenko, M.L. Kheifetz, S.A. Chizhik, et al., Tekhnologii konstruktsionnykh nanostrukturnykh materialov i pokrytiy [Technology of Nanostructured Materials and Coatings], in: P.A. Vityaz, K.A. Solntsev (Eds.), Belaruskaya Navuka Publ., Minsk, 2011, 283 pp. (In Russian). [9] A.M. Rusetsky, P.A. Vityaz, M.L. Kheifetz, L.M. Akulovich, et al., Teoreticheskie osnovy proektirovaniya tekhnologicheskikh kompleksov [Theoretical Basis of Technological Systems Design], in: A.M. Rusetsky (Ed.), Belaruskaya Navuka Publ., Minsk, 2012. 239 pp. (In Russian).
Synergetic technologies of direct layer deposition in aerospace additive manufacturing
447
[10] V.S. Ivanova, A.C. Balankin, I.Zh Bunin, A.A. Oksogoev, Sinergetika i fraktaly v materialovedenii [Synergetics and Fractals in Materials Science]., Nauka Publ., Moscow, 1994. 383 pp. (In Russian). [11] A.I. Gordienko, M.L. Kheifetz, L.M. Kozhouro, et al., Combined Physico-Chemical Treatment: Synergetic Aspect, Technoprint Publ., Minsk, 2004. 200 pp. [12] V.N. Poduraev, Tekhnologiya fiziko-khimicheskikh metodov obrabotki [Technology of Physicochemical Processing]., Mashinostroenie Publ., Moscow, 1985. 264 pp. (In Russian). [13] Zh.I. Alferov, The history and future of semiconductor heterostructures, Semiconductors 32 (1) (1998) 114. [14] M.L. Kheifetz, Additivnye sinergotekhnologii posloynogo sinteza izdeliy iz kompozitsionnykh materialov pri vozdeystvii potokami energii [Additive synergetic technologies of layer-by layer synthesis of articles from composite materials when exposed to energy flows], Naukoyomkie tekhnologii v mashinostroenii [Adv. Technol. Mech. Eng.] 4 (58) (2016) 39 (In Russian). [15] S.A. Chizhik, M.L. Kheifetz, Sinergotekhnologii posloynogo sinteza izdeliy [Synergy technologies of layered synthesis products], Nauka i Innovatsii [Sci. Innov.] 2 (156) (2016) 1316 (In Russian). [16] M.L. Kheifetz, Analiz algoritmov proizvodstva izdeliy po modelyam samovosproizvedeniya von Neumann [The analysis of product manufacturing algorithms according to von Neumann’s self-reproduction models], Doklady Natsional’noi akademii nauk Belarusi [Rep. Natl. Acad. Sci. Belarus] 45 (5) (2001) 119122 (In Russian). [17] M.L. Kheifetz, Modeli i algoritmy proizvodstva izdeliy bez ispol’zovaniya formoobrazuyushchey osnastki [Models and algorithms for the production of articles without the use of forming equipment], Vestsi Natsyanal’nai akademii navuk Belarusi. Seryya fizika-technichnych navuk [Proc. Natl. Acad. Sci. Belarus. Physicotechnical series] (2) (2001) 5962 (In Russian). [18] P.A. Vityaz, V.S. Ivashko, E.D. Manoilo, Teoriya i praktika gazoplamennogo napyleniya [Theory and Practice of Gas Flame Spraying], Nauka i technika Publ, Minsk, 1993. 296 pp. (In Russian). [19] M.L. Kheifetz, L.M. Akulovich, Zh.A. Mrochek, E.Z. Zeveleva, Elektrofizicheskie i elektrokhimicheskie metody obrabotki materialov [Electrophysical and Electrochemical Methods of Materials Processing], The Polotsk State University Publ., Novopolotsk, 2012. 292 pp. (In Russian). [20] E.D. Eidel’man, Excitation of electric instability by heating, PhysicsUspekhi 38 (11) (1995) 12311246. [21] P.A. Vityaz, A.F. Ilyushchenko, M.L. Kheifetz, Operativnoe maketirovanie i proizvodstvo izdeliy slozhnoy formy iz kompozitsionnykh materialov [Rapid prototyping and manufacture of complex-shape articles from composite materials], Naukoyomkie tekhnologii v mashinostroenii [Adv. Technol. Mech. Eng.] (2) (2011) 38 (In Russian). [22] A.A. Shipko, I.L. Pobol, I.G. Urban, Uprochnenie staley i splavov s ispol’zovaniem elektronno-luchevogo nagreva [Strengthening of Steels and Alloys With the Use of Electron Beam Heating], Nauka i technika Publ, Minsk, 1995. 280 pp. (In Russian).
This page intentionally left blank
Index
Note: Page numbers followed by “f ” and “t” refer to figures and tables, respectively. A A20X in aerospace casting industry. See AluminumCopper (AlCu) based casting alloy A205 ABS. See Acrylonitrile butadiene styrene (ABS) Acoustic monitoring, 411412, 411f Acoustical sensors, 181 Acrylonitrile butadiene styrene (ABS), 6, 1617 Addaero Manufacturing, 328 Additive aerospace, 327 AM improving supply chain aerospace industry, 330331 competitive implications of AM aerospace, 340 factors driving AM in aerospace industry, 328329 geography of AM aerospace, 339 materials for AM, 331338 regulatory factors in AM aerospace, 338339 Europe, 339 United States, 339 3D printing technologies for tooling and prototyping, 328 Additive Factory concept, 327328 Additive manufacturing (AM), 13, 67, 9, 1217, 3334, 6768, 8889, 99, 187, 213, 237, 240, 283, 301, 327, 341, 375376, 376f, 385, 387388, 401, 419, 427, 436f. See also Aerospace additive manufacturing; Fusion additive manufacturing (Fusion AM) additive nonmetal technologies, 1417 aerospace requirements and opportunities, 712
AM-built components surface-finished by Extreme ISF Process, 361366 AMT, 1214 applications, 1721 direct digital manufacturing, 1819 rapid prototyping, 1920 rapid tooling, 19 repair, 2021 challenges, 169170 manufacturing limitations, 22 postprocessing realities, 2223 specification and standard development, 23 components in service, 401402 design requirements, 810 functional complexity, 9 property requirements, 10 structural design, 9 DMD technology, 3f fabrication of various material types, 36 and improved aircraft design, 329 improving supply chain aerospace industry, 330331 manufacturing capabilities and benefits, 1012 material economy, 11 part consolidation, 1011 small production runs and turnaround time, 12 material properties and defects, 404406 potential future applications, 2324 powder, 121 quality control for, 4043 risk, 54 software design software, 83
450
Additive manufacturing (AM) (Continued) as limiting factor on AM aerospace, 329 process software, 8384 specifications released and in-work, 2t surface anatomy, 342345 finishing, 355361 texture, 341 texture characterization, 345355 system resolution, 79 technology, 176 of TiAl applications, 235 EBM, 248256, 248f, 252f, 253f fundamentals, 235237 laser metal deposition, 240245 processings, 237240, 237f SEM, 245248 topological optimization for, 6871, 68f Additive Manufacturing Control Plan (AMCP), 5254 Additive Manufacturing Requirements (AMRs), 5859 Additive Manufacturing Standardization Collaborative (AMSC), 39, 402403 Additive Manufacturing Standardization Committee. See Additive Manufacturing Standardization Collaborative (AMSC) Additive materials certification of, 4748 qualification, 4447 quality control in, 4849 Additive metal technologies (AMT), 10, 1214. See also Additive nonmetal technologies in aerospace industry, 13t DED, 1213 PBF, 13 for repair, 1718, 2021 geometry restoration, 21 structural integrity restoration, 21 Additive nonmetal technologies, 1417. See also Additive metal technologies (AMT) in aerospace industry, 15t FDM, 1617 PolyJet, 16
Index
SLA, 1516 SLS, 14 Additive synergetic technologies of layer by layer synthesis, 432434 Additive technologies (AT), 266, 427 determination of changes in structure using samples, 9395 economic effect of application of, 88 MMCs fabrication via, 266268 Advanced metallographic techniques, 113140 background, 114117 chemical analysis, 136140 light optical microscopy, 122124 metallographic sample preparation, 117121 microstructural analysis, 135136 shape and texture analysis, 124135 Aerospace alloy, 311 enterprises, 428 industrial products, 427 materials and requirements, 1 sector, 7 Aerospace additive manufacturing. See also Additive manufacturing (AM) additive synergetic technologies, 432434 EBH of coated surface, 438445 ion implantation and ion deposition of coatings, 434438 system analysis of processing methods, 427432 Aerospace applications, 302. See also Selective laser melting (SLM) gradient 3D cubes, 271f methods and materials, 269270 MMCs fabrication via AT, 266268 on nickel alloy based, 268269 3D NiCrBSi nickel alloy cubes, 271f Aging of powders, 176 AIA. See Automated image analysis (AIA) Airbus, 12, 336, 340 Airbus 320 characteristics, 424t Airbus 380 characteristics, 424t Airbus Defence and Space, 331 Airbus Group, 329 Aircraft, 14 interiors, 337
Index
Airline passenger seats, 420f Airplane seat, 419 Altair, 329 Aluminum, 439 Aluminum alloys, 11. See also Selective laser melting (SLM) alloy and process design, 310320, 310f adaptation of existing high strength alloys, 316319 design of new alloys, 311316 development of composite materials, 319320 processingmicrostructureproperty considerations, 302309 AluminumCopper (AlCu) based casting alloy A205, 318319 Aluminumsilicon (AlSi) based alloy system, 306307 AM. See Additive manufacturing (AM) AMCP. See Additive Manufacturing Control Plan (AMCP) American National Standards Institute (ANSI), 39 American Society of Mechanical Engineers (ASME), 39 American Society of Testing and Materials (ASTM), 3637, 402403 B417 Standard, 104105 EV31A. See Elektron 21 alloys International, 39 WE43C. See Elektron 43 alloys American Welding Society (AWS), 39 AmpliFORGE, 77 AMRs. See Additive Manufacturing Requirements (AMRs) AMSC. See Additive Manufacturing Standardization Collaborative (AMSC) AMT. See Additive metal technologies (AMT) Annealing heat treatment, 286 ANNs. See Artificial neural networks (ANNs) ANSI. See American National Standards Institute (ANSI) Apparent density, 175176 and flow, 101105 of metal powder, 104 APWorks, 331
451
Arcam AB, 327 Arcam Q10 Plus EBM system, 13 ArianeGroup, 168 Aristo Cast, 424 Artificial neural networks (ANNs), 391 ASME. See American Society of Mechanical Engineers (ASME) ASTM. See American Society of Testing and Materials (ASTM) AT. See Additive technologies (AT) Atomization process, 250 Autodesk, 329, 424 Netfabb software, 419 Autodesk Project Escher technology, 419422 Automated image analysis (AIA), 108, 122124 testing programs, 113114 Automated repair process, 23 AWS. See American Welding Society (AWS) B Backscattered electron images, 294f Backscattered electron imaging (BEI), 117, 137, 138f Balling, 405406 BCC structure. See Body-centered cubic structure (BCC structure) BD technologies. See Bed deposition technologies (BD technologies) Bed deposition technologies (BD technologies), 432 BEI. See Backscattered electron imaging (BEI) Belgium-based additive factory, 335 Benard convection cells, 439 Binder jetting, 36, 1314 Body-centered cubic structure (BCC structure), 163 Boeing, 7 787 Dreamliner, 18 Boron, 170171 Brittle temperature range (BTR), 170171 BTR. See Brittle temperature range (BTR) Buoyancy force, 443444 Buy-to-fly ratios, 11
452
C CAD. See Computer-aided design (CAD) Camera-based process monitoring, 180181 Canfield analysis, 190 Carbides of chromium (CrxCy), 265 Carbon, 170171 emissions, 419 nanotubes, 320 Carney test for flow, 101103 Casting process, 237238, 419422 CAVF. See Chemically accelerated vibratory finishing (CAVF) Cell culture medium machine, 439, 440f Cellular automata, 383384 Cellular structure, 438 Center for Innovative Materials and Processing-3D (CIMP-3D), 39 Centrifugal casting process, 237238 Certification, 4344. See also Qualification and certification (Q&C) of additive materials, 4748 CFR. See Code of Federal Regulations (CFR) Chemical analysis, 136140 composition, 174175, 265266 etchants, 135t etching. See Chemical milling process pickling. See Chemical milling process polishing. See Chemical milling process Chemical milling process, 360361 Chemically accelerated vibratory finishing (CAVF), 361 Chromium atoms, 435 Chromium carbide (Cr3C2), 268 CIJ. See Continuous inkjet (CIJ) CIMP-3D. See Center for Innovative Materials and Processing-3D (CIMP3D) Circularity, 130 CLIP. See Continuous DLP (CLIP) “Co-Design to Target” industry solution, 329, 331 Coated surface, EBH of, 438445 cellular structures, 443444 configurations of cellular machines, 442f distribution of microhardness, 444f model of formed multicomponent coating on peaks, 443f
Index
state graph of technological medium cell automaton, 441f Coatings, ion implantation and ion deposition of, 434438, 437f Code of Federal Regulations (CFR), 4748 Cold shuts, 97 Cold spray. See Supersonic particle deposition (SPD) Cold-cracking. See Strain-age cracking (SAC) COMAC, 334 Combustion chamber, 165166 Commercial aviation gas turbine engines, 3334 Commercially pure titanium (CP titanium), 139 Compact preform, 93 Compactness, 130 Composites, 337338 materials development for SLM, 319320 Computed tomography (CT), 3637, 108, 136 Computer-aided design (CAD), 83, 187188 Concept Laser, 327 Consumable material, 217218 Contact acoustic emissions sensors, 411412 Contact profilometry, AM surface characterization by, 349352 Continuous DLP (CLIP), 328 Continuous inkjet (CIJ), 16 Contour scan effect on surface texture, 290293 Contour zone, 250 Conventional crystallographic fatigue crack initiation mechanisms, 38 Conventional manufacturing, 3336, 4445 Conventional methods, 67 Conventional processing methods, 283 Cooling technologies, 165166 CP titanium. See Commercially pure titanium (CP titanium) Crack modeling, 189190 Crack susceptibility of high strength nickel superalloys, 169 Cracking mechanisms, 169 Crystallographic texture, 304 CT. See Computed tomography (CT)
Index
D D&DT community. See Durability and damage tolerance community (D&DT community) D10 particles, 175 D50 particles, 175 D90 particles, 175 Dassault Syste`mes’ 3DEXPERIENCE platform, 329 DD technique. See Direct deposition technique (DD technique) DDC. See Ductility dip cracking (DDC) DDM. See Direct digital manufacturing (DDM) DDT&E. See Design, development, test and evaluation (DDT&E) DED. See Directed energy deposition (DED) Defect data correlation automated, future closed-loop control possibilities, 182184 Dendritic cell boundaries, 90 Dendritic structure of nickel matrix, 275 Density optimization, 145149 Department of Defense (DoD), 38, 402403 AM Roadmap, 406 Deposition process, 217f, 230 Design, development, test and evaluation (DDT&E), 3334 Design for additive manufacturing (DFAM), 67 Detonation coating process, 438 DFAM. See Design for additive manufacturing (DFAM) DHA. See Dispersion-hardened alloys (DHA) DIA. See Dynamic image analysis (DIA) Diffusion processes, 442 Direct deposition technique (DD technique), 432 Direct digital manufacturing (DDM), 14, 1718 direct metal part fabrication, 1819 fixtures and accessories, 19 Direct metal depositions (DMDs), 88, 95, 266267 processes, 22 systems, 18 testing of mechanical properties of samples of parts, 9597
453
Direct metal laser sintering (DMLS), 12 Direct metal part fabrication, 1819 Direct rapid tooling, 1718 Directed energy deposition (DED), 13, 3f, 10, 1213, 52, 99, 214, 266267, 333334, 409410 Dispersion-hardened alloys (DHA), 265266 Dissipative structure formation, 439 Distortion in fusion AM, 189190 Distribution analysis of processing methods, 428431 DMDs. See Direct metal depositions (DMDs) DMLS. See Direct metal laser sintering (DMLS) “Do no harm” Class C designation, 56 DoD. See Department of Defense (DoD) DOD. See Drop on demand (DOD) Down-skin surfaces, 284285 Dreamcatcher generative design tools, 329 Drop on demand (DOD), 16 Ductility dip cracking (DDC), 173 powder material properties, 174177 Durability and damage tolerance community (D&DT community), 5960 Dynamic image analysis (DIA), 110111 E EASA. See European Aviation Safety Agency (EASA) Easton and St. John model, 390 EBAM. See Electron beam additive manufacturing (EBAM) EBF3. See Electron beam free-form fabrication (EBF3) EBH. See Electron-beam heating (EBH) EBM. See Electron beam melting (EBM) EBSD maps, 388, 389f EBW. See Electron beam welding (EBW) Eddy current (ET), 3637 EDM. See Electrical discharge machining (EDM) EDS. See Energy dispersive spectroscopy (EDS) EDX. See Energy dispersive analysis (EDX) EIGA. See Electrode induction melting GA (EIGA) ELB-Schliff’s millGrind, 19
454
Electrical discharge machining (EDM), 81 Electrode induction melting GA (EIGA), 99 Electromagnetic surfacing, 434 Electromagnetic surfacing process, 433 Electron beam, 215217 Electron beam additive manufacturing (EBAM), 4950, 7576 Ti6Al4V macrostructure, 76, 77f Electron beam free-form fabrication (EBF3), 12 Electron beam melting (EBM), 12, 99, 341. See also Selective laser melting (SLM) technology, 240, 241f Ti48Al2Cr2Nb honeycomb, 257f of TiAl, 248256, 248f, 252f, 253f creep properties, 255t element mappings by microprobe measurements, 249f fatigue crack initiating from ceramic particle, 256f Haigh diagram, 256 physical effects during selective melting, 250f room temperature tensile properties of EBM TiAl-4822, 254t tensile properties vs. temperature, 255f Electron beam power, 215216 Electron beam welding (EBW), 12 Electron-beam heating (EBH), 438445 Electropolishing, 360361 Elektron 21 alloys, 424425 Elektron 43 alloys, 424425 Energy density function, 144 energy-dissipation simulation, 436437 fluxes, 427 system analysis of processing methods, 427432, 429t, 430f source for AM, 301 Energy dispersive analysis (EDX), 270 Energy dispersive spectroscopy (EDS), 110 Entrained gas, 143 EOS additive manufacturing technology, 169 M 290 DMLS system, 18 EOSTATE Exposure OT, 180181 MeltPool, 180, 181f, 182
Index
ET. See Eddy current (ET) Etihad Airways Engineering, 336 European Aviation Safety Agency (EASA), 34, 336 Extreme ISF Process AM-built components surface-finished by, 361362 analysis of surface texture parameters, 365366 examples of surface treated by, 362364 improvement of mechanical properties, 364365 F FAA. See Federal Aviation Administration (FAA) FAA Q&C approach. See Federal aviation administration Q&C approach (FAA Q&C approach) Fabrication processes, 60, 82, 411 Face-centered cubic structure (FCC structure), 163 Fasteners, 10 Fatigue crack initiation, 286 performance, 307308 of selective laser melted Ti6AL4V, 286 properties, 293296, 295f, 296f, 297f testing, 288, 288f FCC structure. See Face-centered cubic structure (FCC structure) FDM. See Fused deposition modeling (FDM) FE model. See Finite element model (FE model) Feature cross sections, 127 Federal Aviation Administration (FAA), 34, 401403, 419 Advisory Circular 25.5711D, 60 Advisory Circular 33.701, 6061 Federal aviation administration Q&C approach (FAA Q&C approach), 6061. See also General electric Q&C approach Feedstock attributes, 36 FG structures. See Functional-gradient structures (FG structures)
Index
FGMs. See Functionally graded materials (FGMs) Filtering function, 346347 Finite element model (FE model), 69 Fitting function, 345346 Flammability, 424425 Flowability of powder, 103 Fortus 450 system, 335 Fortus 900mc system, 335 Four-dimension (4D) printing, 23 Fractography, 7778, 151152, 156 Fracture-critical hardware, special considerations for, 3738 Free-form design, 6869, 83 Friction stir additive manufacturing (FSAM), 191192, 192f laser preheating, 197f mechanical properties, 196t of P92 steel, 194f Friction stir processing (FSP), 191192 aluminum deck lid FSP joined to galvanized steel, 199f damage repair, 200f FSP/FSAM cladding, 196f key benefits, 193t PTA in hybrid couple with, 197f Friction stir welding (FSW), 191192 FSAM. See Friction stir additive manufacturing (FSAM) FSP. See Friction stir processing (FSP) FSW. See Friction stir welding (FSW) Functional complexity, 9 Functional-gradient structures (FG structures), 268269 Functionally graded materials (FGMs), 10, 267 Fused deposition modeling (FDM), 14, 1617, 330, 336 Fusion, 384385 Fusion additive manufacturing (Fusion AM), 187188. See also Additive manufacturing (AM) cylindrical bar preform in custom titanium alloy, 205f defects, 189f experimental examples, 198210 foams produced from fusion PTA AM processing, 207f friction stir machine in operation, 192f
455
large columnar microstructure in, 189f Mg-B4C cermet microstructures, 205f surface at high build rates, 200f tensile values of samples extracted from joint regions, 199t TiB2/Ti cermet armor after hit by APM2 muzzle velocity, 206f TiB2/Ti composite integrally built on Ti6Al4V base, 206f Fusion AM. See Fusion additive manufacturing (Fusion AM) G G/R ratio. See Thermal gradient/ solidification rate ratio (G/R ratio) GA. See Gas atomization (GA) Gamma phase, 163 Gap analysis, 3943 Gas atomized powder, 116 gas-discharge electron beam guns, 216 gas-phase deposition, 433434 materials in gas turbines, 164169 pores, 303 GAs. See Genetic algorithms (GAs) Gas atomization (GA), 99, 256 Gaussian filter, 346347 GE. See General Electric (GE) GE Avio Aero, 11, 327328 GE90 jet engines, 34 Geared-Turbofan engine (GTF engine), 235 General Electric (GE), 34, 327, 340 Aviation, 1011 General electric Q&C approach, 4349 certification of additive materials, 4748 qualification of additive materials, 4447 quality control in additive materials, 4849 Generalities superalloys, 163, 164f defect data correlation automated, future closed-loop control possibilities, 182184 material challenges processing perspectives, 169173 challenges with additive manufacturing, 169170 DDC, 173 liquation cracking, 171172 SAC, 172173
456
Generalities superalloys (Continued) solidification cracking, 170171 materials in gas turbines, 164169 process monitoring, 177179 challenges with tradition postprocess inspection techniques, 179 novel QA approaches, 179 quality assurance in AM, 177178 quality assurance tie-in, 181182 sensor types for in situ process monitoring, 179181 Genesis, 9 Genetic algorithms (GAs), 391 Geometry restoration, 21 Gesellschaft fu¨r Elektrometallurgie mbH (GfE), 238239 Gettering, 385386 GfE. See Gesellschaft fu¨r Elektrometallurgie mbH (GfE) Glass-filled nylon, 14 Government agency approaches, 5161 GR plots, 383384 Grade 5 alloy parts, 93, 97 Grinding, 120121 GTF engine. See Geared-Turbofan engine (GTF engine) H Hall flowmeter and cup, 103f, 104105 Hall-Petch effect, 76 “Hard-to-weld” alloys, 170 Hastelloy X, 168 Hatch distance, 144, 147148 Hatch zone, 250 Hausner ratio, 105106 HCF. See High-cycle fatigue (HCF) Healing effect, 182184 Heat accumulation effect, 283284 diffusion phenomena, 251 heat-resistant nickel alloys, 87 heat/mass transfer, 444 transfer, 284285 treatment, 286 Heat affected zone (HAZ). See Liquation cracking High strength alloys, 310 High-cycle fatigue (HCF), 286 High-energy-density processing, 431
Index
High-intensity processes, 434 High-strength thermoplastics, 1617 HIP. See Hot isostatic pressing (HIP) HIPed. See Hot isostatically pressed (HIPed) HIPing. See Hot isostatic pressing (HIPing) Hollow conical electron beam, 216, 217f on consumable wire, 218f Hot corrosion, 163 Hot isostatic pressing (HIP), 36, 7778, 81, 266, 304 Hot isostatic pressing (HIPing), 9596, 286, 360 Hot isostatically pressed (HIPed), 239, 288 Hybrid dissimilar titanium alloy combinations, 198199 Hybrid manufacturing systems, 19 Hydroflouric acid, 136 Hydrogen, 143 pores, 303 Hypereutectic composition, 311312 I ICME. See Integrated computational materials engineering (ICME) ICP-OES, 177 IGF. See Inert gas fusion (IGF) Image acquisition and processing, 129 Image analysis method, 145 In situ characterization, 406 In-process, 177 In-situ process, 7980 Inclination angle and processing parameters, 289290 Inclusion analysis, 108110 Inconel 718 alloy, 73f INCONEL 718, 168 Indirect rapid tooling, 1719 Industrial 3D printing, 213 Industry Q&C approaches general electric Q&C approach, 4349 government agency approaches, 5161 Lockheed martin Q&C approach, 4951 Inert gas fusion (IGF), 177 Inspection processes, 37 Integrated computational materials engineering (ICME), 37, 207210, 375, 377f, 378f fusion process model, 378t limitations, 393395
Index
predicting microstructure, 387391 predicting properties and performance, 391393 process modeling, 379384 solute profiles in TixW system, 384f solidification partitioning, 386387 solute loss or pickup, 385386 Integrated structural integrity rationale, 5657 Interfacial area ratio, 366 “Intermediate” parameter set, 286287 Intermetallic compounds, 442 International Organization for Standardization (ISO), 39, 403 Interstitialcy defects, 435 Ion deposition of coatings, 434438 Ion implantation, 433438 Ion-vacuum processing, 438 ISO. See International Organization for Standardization (ISO) J Japanese Space Exploration Agency (JAXA), 51, 56 Jet engine components and materials, 166f Joint effect, 433434 K Karl Fischer Titration, 107 Kinetic Monte Carlo, 383384 L L-PBF. See Laser-Powder Bed Fusion (LPBF) Lack of fusion (LOF), 405406 Lamellar structure, 443 Langmuir equation, 385386 Laser diffraction, 175 laser-material interaction, 303 polishing/remelting technology, 285 power, 144, 147148 systems, 385 ultrasonics, 409410 Laser cladding (LC), 266267 Laser engineering net-shaping (LENS), 12 Laser melting (LM), 267, 341 refinement, 153156 fractography, 156
457
selection of optimized heat treatment parameters, 153 selection of optimized laser parameters, 153154 tensile properties, 155 Laser melting parameters (LMPs), 153154 Laser metal deposition (LMD), 12, 240, 244t of TiAl, 240245, 241f, 243f Laser powder bed fusion (LPBF), 403404 Laser sintering (LS), 341 Laser-Powder Bed Fusion (L-PBF), 3637, 45, 52 AM process, 5254 Lattice Boltzmann method, 249 Layer-by-layer building principles, 304 synthesis, 432434, 433t LC. See Laser cladding (LC) LENS. See Laser engineering net-shaping (LENS) LENS Blisk Repair Solution system, 21 Life-Limited Parts (LLPs), 60 Light optical microscopy (LOM), 108, 122124 Light-emitting device, 1516 Liquation cracking, 171172, 301302 Liquid bath dimensions, 89, 90f epoxies, 118119 phase, 442 photopolymer resin, 1516 LLPs. See Life-Limited Parts (LLPs) LM. See Laser melting (LM) LMCO. See Lockheed Martin Company (LMCO) LMD. See Laser metal deposition (LMD) LMPs. See Laser melting parameters (LMPs) Lockheed Martin Company (LMCO), 4951 Lockheed Martin Q&C approach, 4951 LOD. See Loss on drying (LOD) LOF. See Lack of fusion (LOF) LOM. See Light optical microscopy (LOM) Long-wave high pass cut-off filter, 346347 nesting index filters, 352353 Loss on drying (LOD), 107, 177 device for measuring moisture content of metal powders, 108f Low pressure turbine (LPT), 235
458
LPBF. See Laser powder bed fusion (LPBF) LPT. See Low pressure turbine (LPT) LS. See Laser sintering (LS) M Machine learning, 182 Magnesium, 423 alloys, 424425 Marangoni Effect, 381 Marshall Space Flight Center (MSFC), 43 MSFC-STD-3716, 52 additive manufactured part classification system, 55f Technical Standard and Specification, 5254 Mass distribution in Seats, 419, 421f finishing techniques, 356357 Material choice, 166 ductility, 173 economy, 11 engineering, 1 extrusion. See Fused deposition modeling (FDM) jetting. See PolyJet properties and defects in AM, 404406 Materialisation, 9 MATLAB, 145 Mechanical anisotropy, 309 Mechanical polishing methods, 285 Meshify, 9 MET. See Metrology (MET) Metal additive manufacturing design. See also Additive manufacturing (AM) cost considerations, 8182 design software, 83 methods and approaches, 6874 part consolidation, 7172 part integration and repair, 7273 topological optimization, 6871, 68f process aspects of design, 7481 part performance, 7478 process software, 8384 product and process design tools, 8284 Metal additive manufacturing processes, 17t, 33, 187188. See also Additive manufacturing (AM)
Index
current Q&C state-of-the-art and gap analysis, 3943 FAA-approved T25 compressor inlet temperature sensor and fuel nozzle, 35f guidance in process qualification or feedstock specification FAA Q&C approach, 6061 warnings, 5960 industry Q&C approaches, 4351 special considerations for fracture-critical hardware, 3738 Metal injection molding (MIM), 99 Metal matrix composites (MMCs), 265, 270f, 318319 fabrication via AT, 266268 on nickel alloy, 268269 Metal powder bed AM systems, 329 fusion technology, 331, 333 Metal Powder Industries Federation, 39 Metal(s), 11, 331333 sintering processes, 22 3D printers, 218219 Metallic Materials Properties Development and Standardization (MMPDS), 43, 403404 Metallographic sample preparation, 117121 grinding and polishing, 120121 mounting, 117119 sampling, 117 Metallographic studies, 93 Metrology (MET), 3637 Microhardness distribution, 444f, 445 Microstructure analysis, 135136, 436 coarsening, 150 dependence of powder particles, 92 prediction, 387391 MIM. See Metal injection molding (MIM) MMCs. See Metal matrix composites (MMCs) MMPDS. See Metallic Materials Properties Development and Standardization (MMPDS) Moisture analysis, 106108 Mold pattern fabrication, 16 Molybdenum-depleted zones, 439
Index
Mounting, 117119 MPIF Standard 03, 101103 MPIF Standard 04, 104105 MPIF Standard 05, 100101 MPIF Standard 28, 101105 MPIF Standard 32, 100 MPIF Standard 46, 105 MPS, 5254 MSFC. See Marshall Space Flight Center (MSFC) Multifunctional structures, 23 Multilayer deposition process, 240241 Multimaterial DED processes, 23 Multiply localized energy adsorption zones, 433 N NADCAP. See National Aerospace and Defense Contractors Accreditation Program (NADCAP) Nanoparticles, 266 Nanoscale inclusions, 266 NASA. See National Aeronautics and Space Administration (NASA) NASA Procedural Requirement (NPR), 5758 National Aeronautics and Space Administration (NASA), 3334, 402 Q&C approach, 3637, 5160 qualification requirements additional guidance, 59 additive manufactured part categories, 5456 general requirements, 5657 industry standards, 59 influence of mission classification, 5758 integrated structural integrity rationale, 5657 process specifications, 59 procurement specifications, 59 tailoring approach, 5758 National Aerospace and Defense Contractors Accreditation Program (NADCAP), 4043 National Institute of Standards and Technology (NIST), 3637 NDE. See Nondestructive evaluation (NDE) Near-net-shape PM method, 239
459
Netfabb software, 9, 419 Next generation generative software, 329 Nickel alloy MMCs on, 268269 powders, 89 surface structure, 91f Nickel sublayer, 441 Nickel superalloys, 163, 166167 Nickel-based super alloys, 11 disc, 235 Niobium, 170171 alloys, 224 NIST. See National Institute of Standards and Technology (NIST) Nitinol, 174175 Nitrogen-based SPD, 14 Non-hardening phases, 314315 Nonconsumable FSAM tool, 191192, 191f Noncontact profilometer, 352355 Nondestructive evaluation (NDE), 3637, 413 AM components in service, 401402 clustering of acoustic metrics, 412f emerging methods, 409413 material properties and defects in AM, 404406 optical and thermal monitoring, 406409 post-production inspection, 409 practical considerations, 413414 regulatory actions and standardization, 402404 Nondestructive inspection, 7980 Nonequilibrium technological process, 428 Nonmetal additive manufacturing technologies, 17t Nonoptical sensor technologies, 181 Nonspherical particles, 176 Nonsteady-state temperature field, 445 Nonweldable alloys, 170 Novel QA approaches, 179 NPR. See NASA Procedural Requirement (NPR) Numerical algorithms, 70 Nylon 12, 6, 14 O OCT. See Optical coherence tomography (OCT)
460
OEM. See Original equipment manufacturer (OEM) OffAxis sensors, 179 OM. See Optical metallography (OM) OmniSurf from Digital Metrology, 351352 OnAxis sensors, 179 Open atmosphere systems, 79 Operating gas, 215216 Optical coherence tomography (OCT), 406407, 413 Optical in situ process monitoring systems for AM, 179181 camera-based process monitoring, 180181 photodiode-based process monitoring, 180 Optical metallography (OM), 272, 272f, 273f Optical monitoring, 406409 Optimum linear energy density, 283284 OptiStruct, 9 Optomec LENS, 385 Original equipment manufacturer (OEM), 37 Overhanging surfaces. See Down-skin surfaces Oxidation resistance, 163 P Packing characteristics, 101 Part categories, 51 Part consolidation in AM, 7172 Part integration and repair, 7273 Part performance in AM, 7478 defects, 77 mechanical properties, 7778 microstructure, 7477 part evaluation, 7980 part quality, 7879 post processing, 8081 Part production controls, 54 Part production plan (PPP), 54 Particle size and distribution, 99101 Particle size distribution (PSD), 175 PBF. See Powder bed fusion (PBF) PBM. See Powder bed method (PBM) PCRT. See Process compensated resonance testing (PCRT) pdfs. See Probability distribution functions (pdfs) Penetrant testing (PT), 3637 “Performance” parameter set, 286287
Index
Personal protection equipment (PPE), 136 PF. See Powder forging (PF) Phase field modeling, 383384 Phosphorus, 170171 Photodiode-based process monitoring, 180 Photodiode-high speed camera hybrid system, 406407 Photopolymer resins, 1516 Photopolymerization based stereolithography, 328 Plasma beam, 200201 Plasma rotating electrode process (PREP), 99, 111112 Plasma transferred arc (PTA), 200201 AM system, 202f large vertical and robotic PTA system, 203f plasma beam in AM build, 201f PM. See Powder metallurgy (PM) Point-based measurement systems, 406 Polished cross saections, 135, 135f Polishing, 120121 Polycarbonate, 1617 PolyJet, 14, 16, 20, 23, 328 Polylactic acid, 1617 Polymerization, 36 Polymers, 334336 AM, 7 bed fusion, 336337 Polyphenylsulfone, 1617 Porosity, 111112, 383 Post processing, 8081 Post selective laser melting heat treatment, 150152 fractography, 151152 optimized HTPs applied on tensile samples, 150t tensile properties, 150151 Post-production inspection, 409 Postprocessing, 19, 36, 177178 NDE, 80 parameter refinement, 153156 realities, 2223 techniques, 38, 150 Postselective laser melting heat treatment of fatigue samples, 288 surface treatment, 285 Powder characteristics, 303304
Index
flow, 101 flowability, 175 layer thickness, 144 packing, 104 powder-based laser AM, 301 structure evaluation of powder particles of different sizes, 8992 Powder bed fusion (PBF), 13, 10, 1213, 266267, 301 process models, 381383 Powder bed method (PBM), 341 Powder forging (PF), 113 Powder metallurgy (PM), 38, 87, 113, 235, 239, 240f PM-PF water atomized powders, 115 PPE. See Personal protection equipment (PPE) PPP. See Part production plan (PPP) Precipitation hardening alloys, 316 Premium Aerotech, 327328 PREP. See Plasma rotating electrode process (PREP) Preprocess, 177 Primary profile, 346347 Primary roughness, 344 Principal structural elements (PSEs), 60 Print services providers, 330 Probability distribution functions (pdfs), 393 Process control methodologies, 3436 map models, 383384 monitoring, 182 specifications, 59 Process compensated resonance testing (PCRT), 3637 Processingmicrostructureproperty considerations, 302309, 306f Procurement specifications, 59 Profile electron beam 3D metal printing, 215f cup made from CP Ti Grade 4 wire, 225f deposited bead depending on deposition parameters, 224f deposited layers formed from wire, 225f deposition process, 217f formation of walls of different thicknesses, 227f gap Z between wire exit hole and substrate, 223f
461
macrostructure of thick wall, 228f results of mechanical testing of Ti6Al4V, 229t scheme of system for running of invented process, 214f spreading of liquid metal, 219f Ti6Al4V specimen structure, 226f, 227f wire as substrate for deposition, 222f PSD. See Particle size distribution (PSD) PSEs. See Principal structural elements (PSEs) PT. See Penetrant testing (PT) PTA. See Plasma transferred arc (PTA) Q Q&C. See Qualification and certification (Q&C) QMP. See Qualification metallurgical process (QMP) QMS. See Quality management system (QMS) Qualification of additive materials, 4447 processes, 4344 Qualification and certification (Q&C), 3334, 39, 4546 state-of-the-art and gap analysis, 3943 directions, and quality control for additive manufacturing, 4043 standardization gaps, 3940, 41t Qualification metallurgical process (QMP), 5254 Quality assurance in AM, 177178 tie-in, 181182 Quality control for additive manufacturing, 4043 in additive materials, 4849 measurements, 99112 apparent density and flow, 101105 inclusion analysis, 108110 moisture analysis, 106108 particle size and distribution, 99101 porosity, 111112 shape factor, 110111 tap density, 105106, 106f Quality management system (QMS), 40, 54 Quantitative characterization techniques, 402
462
R R&D. See Research and development (R&D) Ra value, 289290, 289f of top surface of SLMed cubic samples, 283284 Radiographic testing, 3637 Radius function evaluation, 131, 132f Rapid prototyping, 1720 Rapid tooling, 1719 Rare earth (RE), 302 Reconfigurable tooling, 19 Regulatory actions and standardization, 402404 RepAIR project, 23 Research and development (R&D), 39 Resonance type testing, 405406 RMS. See Root-mean-squared (RMS) Root-mean-squared (RMS), 411412 Rough surface, 343 Roughness. See also Surface roughness component, 346347 sampling length, 347349 S SAC. See Strain-age cracking (SAC) Sacrificial metal or material, 356357 SAE. See Society for Automotive Engineers (SAE) Sampling, 117, 174 SAP, 331 SAW. See Surface acoustic wave (SAW) Scan speed, 144, 147148 Scandium (Sc), 311 Scanning electron microscope (SEM), 110, 270, 275f, 290, 291f, 292f, 293f SDOs. See Standards development organizations (SDOs) Secondary dendrite arm spacing (SDAS), 312 Secondary electron images (SEI), 116 SEI. See Secondary electron images (SEI) Selective laser melted Al-7Si-0.6Mg alloy post selective laser melting heat treatment, 150152 refinement of laser melting and postprocessing parameters, 153156 selective laser melted Al alloy A357, 143149
Index
density optimization, 145149 optimized laser processing parameters, 149t process control, 143144 relative vs. energy density, 148f 3D plots and 2D processing windows, 147f Selective laser melted metallic components, 283285 Selective laser melted Ti6Al4V alloy experimental procedure fatigue testing, 288 material, 286 postselective laser melting heat treatment of fatigue samples, 288 specimens, 286287, 287f, 287t, 288t surface roughness measurements, 288 fatigue performance, 286 fatigue properties, 293296 postselective laser melting surface treatment, 285 surface roughness, 289293 of SLM metallic components, 283285 Selective laser melting (SLM), 12, 88, 95, 99, 143, 240, 266267, 283, 284f, 301. See also Aerospace applications; Electron beam melting (EBM) adaptation of existing high strength alloys, 316319, 316f, 317f alloys in, 302309 composite materials development for, 319320 design of new alloys, 311316, 314f SLM-processing parameters, 145 SLMed aluminum samples, 143 testing of mechanical properties of samples of parts, 9597 of TiAl, 245248 produced TNM-B1 3D-dodecahedron structures, 246f SEM pictures of SLM producing TNM sample, 246f typical microstructures SEM of samples, 247f Selective laser sintering (SLS), 36, 14, 19 Self-organization, 432, 434435 SEM. See Scanning electron microscope (SEM) Semiatin’s approach, 385386
Index
Sensor acoustical, 181 contact acoustic emissions, 411412 OffAxis, 179 types for in situ process monitoring, 179181 optical in situ process monitoring systems for AM, 179181 Service providers, 330 SHAIK, 166167, 167f Shape analysis, 124135 Shape factor, 110111 Short or low-pass cut-off filter, 346347 Short-wave nesting indexes filters, 352353 SHT. See Solution heat treatment (SHT) Siemens, 330331 Sieve analysis, 100101, 175 for AM techniques, 102t Signal processing methods, 184 Silicon (SiC), 265 Simpleware, 9 Single “deposition engine tool”, 333334 Single-focused energy adsorption zones, 432 Sintavia, 328 SLA. See Stereolithography (SLA) SLM. See Selective laser melting (SLM) SLS. See Selective laser sintering (SLS) Slurry ball milling method, 320 Society for Automotive Engineers (SAE), 39 Solid alloys, 265266 Solid-solution strengthening, 164 Solid-state cracking. See Strain-age cracking (SAC) Solid/liquid interface, 444 Solidification, 311 cracking, 170171 partitioning, 386387 solute partitioning, 387f Solute loss or pickup, 385386 Solution heat treatment (SHT), 143 Spaceflight rocket engines, 3334 Spatially resolved acoustic spectroscopy (SRAS), 395, 409410, 410f, 413 SPC. See Statistical process control (SPC) SPD. See Supersonic particle deposition (SPD) Specialized service bureaus, 328 “Speed” parameter set, 286287 Spherical particles, 176
463
SR. See Stress-relief (SR) SRAS. See Spatially resolved acoustic spectroscopy (SRAS) ST-130TM material, 338 Stainless steel, 99 Staircase effect, 284 Staircase feature, 343 Standardization gaps, 3940, 41t regulatory actions and, 402404 standards structure approved by ASTM F42 and ISO TC261, 403f Standards development organizations (SDOs), 39 Statistical process control (SPC), 5254 Stereolithography (SLA), 1416 Strain hardening effects, 307 Strain-age cracking (SAC), 172173 Stratasys FDM technology, 336 industrial thermal extrusion technology, 338 for sacrificial composite tooling, 338 Fortus 900 system, 336 ULTEM 1010 resin, 338 Stress concentration, 431 Stress-relief (SR), 150 Stress-rupture strength of different alloys, 164, 165f Structural integrity restoration, 21 Structure formation in AM determination of changes in structure, 9395 microstructure dependence of powder particles, 92 structure evaluation of powder particles of different sizes, 8992 testing of mechanical properties of samples of parts, 9597 Substantial effect, 434435 Substrate temperature, 144 Sulfur, 170171 Superalloys, 7273, 99 Supersonic particle deposition (SPD), 12, 14, 21 Supersonic particle deposition (SPD), 14 Surface accuracy formation, 431 Surface acoustic wave (SAW), 395 Surface defects, 358
464
Surface finishing of AM components, 355361 basics, 356358 Surface oxidation induced phenomenon, 255 Surface roughness, 303, 344 effect of contour scan on surface texture, 290293 measurements, 288 parameter, 357 Ra vs. inclination angle and processing parameters, 289290 of selective laser melted metallic components, 283285 Surface self-organization phenomena, 427 Surface texture characterization of AM components, 349355 best practices for, 345355 contact profilometry, 349352 noncontact profilometer, 352355 surface texture review, 345349 Surface-distributed energy adsorption zone, 433 Surface-sensitive NDE techniques, 38 Synergetic concept, 432 T Tailoring approach, 5758 Tap density, 105106, 106f, 175176 Texture analysis, 124135 Thermal gradient/solidification rate ratio (G/ R ratio), 305 Thermal monitoring, 406409 Thermal warping effect, 284285 Thermocapillary force, 443444 Thermographic imaging process, 407408 Thermogravimetric analysis, 107 Three-dimension (3D) CAD digital model, 8081 LC, 266267 model, 213 objects, 187188 optical surface metrology, 352 particle, 116 printers, 8788 printing, 8081 technologies for tooling and prototyping, 328 TiAl alloy (TNM), 235 TiB2. See Titanium diboride (TiB2)
Index
TiC. See Titanium carbides (TiC) Titanium, 89, 99, 164, 224, 385, 439 surface structure, 91f Titanium alloy (Ti6Al4V), 11, 13, 443, 445 effects in, 87 Titanium aluminide (TiAl), 11, 235 applications, 235 EBM, 248256 fundamentals, 235237 laser metal deposition, 240245 processings, 237240, 237f AM, 240 casting, 237238 PM, 239 wrought processing, 238239 SEM, 245248 Titanium carbides (TiC), 265, 268 Titanium diboride (TiB2), 268 TM. See Transition metal (TM) TNM. See TiAl alloy (TNM) Tool steel powders, 99 Topology optimization, 340 for AM, 6871, 68f FE topological optimization of bracket, 70f methods, 23 tools, 9 Tradition postprocess inspection techniques, challenges with, 179 Traditional subtractive manufacturing, 19 Transformation behavior, 236 Transition metal (TM), 302 Transmitted light technique, 117 True profile. See Primary profile Tungsten (WC), 265 Turbine blades, 165 Two-dimension (2D) non-contact profile, 352 surfaces, 113114 U ULTEM, 1617 Ultimate tensile strength (UTS), 244, 360361 Ultrafine structures, 267268 Ultrasonic testing (UT), 3637 Unmanned aerial vehicle (UAV), 1617 Up-skin surfaces, 284
Index
US Food and Drug Administration Federal Food, Drug, and Cosmetic Act, 4748 V Vacuum-ion modification, 436437 Vaporization, 385386 Vat photopolymerization. See Stereolithography (SLA) Volumetric techniques, 409 Voluntary consensus standards (VSC), 39 VT20 single-phase pseudo-α alloy, 439 VT22 alloy, 92 W WAAM. See Wire arc AM (WAAM) Warping effect, 284285 Waviness component of surface texture, 346347 WC. See Tungsten (WC) Weight loss on drying, 107 Weight reduction, 328329 Weld cracking, 170
465
Weld-centerline cracking. See Solidification cracking Weldability issues, 169170 Welding, 200201 Wire arc AM (WAAM), 12 Work hardening alloys, 316 Wrought processing of TiAl, 238239, 238f X X-ray computer tomography scan (XRCTscan), 355 X-ray diffraction (XRD), 270, 274f, 435436 X-ray fluoroscopy (XRF), 177 xBeam 3D Metal Printing, 214, 217218, 220221, 223, 225226, 228 xBeam-01 pilot installation, 230, 230f, 231t Y Yield strength (YS), 360361 Z Zero gravity, 218 Zirconium, 170171