Figure 1: Scheme of liquid metal printing process

Effect of Aging Heat Treatment in an Al-4008 Produced byLiquid Metal Printing

C. M. Ladeiro
Department of Metallurgical and Materials Engineering, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto
Frias, 4200-465 PORTO, Portugal (up201806112@fe.up.pt) ORCID 0009-0003-8587-2309
F. L. Nunes
Department of Metallurgical and Materials Engineering, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto
Frias, 4200-465 PORTO, Portugal (up201806193@fe.up.pt) ORCID 0009-0000-0988-4285
M. M. Trindade
Department of Metallurgical and Materials Engineering, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto
Frias, 4200-465 PORTO, Portugal (up201806438@fe.up.pt) ORCID 0009-0008-1397-5321
J. M. Costa
Department of Metallurgical and Materials Engineering, Faculdade de Engenharia, Universidade do Porto and LAETA/INEGI –
Institute of Science and Innovation in Mechanical and Industrial Engineering, Rua Dr. Roberto Frias, 4200-465 PORTO,
Portugal (jose.costa@fe.up.pt) ORCID 0000-0002-1714-4671

Abstract

In today’s world, additive manufacturing (AM) is one of the most popular technologies and has the potential to revolutionize the manufacturing industry. As one of the most recent advances in this industry, liquid metal printing has a growing value in the engineering field. This study aims to evaluate the effect of two heat treatment conditions in an Al-4008 alloy produced by this technique in the microstructure and mechanical properties. It was concluded that the heat treatment (HT) enhances the Si particle coalescence and Fe-rich intermetallic compound precipitation, increasing the sample hardness significantly (50%). Density analysis showed a slight porosity decrease with HT. Tensile tests indicated heat-treated, same-directionally pulled samples exhibited brittleness compared to as-printed ones, while HT increased both yield strength (245 MPa) and ultimate tensile strength (294 MPa).

오늘날 세계에서 적층 제조(AM)는 가장 인기 있는 기술 중 하나이며 제조 산업에 혁명을 일으킬 잠재력을 가지고 있습니다. 이 업계의 가장 최근 발전 중 하나인 액체 금속 인쇄는 엔지니어링 분야에서 그 가치가 커지고 있습니다. 본 연구는 이 기술로 생산된 Al-4008 합금의 두 가지 열처리 조건이 미세 구조 및 기계적 특성에 미치는 영향을 평가하는 것을 목표로 합니다. 열처리(HT)는 Si 입자 유착과 Fe가 풍부한 금속간 화합물 침전을 향상시켜 샘플 경도를 크게(50%) 증가시키는 것으로 결론지었습니다. 밀도 분석에서는 HT를 사용하면 다공성이 약간 감소하는 것으로 나타났습니다. 인장 테스트에서는 동일한 방향으로 당겨진 열처리된 샘플이 인쇄된 샘플에 비해 취성을 보인 반면, HT는 항복 강도(245MPa)와 최대 인장 강도(294MPa)를 모두 증가시켰습니다.

Figure 1: Scheme of liquid metal printing process
Figure 1: Scheme of liquid metal printing process

Influences of the Powder Size and Process Parameters on the Quasi-Stability of Molten Pool Shape in Powder Bed Fusion-Laser Beam of Molybdenum

Influences of the Powder Size and Process Parameters on the Quasi-Stability of Molten Pool Shape in Powder Bed Fusion-Laser Beam of Molybdenum

몰리브덴 분말층 융합-레이저 빔의 용융 풀 형태의 준안정성에 대한 분말 크기 및 공정 매개변수의 영향

Abstract

Formation of a quasi-steady molten pool is one of the necessary conditions for achieving excellent quality in many laser processes. The influences of distribution characteristics of powder sizes on quasi-stability of the molten pool shape during single-track powder bed fusion-laser beam (PBF-LB) of molybdenum and the underlying mechanism were investigated.

The feasibility of improving quasi-stability of the molten pool shape by increasing the laser energy conduction effect and preheating was explored. Results show that an increase in the range of powder sizes does not significantly influence the average laser energy conduction effect in PBF-LB process. Whereas, it intensifies fluctuations of the transient laser energy conduction effect.

It also leads to fluctuations of the replenishment rate of metals, difficulty in formation of the quasi-steady molten pool, and increased probability of incomplete fusion and pores defects. As the laser power rises, the laser energy conduction effect increases, which improves the quasi-stability of the molten pool shape. When increasing the laser scanning speed, the laser energy conduction effect grows.

However, because the molten pool size reduces due to the decreased heat input, the replenishment rate of metals of the molten pool fluctuates more obviously and the quasi-stability of the molten pool shape gets worse. On the whole, the laser energy conduction effect in the PBF-LB process of Mo is low (20-40%). The main factor that affects quasi-stability of the molten pool shape is the amount of energy input per unit length of the scanning path, rather than the laser energy conduction effect.

Moreover, substrate preheating can not only enlarge the molten pool size, particularly the length, but also reduce non-uniformity and discontinuity of surface morphologies of clad metals and inhibit incomplete fusion and pores defects.

준안정 용융 풀의 형성은 많은 레이저 공정에서 우수한 품질을 달성하는 데 필요한 조건 중 하나입니다. 몰리브덴의 단일 트랙 분말층 융합 레이저 빔(PBF-LB) 동안 용융 풀 형태의 준안정성에 대한 분말 크기 분포 특성의 영향과 그 기본 메커니즘을 조사했습니다.

레이저 에너지 전도 효과와 예열을 증가시켜 용융 풀 형태의 준안정성을 향상시키는 타당성을 조사했습니다. 결과는 분말 크기 범위의 증가가 PBF-LB 공정의 평균 레이저 에너지 전도 효과에 큰 영향을 미치지 않음을 보여줍니다. 반면, 과도 레이저 에너지 전도 효과의 변동이 강화됩니다.

이는 또한 금속 보충 속도의 변동, 준안정 용융 풀 형성의 어려움, 불완전 융합 및 기공 결함 가능성 증가로 이어집니다. 레이저 출력이 증가함에 따라 레이저 에너지 전도 효과가 증가하여 용융 풀 모양의 준 안정성이 향상됩니다. 레이저 스캐닝 속도를 높이면 레이저 에너지 전도 효과가 커집니다.

그러나 열 입력 감소로 인해 용융 풀 크기가 줄어들기 때문에 용융 풀의 금속 보충 속도의 변동이 더욱 뚜렷해지고 용융 풀 형태의 준안정성이 악화됩니다.

전체적으로 Mo의 PBF-LB 공정에서 레이저 에너지 전도 효과는 낮다(20~40%). 용융 풀 형상의 준안정성에 영향을 미치는 주요 요인은 레이저 에너지 전도 효과보다는 스캐닝 경로의 단위 길이당 입력되는 에너지의 양입니다.

또한 기판 예열은 용융 풀 크기, 특히 길이를 확대할 수 있을 뿐만 아니라 클래드 금속 표면 형태의 불균일성과 불연속성을 줄이고 불완전한 융합 및 기공 결함을 억제합니다.

References

  1. M. Sharifitabar, F.O. Sadeq, and M.S. Afarani, Synthesis and Kinetic Study of Mo (Si, Al)2 Coatings on the Surface of Molybdenum Through Hot Dipping into a Commercial Al-12 wt.% Si Alloy Melt, Surf. Interfaces, 2021, 24, p 101044.Article CAS Google Scholar 
  2. Z. Zhang, X. Li, and H. Dong, Response of a Molybdenum Alloy to Plasma Nitriding, Int. J. Refract. Met. Hard Mater., 2018, 72, p 388–395.Article CAS Google Scholar 
  3. C. Tan, K. Zhou, M. Kuang, W. Ma, and T. Kuang, Microstructural Characterization and Properties of Selective Laser Melted Maraging Steel with Different Build Directions, Sci. Technol. Adv. Mater., 2018, 19(1), p 746–758.Article CAS Google Scholar 
  4. C. Tan, F. Weng, S. Sui, Y. Chew, and G. Bi, Progress and Perspectives in Laser Additive Manufacturing of Key Aeroengine Materials, Int. J. Mach. Tools Manuf, 2021, 170, p 103804.Article Google Scholar 
  5. S.A. Khairallah and A. Anderson, Mesoscopic Simulation Model of Selective Laser Melting of Stainless Steel Powder, J. Mater. Process. Technol., 2014, 214(11), p 2627–2636.Article CAS Google Scholar 
  6. S.A. Khairallah, A.T. Anderson, A. Rubenchik, and 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., 2016, 108, p 36–45.Article CAS ADS Google Scholar 
  7. K.Q. Le, C. Tang, and C.H. Wong, On the Study of Keyhole-Mode Melting in Selective Laser Melting Process, Int. J. Therm. Sci., 2019, 145, p 105992.Article Google Scholar 
  8. M. Bayat, A. Thanki, S. Mohanty, A. Witvrouw, S. Yang, J. Thorborg, N.S. Tiedje, and J.H. Hattel, Keyhole-Induced Porosities in Laser-Based Powder Bed Fusion (L-PBF) of Ti6Al4V: High-Fidelity Modelling and Experimental Validation, Addit. Manuf., 2019, 30, p 100835.CAS Google Scholar 
  9. B. Liu, G. Fang, L. Lei, and X. Yan, Predicting the Porosity Defects in Selective Laser Melting (SLM) by Molten Pool Geometry, Int. J. Mech. Sci., 2022, 228, p 107478.Article Google Scholar 
  10. W. Ge, J.Y.H. Fuh, and S.J. Na, Numerical Modelling of Keyhole Formation in Selective Laser Melting of Ti6Al4V, J. Manuf. Process., 2021, 62, p 646–654.Article Google Scholar 
  11. W. Ge, S. Han, S.J. Na, and J.Y.H. Fuh, Numerical Modelling of Surface Morphology in Selective Laser Melting, Comput. Mater. Sci., 2021, 186, p 110062.Article Google Scholar 
  12. Y.-C. Wu, C.-H. San, C.-H. Chang, H.-J. Lin, R. Marwan, S. Baba, and W.-S. Hwang, Numerical Modeling of Melt-Pool Behavior In Selective Laser Melting with Random Powder Distribution and Experimental Validation, J. Mater. Process. Technol., 2018, 254, p 72–78.Article Google Scholar 
  13. C. Tang, J.L. Tan, and C.H. Wong, A Numerical Investigation on the Physical Mechanisms of Single Track Defects in Selective Laser Melting, Int. J. Heat Mass Transf., 2018, 126, p 957–968.Article CAS Google Scholar 
  14. X. Zhou, X. Liu, D. Zhang, Z. Shen, and W. Liu, Balling Phenomena in Selective Laser Melted Tungsten, J. Mater. Process. Technol., 2015, 222, p 33–42.Article CAS Google Scholar 
  15. J.D.K. Monroy and J. Ciurana, Study of the Pore Formation on CoCrMo Alloys by Selective Laser Melting Manufacturing Process, Procedia Eng., 2013, 63, p 361–369.Article CAS Google Scholar 
  16. L. Kaserer, J. Braun, J. Stajkovic, K.H. Leitz, B. Tabernig, P. Singer, I. Letofsky-Papst, H. Kestler, and G. Leichtfried, Fully Dense and Crack Free Molybdenum Manufactured by Selective Laser Melting Through Alloying with Carbon, Int. J. Refract. Met. Hard Mater., 2019, 84, p 105000.Article CAS Google Scholar 
  17. T.B.T. Majumdar, E.M.C. Ribeiro, J.E. Frith, and N. Birbilis, Understanding the Effects of PBF Process Parameter Interplay on Ti-6Al-4V Surface Properties, PLoS ONE, 2019, 14, p e0221198.Article CAS PubMed PubMed Central Google Scholar 
  18. A.K.J.-R. Poulin, P. Terriault, and V. Brailovski, Long Fatigue Crack Propagation Behavior of Laser Powder Bed-Fused Inconel 625 with Intentionally- Seeded Porosity, Int. J. Fatigue, 2019, 127, p 144–156.Article CAS Google Scholar 
  19. P. Rebesan, M. Ballan, M. Bonesso, A. Campagnolo, S. Corradetti, R. Dima, C. Gennari, G.A. Longo, S. Mancin, M. Manzolaro, G. Meneghetti, A. Pepato, E. Visconti, and M. Vedani, Pure Molybdenum Manufactured by Laser Powder Bed Fusion: Thermal and Mechanical Characterization at Room and High Temperature, Addit. Manuf., 2021, 47, p 102277.CAS Google Scholar 
  20. D. Wang, C. Yu, J. Ma, W. Liu, and Z. Shen, Densification and Crack Suppression in Selective Laser Melting of Pure Molybdenum, Mater. Des., 2017, 129, p 44–52.Article CAS Google Scholar 
  21. K.-H. Leitz, P. Singer, A. Plankensteiner, B. Tabernig, H. Kestler, and L.S. Sigl, Multi-physical Simulation of Selective Laser Melting, Met. Powder Rep., 2017, 72, p 331–338.Article Google Scholar 
  22. D.G.J. Zhang, Y. Yang, H. Zhang, H. Chen, D. Dai, and K. Lin, Influence of Particle Size on Laser Absorption and Scanning Track Formation Mechanisms of Pure Tungsten Powder During Selective Laser Melting, Engineering, 2019, 5, p 736–745.Article CAS Google Scholar 
  23. L. Caprio, A.G. Demir, and B. Previtali, Influence of Pulsed and Continuous Wave Emission on Melting Efficiency in Selective Laser Melting, J. Mater. Process. Technol., 2019, 266, p 429–441.Article CAS Google Scholar 
  24. D. Gu, M. Xia, and D. Dai, On the Role of Powder Flow Behavior in Fluid Thermodynamics and Laser Processability of Ni-based Composites by Selective Laser Melting, Int. J. Mach. Tools Manuf, 2018, 137, p 67–78.Article Google Scholar 
  25. W.-I. Cho, S.-J. Na, C. Thomy, and F. Vollertsen, Numerical Simulation of Molten Pool Dynamics in High Power Disk Laser Welding, J. Mater. Process. Technol., 2012, 212(1), p 262–275.Article CAS Google Scholar 
  26. S.W. Han, J. Ahn, and S.J. Na, A Study on Ray Tracing Method for CFD Simulations of Laser Keyhole Welding: Progressive Search Method, Weld. World, 2016, 60, p 247–258.Article CAS Google Scholar 
  27. W. Ge, S. Han, Y. Fang, J. Cheon, and S.J. Na, Mechanism of Surface Morphology in Electron Beam Melting of Ti6Al4V Based on Computational Flow Patterns, Appl. Surf. Sci., 2017, 419, p 150–158.Article CAS ADS Google Scholar 
  28. W.-I. Cho, S.-J. Na, C. Thomy, and F. Vollertsen, Numerical Simulation of Molten Pool Dynamics in High Power Disk Laser Welding, J. Mater. Process. Technol., 2012, 212, p 262–275.Article CAS Google Scholar 
  29. W. Ma, J. Ning, L.-J. Zhang, and S.-J. Na, Regulation of Microstructures and Properties of Molybdenum-Silicon-Boron Alloy Subjected to Selective Laser Melting, J. Manuf. Process., 2021, 69, p 593–601.Article Google Scholar 
  30. S. Haeri, Y. Wang, O. Ghita, and J. Sun, Discrete Element Simulation and Experimental Study of Powder Spreading Process in Additive Manufacturing, Powder Technol., 2016, 306, p 45–54.Article Google Scholar 
  31. D. Yao, X. Liu, J. Wang, W. Fan, M. Li, H. Fu, H. Zhang, X. Yang, Q. Zou, and X. An, Numerical Insights on the Spreading of Practical 316 L Stainless Steel Powder in SLM Additive Manufacturing, Powder Technol., 2021, 390, p 197–208.Article CAS Google Scholar 
  32. S. Vock, B. Klöden, A. Kirchner, T. Weißgärber, and B. Kieback, Powders for Powder Bed Fusion: A Review, Prog. Addit. Manuf., 2019, 4, p 383–397.Article Google Scholar 
  33. X. Luo, C. Yang, Z.Q. Fu, L.H. Liu, H.Z. Lu, H.W. Ma, Z. Wang, D.D. Li, L.C. Zhang, and Y.Y. Li, Achieving Ultrahigh-Strength in Beta-Type Titanium Alloy by Controlling the Melt Pool Mode in Selective Laser Melting, Mater. Sci. Eng. A, 2021, 823, p 141731.Article CAS Google Scholar 
  34. J. Braun, L. Kaserer, J. Stajkovic, K.-H. Leitz, B. Tabernig, P. Singer, P. Leibenguth, C. Gspan, H. Kestler, and G. Leichtfried, Molybdenum and Tungsten Manufactured by Selective Laser Melting: Analysis of Defect Structure and Solidification Mechanisms, Int. J. Refract. Met. Hard Mater., 2019, 84, p 104999.Article CAS Google Scholar 
  35. L. Kaserera, J. Brauna, J. Stajkovica, K.-H. Leitzb, B. Tabernigb, P. Singerb, I. Letofsky-Papstc, H. Kestlerb, and G. Leichtfried, Fully Dense and Crack Free Molybdenum Manufactured by Selective Laser Melting Through Alloying with Carbon, Int. J. Refract Metal Hard Mater., 2019, 84, p 105000.Article Google Scholar 
Figure 3. The simulated temperature distribution and single-layer multi-track isothermograms of LPBF Hastelloy X, located at the bottom of the powder bed, are presented for various laser energy densities. (a) depicts the single-point temperature distribution at the bottom of the powder bed, followed by the isothermograms corresponding to laser energy densities of (b) 31 J/mm3 , (c) 43 J/mm3 , (d) 53 J/mm3 , (e) 67 J/mm3 , and (f) 91 J/mm3 .

An integrated multiscale simulation guiding the processing optimisation for additively manufactured nickel-based superalloys

적층 가공된 니켈 기반 초합금의 가공 최적화를 안내하는 통합 멀티스케일 시뮬레이션

Xing He, Bing Yang, Decheng Kong, Kunjie Dai, Xiaoqing Ni, Zhanghua Chen
& Chaofang Dong

ABSTRACT

Microstructural defects in laser powder bed fusion (LPBF) metallic materials are correlated with processing parameters. A multi-physics model and a crystal plasticity framework are employed to predict microstructure growth in molten pools and assess the impact of manufacturing defects on plastic damage parameters. Criteria for optimising the LPBF process are identified, addressing microstructural defects and tensile properties of LPBF Hastelloy X at various volumetric energy densities (VED). The results show that higher VED levels foster a specific Goss texture {110} <001>, with irregular lack of fusion defects significantly affecting plastic damage, especially near the material surface. A critical threshold emerges between manufacturing defects and grain sizes in plastic strain accumulation. The optimal processing window for superior Hastelloy X mechanical properties ranges from 43 to 53 J/mm3 . This work accelerates the development of superior strengthductility alloys via LPBF, streamlining the trial-and-error process and reducing associated costs.

Figure 3. The simulated temperature distribution and single-layer multi-track isothermograms of LPBF Hastelloy X, located at the bottom of the powder bed, are presented for various laser energy densities. (a) depicts the single-point temperature distribution at the bottom of the powder bed, followed by the isothermograms corresponding to laser energy densities of (b) 31 J/mm3 , (c) 43 J/mm3 , (d) 53 J/mm3 , (e) 67 J/mm3 , and (f) 91 J/mm3 .
Figure 3. The simulated temperature distribution and single-layer multi-track isothermograms of LPBF Hastelloy X, located at the bottom of the powder bed, are presented for various laser energy densities. (a) depicts the single-point temperature distribution at the bottom of the powder bed, followed by the isothermograms corresponding to laser energy densities of (b) 31 J/mm3 , (c) 43 J/mm3 , (d) 53 J/mm3 , (e) 67 J/mm3 , and (f) 91 J/mm3 .

References
[1] DebRoy T, Wei HL, Zuback JS, et al. Additive manufacturing of metallic components – process, structure and properties. Prog Mater Sci. 2018;92:112–224. doi:10.
1016/j.pmatsci.2017.10.001
[2] Mostafaei A, Ghiaasiaan R, Ho IT, et al. Additive manufacturing of nickel-based superalloys: A state-of-the-art
review on process-structure-defect-property relationship.
Prog Mater Sci. 2023;136:101108. doi:10.1016/j.pmatsci.
2023.101108
[3] Akande IG, Oluwole OO, Fayomi OSI, et al. Overview of
mechanical, microstructural, oxidation properties and
high-temperature applications of superalloys. Mater
Today Proc. 2021;43:2222–2231. doi:10.1016/j.matpr.
2020.12.523
[4] Sanchez S, Smith P, Xu Z, et al. Powder bed fusion of
nickel-based superalloys: a review. Int J Machine Tools
Manuf. 2021;165:103729. doi:10.1016/j.ijmachtools.2021.
103729
[5] Xie Y, Teng Q, Shen M, et al. The role of overlap region
width in multi-laser powder bed fusion of Hastelloy X
superalloy. Virtual Phys Prototyp. 2023;18(1):e2142802.
doi:10.1080/17452759.2022.2142802
[6] Yuan W, Chen H, Cheng T, et al. Effects of laser scanning
speeds on different states of the molten pool during
selective laser melting: simulation and experiment.
Mater Des. 2020;189:108542. doi:10.1016/j.matdes.2020.
108542
[7] He X, Kong D, Zhou Y, et al. Powder recycling effects on
porosity development and mechanical properties of
Hastelloy X alloy during laser powder bed fusion
process. Addit Manuf. 2022;55:102840. doi:10.1016/j.
addma.2022.102840
[8] Sanaei N, Fatemi A. Defects in additive manufactured
metals and their effect on fatigue performance: a stateof-the-art review. Prog Mater Sci. 2021;117:100724.
doi:10.1016/j.pmatsci.2020.100724
[9] Pourbabak S, Montero-Sistiaga ML, Schryvers D, et al.
Microscopic investigation of as built and hot isostatic
pressed Hastelloy X processed by selective laser
melting. Mater Charact. 2019;153:366–371. doi:10.1016/j.
matchar.2019.05.024
[10] He X, Wang L, Kong D, et al. Recrystallization effect on
surface passivation of Hastelloy X alloy fabricated by
laser powder bed fusion. J Mater Sci Technol.
2023;163:245–258. doi:https://doi.org/10.1016j.jmst.
2023.06.003.
[11] Sabzi HE, Maeng S, Liang X, et al. Controlling crack formation and porosity in laser powder bed fusion: alloy
design and process optimisation. Addit Manuf.
2020;34:101360. doi:10.1016/j.addma.2020.101360
[12] Yu C, Chen N, Li R, et al. Selective laser melting of GH3536
superalloy: microstructure, mechanical properties, and
hydrocyclone manufacturing. Adv Powder Mater. 2023:

doi:10.1016/j.apmate.2023.100134
[13] Ye C, Zhang C, Zhao J, et al. Effects of post-processing on
the surface finish, porosity, residual stresses, and fatigue
performance of additive manufactured metals: a review.
J Mater Eng Perform. 2021;30(9):6407–6425. doi:10.
1007/s11665-021-06021-7
[14] Zhang W, Zheng Y, Liu F, et al. Effect of solution temperature on the microstructure and mechanical properties of
Hastelloy X superalloy fabricated by laser directed energy
deposition. Mater Sci Eng A. 2021;820:141537. doi:10.
1016/j.msea.2021.141537
[15] Lehmann T, Rose D, Ranjbar E, et al. Large-scale metal
additive manufacturing: a holistic review of the state of the art and challenges. Int Mater Rev. 2021;67(4):410–459. doi:10.1080/09506608.2021.1971427

[16] Wu S, Hu Y, Yang B, et al. Review on defect characterization and structural integrity assessment method of additively manufactured materials. J Mech Eng. 2021;57 (22):3–34. doi:10.3901/JME.2021.22.003

[17] Keller C, Mokhtari M, Vieille B, et al. Influence of a rescanning strategy with different laser powers on the microstructure and mechanical properties of Hastelloy X elaborated by powder bed fusion. Mater Sci Eng A. 2021;803:140474. doi:10.1016/j.msea.2020.140474

[18] Keshavarzkermani A, Marzbanrad E, Esmaeilizadeh R,et al. An investigation into the effect of process parameters on melt pool geometry, cell spacing, and grain refinement during laser powder bed fusion. Optics & Laser Technol. 2019;116:83–91. doi:10.1016/j.optlastec. 2019.03.012

[19] Watring DS, Benzing JT, Hrabe N, et al. Effects of laserenergy density and build orientation on the structureproperty relationships in as-built Inconel 718 manufactured by laser powder bed fusion. Addit Manuf. 2020;36:101425. doi:10.1016/j.addma.2020.101425

[20] Xiao H, Liu X, Xiao W, et al. Influence of molten-pool cooling rate on solidification structure and mechanical property of laser additive manufactured Inconel 718. J Mater Res Technol. 2022;19:4404–4416. doi:10.1016/j. jmrt.2022.06.162

[21] Wang J, Zhu R, Liu Y, et al. Understanding melt pool characteristics in laser powder bed fusion: An overview of single- and multi-track melt pools for process optimization. Adv Powder Mater. 2023;2(4):100137. doi:10.1016/j. apmate.2023.100137

[22] Li Z, Deng Y, Yao B, et al. Effect of laser scan speed on pool size and densification of selective laser melted CoCr alloy under constant laser energy density. Laser Optoelectronics Progress. 2022;59(7):0736001. doi:10. 3788/LOP202259.0736001

[23] Zhang J, Yuan W, Song B, et al. Towards understanding metallurgical defect formation of selective laser melted wrought aluminum alloys. Adv Powder Mater. 2022;1 (4):100035. doi:10.1016/j.apmate.2022.100035

[24] Rui H, Meiping W, Chen C, et al. Effects of laser energy density on microstructure and corrosion resistance of FeCrNiMnAl high entropy alloy coating. Optics & Laser Technol. 2022;152:108188. doi:https://doi.org/10.1016j. optlastec.2022.108188.

[25] Zhao Y, Sun W, Wang Q, et al. Effect of beam energy density characteristics on microstructure and mechanical properties of nickel-based alloys manufactured by laser directed energy deposition. J Mater Process Technol. 2023;319:118074. doi:10.1016/j.jmatprotec.2023.118074

[26] Tan P, Zhou M, Tang C, et al. Multiphysics modelling of powder bed fusion for polymers. Virtual Phys Prototyp. 2023;18(1):e2257191. doi:10.1080/17452759.2023. 2257191

[27] Tan P, Shen F, Shian Tey W, et al. A numerical study on the packing quality of fibre/polymer composite powder for powder bed fusion additive manufacturing. Virtual Phys Prototyp. 2021;16(sup1):S1–S18. doi:10.1080/17452759. 2021.1922965

[28] Kusano M, Watanabe M. Microstructure control of Hastelloy X by geometry-induced elevation of sample temperature during a laser powder bed fusion process. Mater Des. 2022;222:111016. doi:10.1016/j.matdes.2022. 111016

[29] Lee YS, Zhang W. Modeling of heat transfer, fluid flow and solidification microstructure of nickel-base superalloy fabricated by laser powder bed fusion. Addit Manuf. 2016;12:178–188. doi:10.1016/j.addma.2016.05.003

[30] Lv F, Liang HX, Xie DQ, et al. On the role of laser in situ remelting into pore elimination of Ti-6Al-4V components fabricated by selective laser melting. J Alloys Compd. 2021;854:156866. doi:10.1016/j.jallcom.2020.156866

[31] Prithivirajan V, Sangid MD. The role of defects and critical pore size analysis in the fatigue response of additively manufactured IN718 via crystal plasticity. Mater Des. 2018;150:139–153. doi:10.1016/j.matdes.2018.04.022

[32] Huang Y. A user-material subroutine incroporating single crystal plasticity in the ABAQUS finite element program. Cambridge: Harvard University Press; 1991.

[33] Pilgar CM, Fernandez AM, Lucarini S, et al. Effect of printing direction and thickness on the mechanical behavior of SLM fabricated Hastelloy-X. Int J Plasticity. 2022;153:103250. doi:10.1016/j.ijplas.2022.103250

[34] Garlea E, Choo H, Sluss CC, et al. Variation of elastic mechanical properties with texture, porosity, and defect characteristics in laser powder bed fusion 316L stainless steel. Mater Sci Eng A. 2019;763:138032. doi:10.1016/j. msea.2019.138032

[35] Sanchez-Mata O, Wang X, Muñiz-Lerma JA, et al. Dependence of mechanical properties on crystallographic orientation in nickel-based superalloy Hastelloy X fabricated by laser powder bed fusion. J Alloys Compd. 2021;865:158868. doi:10.1016/j.jallcom.2021. 158868

[36] Gu H, Wei C, Li L, et al. Multi-physics modelling of molten

pool development and track formation in multi-track, multi-layer and multi-material selective laser melting. Int J Heat Mass Transf. 2020;151:119458. doi:10.1016/j. ijheatmasstransfer.2020.119458

[37] Johnson L, Mahmoudi M, Zhang B, et al. Assessing printability maps in additive manufacturing of metal alloys. Acta Mater. 2019;176:199–210. doi:10.1016/j.actamat. 2019.07.005

[38] Wang S, Ning J, Zhu L, et al. Role of porosity defects in metal 3D printing: formation mechanisms, impacts on properties and mitigation strategies. Mater Today. 2022;59:133–160. doi:10.1016/j.mattod.2022.08.014

[39] Guo Y, Collins DM, Tarleton E, et al. Measurements of stress fields near a grain boundary: exploring blocked arrays of dislocations in 3D. Acta Mater. 2015;96:229–doi:10.1016/j.actamat.2015.05.041
[40] Kong D, Dong C, Ni X, et al. Hetero-deformation-induced
stress in additively manufactured 316L stainless steel.
Mater Res Lett. 2020;8(10):390–397. doi:10.1080/
21663831.2020.1775149

Figure 1. Experimental setup and materials. (a) Schematic of the DED process, where three types of base materials were adopted—B1 (IN718), B2 (IN625), and B3 (SS316L), and two types of powder materials were adopted—P1 (IN718) and P2 (SS316L). (b) In situ high-speed imaging of powder flow and the SEM images of IN718 and SS316L powder particle. (c) Powder size statistics, and (d) element composition of powder IN718 (P1) and SS316L (P2).

Printability disparities in heterogeneous materialcombinations via laser directed energy deposition:a comparative stud

Jinsheng Ning1,6, Lida Zhu1,6,∗, Shuhao Wang2, Zhichao Yang1, Peihua Xu1,Pengsheng Xue3, Hao Lu1, Miao Yu1, Yunhang Zhao1, Jiachen Li4, Susmita Bose5 and Amit Bandyopadhyay5,∗

Abstract

적층 제조는 바이메탈 및 다중 재료 구조의 제작 가능성을 제공합니다. 그러나 재료 호환성과 접착성은 부품의 성형성과 최종 품질에 직접적인 영향을 미칩니다. 적합한 프로세스를 기반으로 다양한 재료 조합의 기본 인쇄 가능성을 이해하는 것이 중요합니다.

여기에서는 두 가지 일반적이고 매력적인 재료 조합(니켈 및 철 기반 합금)의 인쇄 적성 차이가 레이저 지향 에너지 증착(DED)을 통해 거시적 및 미시적 수준에서 평가됩니다.

증착 프로세스는 현장 고속 이미징을 사용하여 캡처되었으며, 용융 풀 특징 및 트랙 형태의 차이점은 특정 프로세스 창 내에서 정량적으로 조사되었습니다. 더욱이, 다양한 재료 쌍으로 처리된 트랙과 블록의 미세 구조 다양성이 비교적 정교해졌고, 유익한 다중 물리 모델링을 통해 이종 재료 쌍 사이에 제시된 기계적 특성(미세 경도)의 불균일성이 합리화되었습니다.

재료 쌍의 서로 다른 열물리적 특성에 의해 유발된 용융 흐름의 차이와 응고 중 결과적인 요소 혼합 및 국부적인 재합금은 재료 조합 간의 인쇄 적성에 나타난 차이점을 지배합니다.

이 작업은 서로 다른 재료의 증착에서 현상학적 차이에 대한 심층적인 이해를 제공하고 바이메탈 부품의 보다 안정적인 DED 성형을 안내하는 것을 목표로 합니다.

Additive manufacturing provides achievability for the fabrication of bimetallic and
multi-material structures; however, the material compatibility and bondability directly affect the
parts’ formability and final quality. It is essential to understand the underlying printability of
different material combinations based on an adapted process. Here, the printability disparities of
two common and attractive material combinations (nickel- and iron-based alloys) are evaluated
at the macro and micro levels via laser directed energy deposition (DED). The deposition
processes were captured using in situ high-speed imaging, and the dissimilarities in melt pool
features and track morphology were quantitatively investigated within specific process
windows. Moreover, the microstructure diversity of the tracks and blocks processed with varied
material pairs was comparatively elaborated and, complemented with the informative
multi-physics modeling, the presented non-uniformity in mechanical properties (microhardness)
among the heterogeneous material pairs was rationalized. The differences in melt flow induced
by the unlike thermophysical properties of the material pairs and the resulting element
intermixing and localized re-alloying during solidification dominate the presented dissimilarity
in printability among the material combinations. This work provides an in-depth understanding
of the phenomenological differences in the deposition of dissimilar materials and aims to guide
more reliable DED forming of bimetallic parts.

Figure 1. Experimental setup and materials. (a) Schematic of the DED process, where three types of base materials were adopted—B1
(IN718), B2 (IN625), and B3 (SS316L), and two types of powder materials were adopted—P1 (IN718) and P2 (SS316L). (b) In situ
high-speed imaging of powder flow and the SEM images of IN718 and SS316L powder particle. (c) Powder size statistics, and (d) element
composition of powder IN718 (P1) and SS316L (P2).
Figure 1. Experimental setup and materials. (a) Schematic of the DED process, where three types of base materials were adopted—B1 (IN718), B2 (IN625), and B3 (SS316L), and two types of powder materials were adopted—P1 (IN718) and P2 (SS316L). (b) In situ high-speed imaging of powder flow and the SEM images of IN718 and SS316L powder particle. (c) Powder size statistics, and (d) element composition of powder IN718 (P1) and SS316L (P2).
Figure 2. Deposition process and the track morphology. (a)–(c) Display the in situ captured tableaux of melt propagation and some physical
features during depositing for P1B1, P1B2, and P1B3, respectively. (d) The profiles of the melt pool at a frame of (t0 + 1) ms, and the flow
streamlines in the molten pool of each case. (e) The outer surface of the formed tracks, in which the colored arrows mark the scanning
direction. (f) Cross-section of the tracks. The parameter set used for in situ imaging was P-1000 W, S-600 mm·min–1, F-18 g·min–1. All the
scale bars are 2 mm.
Figure 2. Deposition process and the track morphology. (a)–(c) Display the in situ captured tableaux of melt propagation and some physical features during depositing for P1B1, P1B2, and P1B3, respectively. (d) The profiles of the melt pool at a frame of (t0 + 1) ms, and the flow streamlines in the molten pool of each case. (e) The outer surface of the formed tracks, in which the colored arrows mark the scanning direction. (f) Cross-section of the tracks. The parameter set used for in situ imaging was P-1000 W, S-600 mm·min–1, F-18 g·min–1. All the scale bars are 2 mm.

References

[1] Tan C L, Weng F, Sui S, Chew Y and Bi G J 2021 Progress and perspectives in laser additive manufacturing of key aeroengine materials Int. J. Mach. Tools Manuf. 170 103804
[2] Bandyopadhyay A, Traxel K D, Lang M, Juhasz M, Eliaz N and Bose S 2022 Alloy design via additive manufacturing: advantages, challenges, applications and perspectives Mater. Today 52 207–24
[3] Sui S, Chew Y, Weng F, Tan C L, Du Z L and Bi G J 2022 Study of the intrinsic mechanisms of nickel additive for grain refinement and strength enhancement of laser aided additively manufactured Ti–6Al–4V Int. J. Extrem. Manuf. 4 035102
[4] Xue P S, Zhu L D, Xu P H, Ren Y, Xin B, Meng G R, Yang Z C and Liu Z 2021 Research on process optimization and microstructure of CrCoNi medium-entropy alloy formed by laser metal deposition Opt. Laser Technol. 142 107167
[5] Bandyopadhyay A, Traxel K D and Bose S 2021 Nature-inspired materials and structures using 3D printing Mater. Sci. Eng. R 145 100609
[6] Zuback J S, Palmer T A and DebRoy T 2019 Additive manufacturing of functionally graded transition joints between ferritic and austenitic alloys J. Alloys Compd. 770 995–1003
[7] Feenstra D R, Banerjee R, Fraser H L, Huang A, Molotnikov A and Birbilis N 2021 Critical review of the state of the art in multi-material fabrication via directed energy deposition Curr. Opin. Solid State Mater. Sci. 25 100924
[8] Wei C, Zhang Z Z, Cheng D X, Sun Z, Zhu M H and Li L 2021 An overview of laser-based multiple metallic material additive manufacturing: from macro- to micro-scales Int. J. Extrem. Manuf. 3 012003
[9] Gu D D, Shi X Y, Poprawe R, Bourell D L, Setchi R and Zhu J H 2021 Material-structure-performance integrated laser-metal additive manufacturing Science 372 eabg1487
[10] Bandyopadhyay A and Heer B 2018 Additive manufacturing of multi-material structures Mater. Sci. Eng. R 129 1–16
[11] Tammas-Williams S and Todd I 2017 Design for additive manufacturing with site-specific properties in metals and alloys Scr. Mater. 135 105–10
[12] Chen W, Gu D D, Yang J K, Yang Q, Chen J and Shen X F 2022 Compressive mechanical properties and shape memory effect of NiTi gradient lattice structures fabricated by laser powder bed fusion Int. J. Extrem. Manuf. 4 045002
[13] Svetlizky D, Das M, Zheng B L, Vyatskikh A L, Bose S, Bandyopadhyay A, Schoenung J M, Lavernia E J and Eliaz N 2021 Directed energy deposition (DED) additive manufacturing: physical characteristics, defects, challenges and applications Mater. Today 49 271–95
[14] Panwisawas C, Tang Y T and Reed R C 2020 Metal 3D printing as a disruptive technology for superalloys Nat. Commun. 11 2327
[15] Wang S H, Ning J S, Zhu L D, Yang Z C, Yan W T, Dun Y C, Xue P S, Xu P H, Bose S and Bandyopadhyay A 2022 Role of porosity defects in metal 3D printing: formation mechanisms, impacts on properties and mitigation strategies Mater. Today 59 133–60
[16] DebRoy T, Mukherjee T, Milewski J O, Elmer J W, Ribic B, Blecher J J and Zhang W 2019 Scientific, technological and economic issues in metal printing and their solutions Nat. Mater. 18 1026–32
[17] Afrouzian A, Groden C J, Field D P, Bose S and Bandyopadhyay A 2022 Additive manufacturing of Ti-Ni bimetallic structures Mater. Des. 215 110461
[18] Bandyopadhyay A, Zhang Y N and Onuike B 2022 Additive manufacturing of bimetallic structures Virtual Phys. Prototyp. 17 256–94
[19] Onuike B, Heer B and Bandyopadhyay A 2018 Additive manufacturing of Inconel 718—copper alloy bimetallic structure using laser engineered net shaping (LENSTM) Addit. Manuf. 21 133–40
[20] Sahasrabudhe H, Harrison R, Carpenter C and Bandyopadhyay A 2015 Stainless steel to titanium bimetallic structure using LENSTM Addit. Manuf. 5 1–8
[21] Li B Y, Han C J, Lim C W J and Zhou K 2022 Interface formation and deformation behaviors of an additively manufactured nickel-aluminum-bronze/15-5 PH multimaterial via laser-powder directed energy deposition Mater. Sci. Eng. A 829 142101
[22] Zhang X C, Pan T, Chen Y T, Li L, Zhang Y L and Liou F 2021 Additive manufacturing of copper-stainless steel hybrid components using laser-aided directed energy deposition J. Mater. Sci. Technol. 80 100–16
[23] Shinjo J and Panwisawas C 2022 Chemical species mixing during direct energy deposition of bimetallic systems using titanium and dissimilar refractory metals for repair and biomedical applications Addit. Manuf. 51 102654
[24] Wang D et al 2022 Recent progress on additive manufacturing of multi-material structures with laser powder bed fusion Virtual Phys. Prototyp. 17 329–65
[25] Lin X, Yue T M, Yang H O and Huang W D 2005 Laser rapid forming of SS316L/Rene88DT graded material Mater. Sci. Eng. A 391 325–36
[26] Melzer D, Dˇzugan J, Koukolíková M, Rzepa S and Vavˇrík J 2021 Structural integrity and mechanical properties of the functionally graded material based on 316L/IN718 processed by DED technology Mater. Sci. Eng. A 811 141038
[27] Melzer D, Dˇzugan J, Koukolíková M, Rzepa S, Dlouh´y J, Brázda M and Bucki T 2022 Fracture characterisation of vertically build functionally graded 316L stainless steel with Inconel 718 deposited by directed energy deposition process Virtual Phys. Prototyp. 17 821–40
[28] Zhang Y N and Bandyopadhyay A 2018 Direct fabrication of compositionally graded Ti-Al2O3 multi-material structures using laser engineered net shaping Addit. Manuf. 21 104–11
[29] Ben-Artzy A, Reichardt A, Borgonia P J, Dillon R P, McEnerney B, Shapiro A A and Hosemann P 2021 Compositionally graded SS316 to C300 maraging steel using additive manufacturing Mater. Des. 201 109500
[30] Tan C L, Liu Y C, Weng F, Ng F L, Su J L, Xu Z K, Ngai X D and Chew Y 2022 Additive manufacturing of voxelized heterostructured materials with hierarchical phases Addit. Manuf. 54 102775
[31] Chen J, Yang Y Q, Song C H, Zhang M K, Wu S B and Wang D 2019 Interfacial microstructure and mechanical properties of 316L/CuSn10 multi-material bimetallic structure fabricated by selective laser melting Mater. Sci. Eng. A 752 75–85
[32] Wei C, Gu H, Gu Y C, Liu L C, Huang Y H, Cheng D X, Li Z Q and Li L 2022 Abnormal interfacial bonding mechanisms of multi-material additive-manufactured tungsten–stainless steel sandwich structure Int. J. Extrem. Manuf. 4 025002
[33] Zhang Y N and Bandyopadhyay A 2021 Influence of compositionally graded interface on microstructure and compressive deformation of 316L stainless steel to Al12Si aluminum alloy bimetallic structures ACS Appl. Mater. Interfaces 13 9174–85
[34] Wei C et al 2022 Cu10Sn to Ti6Al4V bonding mechanisms in laser-based powder bed fusion multiple material additive 15 Int. J. Extrem. Manuf. 6 (2024) 025001 J Ning et al manufacturing with different build strategies Addit. Manuf. 51 102588
[35] Li W, Karnati S, Kriewall C, Liou F, Newkirk J, Brown Taminger K M and Seufzer W J 2017 Fabrication and characterization of a functionally graded material from Ti-6Al-4V to SS316 by laser metal deposition Addit. Manuf. 14 95–104
[36] Shi Q M, Zhong G Y, Sun Y, Politis C and Yang S F 2021 Effects of laser melting+remelting on interfacial macrosegregation and resulting microstructure and microhardness of laser additive manufactured H13/IN625 bimetals J. Manuf. Process. 71 345–55
[37] Zhang W X, Hou W Y, Deike L and Arnold C 2022 Understanding the Rayleigh instability in humping phenomenon during laser powder bed fusion process Int. J. Extrem. Manuf. 4 015201
[38] Chen Y W, Zhang X, Li M M, Xu R Q, Zhao C and Sun T 2020 Laser powder bed fusion of Inconel 718 on 316 stainless steel Addit. Manuf. 36 101500
[39] Yang Z C, Wang S H, Zhu L D, Ning J S, Xin B, Dun Y C and Yan W T 2022 Manipulating molten pool dynamics during metal 3D printing by ultrasound Appl. Phys. Rev. 9 021416
[40] Hofmann D C, Roberts S, Otis R, Kolodziejska J, Dillon R P, Suh J O, Shapiro A A, Liu Z K and Borgonia J P 2014 Developing gradient metal alloys through radial deposition additive manufacturing Sci. Rep. 4 5357
[41] Tumkur T U et al 2021 Nondiffractive beam shaping for enhanced optothermal control in metal additive manufacturing Sci. Adv. 7 eabg9358
[42] Scipioni Bertoli U, Guss G, Wu S, Matthews M J and Schoenung J M 2017 In-situ characterization of laser-powder interaction and cooling rates through high-speed imaging of powder bed fusion additive manufacturing Mater. Des. 135 385–96
[43] Siva Prasad H, Brueckner F and Kaplan A F H 2020 Powder incorporation and spatter formation in high deposition rate blown powder directed energy deposition Addit. Manuf. 35 101413
[44] Ebrahimi A, Kleijn C R and Richardson I M 2021 Numerical study of molten metal melt pool behaviour during conduction-mode laser spot melting J. Appl. Phys. 54 105304
[45] Mumtaz K A and Hopkinson N 2010 Selective laser melting of thin wall parts using pulse shaping J. Mater. Process. Technol. 210 279–87
[46] Sikandar Iquebal A, Yadav A, Botcha B, Krishna Gorthi R and Bukkapatnam S 2022 Tracking and quantifying spatter characteristics in a laser directed energy deposition process using Kalman filter Manuf. Lett. 33 692–700
[47] Criales L E, Arısoy Y M, Lane B, Moylan S, Donmez A and Özel T 2017 Laser powder bed fusion of nickel alloy 625: experimental investigations of effects of process parameters on melt pool size and shape with spatter analysis Int. J. Mach. Tools Manuf. 121 22–36
[48] Coen V, Goossens L and van Hooreweder B 2022 Methodology and experimental validation of analytical melt pool models for laser powder bed fusion J. Mater. Process. Technol. 304 117547
[49] Zhao C, Shi B, Chen S L, Du D, Sun T, Simonds B J, Fezzaa K and Rollett A D 2022 Laser melting modes in metal powder bed fusion additive manufacturing Rev. Mod. Phys. 94 045002
[50] Wang J H, Han F Z, Chen S F and Ying W S 2019 A novel model of laser energy attenuation by powder particles for laser solid forming Int. J. Mach. Tools Manuf. 145 103440
[51] Haley J C, Schoenung J M and Lavernia E J 2018 Observations of particle-melt pool impact events in directed energy deposition Addit. Manuf. 22 368–74
[52] Chen Y H et al 2021 Correlative synchrotron x-ray imaging and diffraction of directed energy deposition additive manufacturing Acta Mater. 209 116777
[53] Khorasani M, Ghasemi A, Leary M, Cordova L, Sharabian E, Farabi E, Gibson I, Brandt M and Rolfe B 2022 A comprehensive study on meltpool depth in laser-based powder bed fusion of Inconel 718 Int. J. Adv. Manuf. Technol. 120 2345–62
[54] Shamsaei N, Yadollahi A, Bian L and Thompson S M 2015 An overview of direct laser deposition for additive manufacturing; part II: mechanical behavior, process parameter optimization and control Addit. Manuf. 8 12–35
[55] Ghanavati R, Naffakh-Moosavy H, Moradi M and Eshraghi M 2022 Printability and microstructure of directed energy deposited SS316l-IN718 multi-material: numerical modeling and experimental analysis Sci. Rep. 12 16600
[56] Galbusera F, Demir A G, Platl J, Turk C, Schnitzer R and Previtali B 2022 Processability and cracking behaviour of novel high-alloyed tool steels processed by laser powder bed fusion J. Mater. Process. Technol. 302 117435
[57] Wang A et al 2023 Effects of processing parameters on pore defects in blue laser directed energy deposition of aluminum by in and ex situ observation J. Mater. Process. Technol. 319 118068
[58] Hinojos A, Mireles J, Reichardt A, Frigola P, Hosemann P, Murr L E and Wicker R B 2016 Joining of Inconel 718 and 316 stainless steel using electron beam melting additive manufacturing technology Mater. Des. 94 17–27
[59] Yang Z C, Zhu L D, Wang S H, Ning J S, Dun Y C, Meng G R, Xue P S, Xu P H and Xin B 2021 Effects of ultrasound on multilayer forming mechanism of Inconel 718 in directed energy deposition Addit. Manuf. 48 102462
[60] Yao L M, Huang S, Ramamurty U and Xiao Z M 2021 On the formation of “Fish-scale” morphology with curved grain interfacial microstructures during selective laser melting of dissimilar alloys Acta Mater. 220 117331
[61] Ghanavati R, Naffakh-Moosavy H and Moradi M 2021 Additive manufacturing of thin-walled SS316L-IN718 functionally graded materials by direct laser metal deposition J. Mater. Res. Technol. 15 2673–85
[62] Chen N N, Khan H A, Wan Z X, Lippert J, Sun H, Shang S L, Liu Z K and Li J J 2020 Microstructural characteristics and crack formation in additively manufactured bimetal material of 316L stainless steel and Inconel 625 Addit. Manuf. 32 101037
[63] Xiao Y H, Wan Z X, Liu P W, Wang Z, Li J J and Chen L 2022 Quantitative simulations of grain nucleation and growth at additively manufactured bimetallic interfaces of SS316L and IN625 J. Mater. Process. Technol. 302 117506
[64] Mukherjee T, DebRoy T, Lienert T J, Maloy S A and Hosemann P 2021 Spatial and temporal variation of hardness of a printed steel part Acta Mater. 209 116775
[65] Dinda G P, Dasgupta A K and Mazumder J 2021 Texture control during laser deposition of nickel-based superalloy Scr. Mater. 67 503–6
[66] Tan Z E, Pang J H L, Kaminski J and Pepin H 2019 Characterisation of porosity, density, and microstructure of directed energy deposited stainless steel AISI 316L Addit. Manuf. 25 286–96
[67] Wolff S J, Gan Z T, Lin S, Bennett J L, Yan W T, Hyatt G, Ehmann K F, Wagner G J, Liu W K and Cao J 2019 Experimentally validated predictions of thermal history and microhardness in laser-deposited Inconel 718 on carbon steel Addit. Manuf. 27 540–51 16 Int. J. Extrem. Manuf. 6 (2024) 025001 J Ning et al
[68] Zhang L, Wen M, Imade M, Fukuyama S and Yokogawa K 2008 Effect of nickel equivalent on hydrogen gas embrittlement of austenitic stainless steels based on type 316 at low temperatures Acta Mater. 56 3414–21
[69] Zuback J S and DebRoy T 2018 The hardness of additively manufactured alloys Materials 11 2070
[70] Adomako N K, Lewandowski J J, Arkhurst B M, Choi H, Chang H J and Kim J H 2022 Microstructures and mechanical properties of multi-layered materials composed of Ti-6Al-4V, vanadium, and 17–4PH stainless steel produced by directed energy deposition Addit. Manuf. 59 103174

Figure 5. Simulation of the molten pool under low-speed scanning (1.06 m/s). (a) Sequential solidification of the molten pool at the end of the melt track for laser powers of 190 and 340 W, respectively. (b) Recoil pressure on the molten pool at the keyhole for laser powers of 190 and 340 W, respectively. (c) The force diagram of the melt at the back of the keyhole at t = 750 μs in case B. (d) Temperature gradient at the solid–liquid interface of the molten pool at the moment the laser is deactivated in case A. (e) Temperature gradient at the solid–liquid interface of the molten pool at the moment the laser is deactivated in case B.

Revealing formation mechanism of end of processdepression in laser powder bed fusion by multiphysics meso-scale simulation

다중물리 메조 규모 시뮬레이션을 통해 레이저 분말층 융합에서 공정 종료의 함몰 형성 메커니즘 공개

Haodong Chen a,b, Xin Lin a,b,c, Yajing Sund, Shuhao Wanga,b, Kunpeng Zhu a,b,c and Binbin Dana,b

To link to this article: https://doi.org/10.1080/17452759.2024.2326599

ABSTRACT

Unintended end-of-process depression (EOPD) commonly occurs in laser powder bed fusion (LPBF), leading to poor surface quality and lower fatigue strength, especially for many implants. In this study, a high-fidelity multi-physics meso-scale simulation model is developed to uncover the forming mechanism of this defect. A defect-process map of the EOPD phenomenon is obtained using this simulation model. It is found that the EOPD formation mechanisms are different under distinct regions of process parameters. At low scanning speeds in keyhole mode, the long-lasting recoil pressure and the large temperature gradient easily induce EOPD. While at high scanning speeds in keyhole mode, the shallow molten pool morphology and the large solidification rate allow the keyhole to evolve into an EOPD quickly. Nevertheless, in the conduction mode, the Marangoni effects along with a faster solidification rate induce EOPD. Finally, a ‘step’ variable power strategy is proposed to optimise the EOPD defects for the case with high volumetric energy density at low scanning speeds. This work provides a profound understanding and valuable insights into the quality control of LPBF fabrication.

의도하지 않은 공정 종료 후 함몰(EOPD)은 LPBF(레이저 분말층 융합)에서 흔히 발생하며, 특히 많은 임플란트의 경우 표면 품질이 떨어지고 피로 강도가 낮아집니다. 본 연구에서는 이 결함의 형성 메커니즘을 밝히기 위해 충실도가 높은 다중 물리학 메조 규모 시뮬레이션 모델을 개발했습니다.

이 시뮬레이션 모델을 사용하여 EOPD 현상의 결함 프로세스 맵을 얻습니다. EOPD 형성 메커니즘은 공정 매개변수의 별개 영역에서 서로 다른 것으로 밝혀졌습니다.

키홀 모드의 낮은 스캔 속도에서는 오래 지속되는 반동 압력과 큰 온도 구배로 인해 EOPD가 쉽게 유발됩니다. 키홀 모드에서 높은 스캐닝 속도를 유지하는 동안 얕은 용융 풀 형태와 큰 응고 속도로 인해 키홀이 EOPD로 빠르게 진화할 수 있습니다.

그럼에도 불구하고 전도 모드에서는 더 빠른 응고 속도와 함께 마랑고니 효과가 EOPD를 유발합니다. 마지막으로, 낮은 스캐닝 속도에서 높은 체적 에너지 밀도를 갖는 경우에 대해 EOPD 결함을 최적화하기 위한 ‘단계’ 가변 전력 전략이 제안되었습니다.

이 작업은 LPBF 제조의 품질 관리에 대한 심오한 이해와 귀중한 통찰력을 제공합니다.

Figure 5. Simulation of the molten pool under low-speed scanning (1.06 m/s). (a) Sequential solidification of the molten pool at the
end of the melt track for laser powers of 190 and 340 W, respectively. (b) Recoil pressure on the molten pool at the keyhole for laser
powers of 190 and 340 W, respectively. (c) The force diagram of the melt at the back of the keyhole at t = 750 μs in case B. (d) Temperature gradient at the solid–liquid interface of the molten pool at the moment the laser is deactivated in case A. (e) Temperature
gradient at the solid–liquid interface of the molten pool at the moment the laser is deactivated in case B.
Figure 5. Simulation of the molten pool under low-speed scanning (1.06 m/s). (a) Sequential solidification of the molten pool at the end of the melt track for laser powers of 190 and 340 W, respectively. (b) Recoil pressure on the molten pool at the keyhole for laser powers of 190 and 340 W, respectively. (c) The force diagram of the melt at the back of the keyhole at t = 750 μs in case B. (d) Temperature gradient at the solid–liquid interface of the molten pool at the moment the laser is deactivated in case A. (e) Temperature gradient at the solid–liquid interface of the molten pool at the moment the laser is deactivated in case B.

References

[1] Zhang C, Li Z, Zhang J, et al. Additive manufacturing of magnesium matrix composites: comprehensive review of recent progress and research perspectives. J Mag
Alloys. 2023. doi:10.1016/j.jma.2023.02.005
[2] Webster S, Lin H, Carter III FM, et al. Physical mechanisms in hybrid additive manufacturing: a process design framework. J Mater Process Technol. 2022;291:117048. doi:10. 1016/j.jmatprotec.2021.117048
[3] Wang S, Ning J, Zhu L, et al. Role of porosity defects in metal 3D printing: formation mechanisms, impacts on properties and mitigation strategies. Mater Today. 2022. doi:10.1016/j.mattod.2022.08.014
[4] Wei C, Li L. Recent progress and scientific challenges in multi-material additive manufacturing via laser-based powder bed fusion. Virtual Phys Prototyp. 2021;16 (3):347–371. doi:10.1080/17452759.2021.1928520
[5] Lin X, Wang Q, Fuh JYH, et al. Motion feature based melt pool monitoring for selective laser melting process. J Mater Process Technol. 2022;303:117523. doi:10.1016/j. jmatprotec.2022.117523
[6] Gockel J, Sheridan L, Koerper B, et al. The influence of additive manufacturing processing parameters on surface roughness and fatigue life. Int J Fatigue. 2019;124:380–388. doi:10.1016/j.ijfatigue.2019.03.025
[7] Nicoletto G. Influence of rough as-built surfaces on smooth and notched fatigue behavior of L-PBF AlSi10Mg. Addit Manuf. 2020;34:101251. doi:10.1016/j. addma.2020.101251
[8] Spece H, Yu T, Law AW, et al. 3D printed porous PEEK created via fused filament fabrication for osteoconductive orthopaedic surfaces. J Mech Behav Biomed Mater. 2020;109:103850. doi:10.1115/1.0004270v
[9] Andrukhov O, Huber R, Shi B, et al. Proliferation, behavior, and differentiation of osteoblasts on surfaces of different microroughness. Dent Mater. 2016;32(11):1374–1384. doi:10.1016/j.dental.2016.08.217
[10] Dai N, Zhang LC, Zhang J, et al. Corrosion behavior of selective laser melted Ti-6Al-4 V alloy in NaCl solution. Corros Sci. 2016;102:484–489. doi:10.1016/j.corsci.2015. 10.041
[11] Li EL, Wang L, Yu AB, et al. A three-phase model for simulation of heat transfer and melt pool behaviour in laser powder bed fusion process. Powder Technol. 2021;381:298–312. doi:10.1016/j.powtec.2020.11.061
[12] Liao B, Xia RF, Li W, et al. 3D-printed ti6al4v scaffolds with graded triply periodic minimal surface structure for bone tissue engineering. J Mater Eng Perform. 2021;30:4993– 5004. doi:10.1007/s11665-021-05580-z
[13] Li E, Zhou Z, Wang L, et al. Melt pool dynamics and pores formation in multi-track studies in laser powder bed fusion process. Powder Technol. 2022;405:117533. doi:10.1016/j.powtec.2022.117533
[14] Guo L, Geng S, Gao X, et al. Numerical simulation of heat transfer and fluid flow during nanosecond pulsed laser processing of Fe78Si9B13 amorphous alloys. Int J Heat Mass Transfer. 2021;170:121003. doi:10.1016/j.ijheatma sstransfer.2021.121003
[15] Guo L, Li Y, Geng S, et al. Numerical and experimental analysis for morphology evolution of 6061 aluminum alloy during nanosecond pulsed laser cleaning. Surf Coat Technol. 2022;432:128056. doi:10.1016/j.surfcoat. 2021.128056
[16] Li S, Liu D, Mi H, et al. Numerical simulation on evolution process of molten pool and solidification characteristics of melt track in selective laser melting of ceramic powder. Ceram Int. 2022;48(13):18302–18315. doi:10. 1016/j.ceramint.2022.03.089
[17] Aboulkhair NT, Maskery I, Tuck C, et al. On the formation of AlSi10Mg single tracks and layers in selective laser melting: microstructure and nano-mechanical properties. J Mater Process Technol. 2016;230:88–98. doi:10.1016/j. jmatprotec.2015.11.016
[18] Thijs L, Kempen K, Kruth JP, et al. Fine-structured aluminium products with controllable texture by selective laser melting of pre-alloyed AlSi10Mg powder. Acta Mater. 2013;61(5):1809–1819. doi:10.1016/j.actamat.2012.11.052
[19] Qiu C, Adkins NJE, Attallah MM. Microstructure and tensile properties of selectively laser-melted and of HIPed laser-melted Ti–6Al–4 V. Mater Sci Eng A. 2013;578:230–239. doi:10.1016/j.msea.2013.04.099
[20] Kazemi Z, Soleimani M, Rokhgireh H, et al. Melting pool simulation of 316L samples manufactured by selective laser melting method, comparison with experimental results. Int J Therm Sci. 2022;176:107538. doi:10.1016/j. ijthermalsci.2022.107538
[21] Cao L. Workpiece-scale numerical simulations of SLM molten pool dynamic behavior of 316L stainless steel. Comput Math Appl. 2021;96:209–228. doi:10.1016/j. camwa.2020.04.020
[22] Liu B, Fang G, Lei L, et al. Predicting the porosity defects in selective laser melting (SLM) by molten pool geometry. Int J Mech Sci. 2022;228:107478. doi:10.1016/j.ijmecsci. 2022.107478
[23] Ur Rehman A, Pitir F, Salamci MU. Full-field mapping and flow quantification of melt pool dynamics in laser powder bed fusion of SS316L. Materials. 2021;14(21):6264. doi:10. 3390/ma14216264
[24] Chia HY, Wang L, Yan W. Influence of oxygen content on melt pool dynamics in metal additive manufacturing: high-fidelity modeling with experimental validation. Acta Mater. 2023;249:118824. doi:10.1016/j.actamat. 2023.118824
[25] Cheng B, Loeber L, Willeck H, et al. Computational investigation of melt pool process dynamics and pore formation in laser powder bed fusion. J Mater Eng Perform. 2019;28:6565–6578. doi:10.1007/s11665-019- 04435-y
[26] Li X, Guo Q, Chen L, et al. Quantitative investigation of gas flow, powder-gas interaction, and powder behavior under different ambient pressure levels in laser powder bed fusion. Int J Mach Tools Manuf. 2021;170:103797. doi:10.1016/j.ijmachtools.2021.103797
[27] Wu Y, Li M, Wang J, et al. Powder-bed-fusion additive manufacturing of molybdenum: process simulation, optimization, and property prediction. Addit Manuf. 2022;58:103069. doi:10.1016/j.addma.2022.103069
[28] Wu S, Yang Y, Huang Y, et al. Study on powder particle behavior in powder spreading with discrete element method and its critical implications for binder jetting additive manufacturing processes. Virtual Phys Prototyp. 2023;18(1):e2158877. doi:10.1080/17452759.2022.2158877
[29] Klassen A, Schakowsky T, Kerner C. Evaporation model for beam based additive manufacturing using free surface lattice Boltzmann methods. J Phys D Appl Phys. 2014;47 (27):275303. doi:10.1088/0022-3727/47/27/275303
[30] Cao L. Mesoscopic-scale numerical simulation including the influence of process parameters on slm single-layer multi-pass formation. Metall Mater Trans A. 2020;51:4130–4145. doi:10.1007/s11661-020-05831-z
[31] Zhuang JR, Lee YT, Hsieh WH, et al. Determination of melt pool dimensions using DOE-FEM and RSM with process window during SLM of Ti6Al4V powder. Opt Laser Technol. 2018;103:59–76. doi:10.1016/j.optlastec.2018. 01.013
[32] Li Y, Gu D. Thermal behavior during selective laser melting of commercially pure titanium powder: numerical simulation and experimental study. Addit Manuf. 2014;1–4:99–109. doi:10.1016/j.addma.2014.09.001
[33] Dai D, Gu D. Thermal behavior and densification mechanism during selective laser melting of copper matrix composites: simulation and experiments. Mater Des. 2014;55 (0):482–491. doi:10.1016/j.matdes.2013.10.006
[34] Wang S, Zhu L, Dun Y, et al. Multi-physics modeling of direct energy deposition process of thin-walled structures: defect analysis. Comput Mech. 2021;67:c1229– c1242. doi:10.1007/s00466-021-01992-9
[35] Wu J, Zheng J, Zhou H, et al. Molten pool behavior and its mechanism during selective laser melting of polyamide 6 powder: single track simulation and experiments. Mater Res Express. 2019;6. doi:10.1088/2053-1591/ab2747
[36] Cho JH, Farson DF, Milewski JO, et al. Weld pool flows during initial stages of keyhole formation in laser welding. J Phys D Appl Phys. 2009;42. doi:10.1088/0022- 3727/42/17/175502
[37] Sinha KN. Identification of a suitable volumetric heat source for modelling of selective laser melting of Ti6Al4V powder using numerical and experimental validation approach. Int J Adv Manuf Technol. 2018;99:2257–2270. doi:10.1007/s00170-018-2631-4
[38] Fu CH, Guo YB. Three-dimensional temperature gradient mechanism in selective laser melting of Ti-6Al-4V. J Manuf Sci Eng. 2014;136(6):061004. doi:10.1115/1.4028539
[39] Ansari P, Rehman AU, Pitir F, et al. Selective laser melting of 316 l austenitic stainless steel: detailed process understanding using multiphysics simulation and experimentation. Metals. 2021;11(7):1076. doi:10.3390/met11071076
[40] Zhao C, Shi B, Chen S, et al. Laser melting modes in metal powder bed fusion additive manufacturing. Rev Mod Phys. 2022;94(4):045002. doi:10.1103/revmodphys.94. 045002
[41] Bertoli US, Wolfer AJ, Matthews MJ, et al. On the limitations of volumetric energy density as a design parameter for selective laser melting. Mater Des. 2017;113:331–340. doi:10.1016/j.matdes.2016.10.037
[42] Dash A, Kamaraj A. Prediction of the shift in melting mode during additive manufacturing of 316L stainless steel. Mater Today Commun. 2023: 107238. doi:10.1016/j. mtcomm.2023.107238
[43] Majeed M, Khan HM, Rasheed I. Finite element analysis of melt pool thermal characteristics with passing laser in SLM process. Optik. 2019;194:163068. doi:10.1016/j.ijleo. 2019.163068

Schematic diagram of HP-LPBF melting process.

Modeling and numerical studies of high-precision laser powder bed fusion

Yi Wei ;Genyu Chen;Nengru Tao;Wei Zhou
https://doi.org/10.1063/5.0191504

In order to comprehensively reveal the evolutionary dynamics of the molten pool and the state of motion of the fluid during the high-precision laser powder bed fusion (HP-LPBF) process, this study aims to deeply investigate the specific manifestations of the multiphase flow, solidification phenomena, and heat transfer during the process by means of numerical simulation methods. Numerical simulation models of SS316L single-layer HP-LPBF formation with single and double tracks were constructed using the discrete element method and the computational fluid dynamics method. The effects of various factors such as Marangoni convection, surface tension, vapor recoil, gravity, thermal convection, thermal radiation, and evaporative heat dissipation on the heat and mass transfer in the molten pool have been paid attention to during the model construction process. The results show that the molten pool exhibits a “comet” shape, in which the temperature gradient at the front end of the pool is significantly larger than that at the tail end, with the highest temperature gradient up to 1.69 × 108 K/s. It is also found that the depth of the second track is larger than that of the first one, and the process parameter window has been determined preliminarily. In addition, the application of HP-LPBF technology helps to reduce the surface roughness and minimize the forming size.

Topics

Heat transferNonequilibrium thermodynamicsSolidification processComputer simulationDiscrete element methodLasersMass transferFluid mechanicsComputational fluid dynamicsMultiphase flows

I. INTRODUCTION

Laser powder bed fusion (LPBF) has become a research hotspot in the field of additive manufacturing of metals due to its advantages of high-dimensional accuracy, good surface quality, high density, and high material utilization.1,2 With the rapid development of electronics, medical, automotive, biotechnology, energy, communication, and optics, the demand for microfabrication technology is increasing day by day.3 High-precision laser powder bed fusion (HP-LPBF) is one of the key manufacturing technologies for tiny parts in the fields of electronics, medical, automotive, biotechnology, energy, communication, and optics because of its process characteristics such as small focal spot diameter, small powder particle size, and thin powder layup layer thickness.4–13 Compared with LPBF, HP-LPBF has the significant advantages of smaller focal spot diameter, smaller powder particle size, and thinner layer thickness. These advantages make HP-LPBF perform better in producing micro-fine parts, high surface quality, and parts with excellent mechanical properties.

HP-LPBF is in the exploratory stage, and researchers have already done some exploratory studies on the focal spot diameter, the amount of defocusing, and the powder particle size. In order to explore the influence of changing the laser focal spot diameter on the LPBF process characteristics of the law, Wildman et al.14 studied five groups of different focal spot diameter LPBF forming 316L stainless steel (SS316L) processing effect, the smallest focal spot diameter of 26 μm, and the results confirm that changing the focal spot diameter can be achieved to achieve the energy control, so as to control the quality of forming. Subsequently, Mclouth et al.15 proposed the laser out-of-focus amount (focal spot diameter) parameter, which characterizes the distance between the forming plane and the laser focal plane. The laser energy density was controlled by varying the defocusing amount while keeping the laser parameters constant. Sample preparation at different focal positions was investigated, and their microstructures were characterized. The results show that the samples at the focal plane have finer microstructure than those away from the focal plane, which is the effect of higher power density and smaller focal spot diameter. In order to explore the influence of changing the powder particle size on the characteristics of the LPBF process, Qian et al.16 carried out single-track scanning simulations on powder beds with average powder particle sizes of 70 and 40 μm, respectively, and the results showed that the melt tracks sizes were close to each other under the same process parameters for the two particle-size distributions and that the molten pool of powder beds with small particles was more elongated and the edges of the melt tracks were relatively flat. In order to explore the superiority of HP-LPBF technology, Xu et al.17 conducted a comparative analysis of HP-LPBF and conventional LPBF of SS316L. The results showed that the average surface roughness of the top surface after forming by HP-LPBF could reach 3.40 μm. Once again, it was verified that HP-LPBF had higher forming quality than conventional LPBF. On this basis, Wei et al.6 comparatively analyzed the effects of different laser focal spot diameters on different powder particle sizes formed by LPBF. The results showed that the smaller the laser focal spot diameter, the fewer the defects on the top and side surfaces. The above research results confirm that reducing the laser focal spot diameter can obtain higher energy density and thus better forming quality.

LPBF involves a variety of complex systems and mechanisms, and the final quality of the part is influenced by a large number of process parameters.18–24 Some research results have shown that there are more than 50 factors affecting the quality of the specimen. The influencing factors are mainly categorized into three main groups: (1) laser parameters, (2) powder parameters, and (3) equipment parameters, which interact with each other to determine the final specimen quality. With the continuous development of technologies such as computational materials science and computational fluid dynamics (CFD), the method of studying the influence of different factors on the forming quality of LPBF forming process has been shifted from time-consuming and laborious experimental characterization to the use of numerical simulation methods. As a result, more and more researchers are adopting this approach for their studies. Currently, numerical simulation studies on LPBF are mainly focused on the exploration of molten pool, temperature distribution, and residual stresses.

  1. Finite element simulation based on continuum mechanics and free surface fluid flow modeling based on fluid dynamics are two common approaches to study the behavior of LPBF molten pool.25–28 Finite element simulation focuses on the temperature and thermal stress fields, treats the powder bed as a continuum, and determines the molten pool size by plotting the elemental temperature above the melting point. In contrast, fluid dynamics modeling can simulate the 2D or 3D morphology of the metal powder pile and obtain the powder size and distribution by certain algorithms.29 The flow in the molten pool is mainly affected by recoil pressure and the Marangoni effect. By simulating the molten pool formation, it is possible to predict defects, molten pool shape, and flow characteristics, as well as the effect of process parameters on the molten pool geometry.30–34 In addition, other researchers have been conducted to optimize the laser processing parameters through different simulation methods and experimental data.35–46 Crystal growth during solidification is studied to further understand the effect of laser parameters on dendritic morphology and solute segregation.47–54 A multi-scale system has been developed to describe the fused deposition process during 3D printing, which is combined with the conductive heat transfer model and the dendritic solidification model.55,56
  2. Relevant scholars have adopted various different methods for simulation, such as sequential coupling theory,57 Lagrangian and Eulerian thermal models,58 birth–death element method,25 and finite element method,59 in order to reveal the physical phenomena of the laser melting process and optimize the process parameters. Luo et al.60 compared the LPBF temperature field and molten pool under double ellipsoidal and Gaussian heat sources by ANSYS APDL and found that the diffusion of the laser energy in the powder significantly affects the molten pool size and the temperature field.
  3. The thermal stresses obtained from the simulation correlate with the actual cracks,61 and local preheating can effectively reduce the residual stresses.62 A three-dimensional thermodynamic finite element model investigated the temperature and stress variations during laser-assisted fabrication and found that powder-to-solid conversion increases the temperature gradient, stresses, and warpage.63 Other scholars have predicted residual stresses and part deflection for LPBF specimens and investigated the effects of deposition pattern, heat, laser power, and scanning strategy on residual stresses, noting that high-temperature gradients lead to higher residual stresses.64–67 

In short, the process of LPBF forming SS316L is extremely complex and usually involves drastic multi-scale physicochemical changes that will only take place on a very small scale. Existing literature employs DEM-based mesoscopic-scale numerical simulations to investigate the effects of process parameters on the molten pool dynamics of LPBF-formed SS316L. However, a few studies have been reported on the key mechanisms of heating and solidification, spatter, and convective behavior of the molten pool of HP-LPBF-formed SS316L with small laser focal spot diameters. In this paper, the geometrical properties of coarse and fine powder particles under three-dimensional conditions were first calculated using DEM. Then, numerical simulation models for single-track and double-track cases in the single-layer HP-LPBF forming SS316L process were developed at mesoscopic scale using the CFD method. The flow genesis of the melt in the single-track and double-track molten pools is discussed, and their 3D morphology and dimensional characteristics are discussed. In addition, the effects of laser process parameters, powder particle size, and laser focal spot diameter on the temperature field, characterization information, and defects in the molten pool are discussed.

II. MODELING

A. 3D powder bed modeling

HP-LPBF is an advanced processing technique for preparing target parts layer by layer stacking, the process of which involves repetitive spreading and melting of powders. In this process, both the powder spreading and the morphology of the powder bed are closely related to the results of the subsequent melting process, while the melted surface also affects the uniform distribution of the next layer of powder. For this reason, this chapter focuses on the modeling of the physical action during the powder spreading process and the theory of DEM to establish the numerical model of the powder bed, so as to lay a solid foundation for the accuracy of volume of fluid (VOF) and CFD.

1. DEM

DEM is a numerical technique for calculating the interaction of a large number of particles, which calculates the forces and motions of the spheres by considering each powder sphere as an independent unit. The motion of the powder particles follows the laws of classical Newtonian mechanics, including translational and rotational,38,68–70 which are expressed as follows:����¨=���+∑��ij,

(1)����¨=∑�(�ij×�ij),

(2)

where �� is the mass of unit particle i in kg, ��¨ is the advective acceleration in m/s2, And g is the gravitational acceleration in m/s2. �ij is the force in contact with the neighboring particle � in N. �� is the rotational inertia of the unit particle � in kg · m2. ��¨ is the unit particle � angular acceleration in rad/s2. �ij is the vector pointing from unit particle � to the contact point of neighboring particle �⁠.

Equations (1) and (2) can be used to calculate the velocity and angular velocity variations of powder particles to determine their positions and velocities. A three-dimensional powder bed model of SS316L was developed using DEM. The powder particles are assumed to be perfect spheres, and the substrate and walls are assumed to be rigid. To describe the contact between the powder particles and between the particles and the substrate, a non-slip Hertz–Mindlin nonlinear spring-damping model71 was used with the following expression:�hz=��������+��[(�����ij−�eff����)−(�����+�eff����)],

(3)

where �hz is the force calculated using the Hertzian in M. �� and �� are the radius of unit particles � and � in m, respectively. �� is the overlap size of the two powder particles in m. ��⁠, �� are the elastic constants in the normal and tangential directions, respectively. �ij is the unit vector connecting the centerlines of the two powder particles. �eff is the effective mass of the two powder particles in kg. �� and �� are the viscoelastic damping constants in the normal and tangential directions, respectively. �� and �� are the components of the relative velocities of the two powder particles. ��� is the displacement vector between two spherical particles. The schematic diagram of overlapping powder particles is shown in Fig. 1.

FIG. 1.

VIEW LARGEDOWNLOAD SLIDE

Schematic diagram of overlapping powder particles.

Because the particle size of the powder used for HP-LPBF is much smaller than 100 μm, the effect of van der Waals forces must be considered. Therefore, the cohesive force �jkr of the Hertz–Mindlin model was used instead of van der Waals forces,72 with the following expression:�jkr=−4��0�*�1.5+4�*3�*�3,

(4)1�*=(1−��2)��+(1−��2)��,

(5)1�*=1��+1��,

(6)

where �* is the equivalent Young’s modulus in GPa; �* is the equivalent particle radius in m; �0 is the surface energy of the powder particles in J/m2; α is the contact radius in m; �� and �� are the Young’s modulus of the unit particles � and �⁠, respectively, in GPa; and �� and �� are the Poisson’s ratio of the unit particles � and �⁠, respectively.

2. Model building

Figure 2 shows a 3D powder bed model generated using DEM with a coarse powder geometry of 1000 × 400 × 30 μm3. The powder layer thickness is 30 μm, and the powder bed porosity is 40%. The average particle size of this spherical powder is 31.7 μm and is normally distributed in the range of 15–53 μm. The geometry of the fine powder was 1000 × 400 × 20 μm3, with a layer thickness of 20 μm, and the powder bed porosity of 40%. The average particle size of this spherical powder is 11.5 μm and is normally distributed in the range of 5–25 μm. After the 3D powder bed model is generated, it needs to be imported into the CFD simulation software for calculation, and the imported geometric model is shown in Fig. 3. This geometric model is mainly composed of three parts: protective gas, powder bed, and substrate. Under the premise of ensuring the accuracy of the calculation, the mesh size is set to 3 μm, and the total number of coarse powder meshes is 1 704 940. The total number of fine powder meshes is 3 982 250.

FIG. 2.

VIEW LARGEDOWNLOAD SLIDE

Three-dimensional powder bed model: (a) coarse powder, (b) fine powder.

FIG. 3.

VIEW LARGEDOWNLOAD SLIDE

Geometric modeling of the powder bed computational domain: (a) coarse powder, (b) fine powder.

B. Modeling of fluid mechanics simulation

In order to solve the flow, melting, and solidification problems involved in HP-LPBF molten pool, the study must follow the three governing equations of conservation of mass, conservation of energy, and conservation of momentum.73 The VOF method, which is the most widely used in fluid dynamics, is used to solve the molten pool dynamics model.

1. VOF

VOF is a method for tracking the free interface between the gas and liquid phases on the molten pool surface. The core idea of the method is to define a volume fraction function F within each grid, indicating the proportion of the grid space occupied by the material, 0 ≤ F ≤ 1 in Fig. 4. Specifically, when F = 0, the grid is empty and belongs to the gas-phase region; when F = 1, the grid is completely filled with material and belongs to the liquid-phase region; and when 0 < F < 1, the grid contains free surfaces and belongs to the mixed region. The direction normal to the free surface is the direction of the fastest change in the volume fraction F (the direction of the gradient of the volume fraction), and the direction of the gradient of the volume fraction can be calculated from the values of the volume fractions in the neighboring grids.74 The equations controlling the VOF are expressed as follows:𝛻����+�⋅(��→)=0,

(7)

where t is the time in s and �→ is the liquid velocity in m/s.

FIG. 4.

VIEW LARGEDOWNLOAD SLIDE

Schematic diagram of VOF.

The material parameters of the mixing zone are altered due to the inclusion of both the gas and liquid phases. Therefore, in order to represent the density of the mixing zone, the average density �¯ is used, which is expressed as follows:72�¯=(1−�1)�gas+�1�metal,

(8)

where �1 is the proportion of liquid phase, �gas is the density of protective gas in kg/m3, and �metal is the density of metal in kg/m3.

2. Control equations and boundary conditions

Figure 5 is a schematic diagram of the HP-LPBF melting process. First, the laser light strikes a localized area of the material and rapidly heats up the area. Next, the energy absorbed in the region is diffused through a variety of pathways (heat conduction, heat convection, and surface radiation), and this process triggers complex phase transition phenomena (melting, evaporation, and solidification). In metals undergoing melting, the driving forces include surface tension and the Marangoni effect, recoil due to evaporation, and buoyancy due to gravity and uneven density. The above physical phenomena interact with each other and do not occur independently.

FIG. 5.

VIEW LARGEDOWNLOAD SLIDE

Schematic diagram of HP-LPBF melting process.

  1. Laser heat sourceThe Gaussian surface heat source model is used as the laser heat source model with the following expression:�=2�0����2exp(−2�12��2),(9)where � is the heat flow density in W/m2, �0 is the absorption rate of SS316L, �� is the radius of the laser focal spot in m, and �1 is the radial distance from the center of the laser focal spot in m. The laser focal spot can be used for a wide range of applications.
  2. Energy absorptionThe formula for calculating the laser absorption �0 of SS316L is as follows:�0=0.365(�0[1+�0(�−20)]/�)0.5,(10)where �0 is the direct current resistivity of SS316L at 20 °C in Ω m, �0 is the resistance temperature coefficient in ppm/°C, � is the temperature in °C, and � is the laser wavelength in m.
  3. Heat transferThe basic principle of heat transfer is conservation of energy, which is expressed as follows:𝛻𝛻𝛻�(��)��+�·(��→�)=�·(�0����)+��,(11)where � is the density of liquid phase SS316L in kg/m3, �� is the specific heat capacity of SS316L in J/(kg K), 𝛻� is the gradient operator, t is the time in s, T is the temperature in K, 𝛻�� is the temperature gradient, �→ is the velocity vector, �0 is the coefficient of thermal conduction of SS316L in W/(m K), and  �� is the thermal energy dissipation term in the molten pool.
  4. Molten pool flowThe following three conditions need to be satisfied for the molten pool to flow:
    • Conservation of mass with the following expression:𝛻�·(��→)=0.(12)
    • Conservation of momentum (Navier–Stokes equation) with the following expression:𝛻𝛻𝛻𝛻���→��+�(�→·�)�→=�·[−pI+�(��→+(��→)�)]+�,(13)where � is the pressure in Pa exerted on the liquid phase SS316L microelement, � is the unit matrix, � is the fluid viscosity in N s/m2, and � is the volumetric force (gravity, atmospheric pressure, surface tension, vapor recoil, and the Marangoni effect).
    • Conservation of energy, see Eq. (11)
  5. Surface tension and the Marangoni effectThe effect of temperature on the surface tension coefficient is considered and set as a linear relationship with the following expression:�=�0−��dT(�−��),(14)where � is the surface tension of the molten pool at temperature T in N/m, �� is the melting temperature of SS316L in K, �0 is the surface tension of the molten pool at temperature �� in Pa, and σdσ/ dT is the surface tension temperature coefficient in N/(m K).In general, surface tension decreases with increasing temperature. A temperature gradient causes a gradient in surface tension that drives the liquid to flow, known as the Marangoni effect.
  6. Metal vapor recoilAt higher input energy densities, the maximum temperature of the molten pool surface reaches the evaporation temperature of the material, and a gasification recoil pressure occurs vertically downward toward the molten pool surface, which will be the dominant driving force for the molten pool flow.75 The expression is as follows:��=0.54�� exp ���−���0���,(15)where �� is the gasification recoil pressure in Pa, �� is the ambient pressure in kPa, �� is the latent heat of evaporation in J/kg, �0 is the gas constant in J/(mol K), T is the surface temperature of the molten pool in K, and Te is the evaporation temperature in K.
  7. Solid–liquid–gas phase transitionWhen the laser hits the powder layer, the powder goes through three stages: heating, melting, and solidification. During the solidification phase, mutual transformations between solid, liquid, and gaseous states occur. At this point, the latent heat of phase transition absorbed or released during the phase transition needs to be considered.68 The phase transition is represented based on the relationship between energy and temperature with the following expression:�=�����,(�<��),�(��)+�−����−����,(��<�<��)�(��)+(�−��)����,(��<�),,(16)where �� and �� are solid and liquid phase density, respectively, of SS316L in kg/m3. �� and �� unit volume of solid and liquid phase-specific heat capacity, respectively, of SS316L in J/(kg K). �� and ��⁠, respectively, are the solidification temperature and melting temperature of SS316L in K. �� is the latent heat of the phase transition of SS316L melting in J/kg.

3. Assumptions

The CFD model was computed using the commercial software package FLOW-3D.76 In order to simplify the calculation and solution process while ensuring the accuracy of the results, the model makes the following assumptions:

  1. It is assumed that the effects of thermal stress and material solid-phase thermal expansion on the calculation results are negligible.
  2. The molten pool flow is assumed to be a Newtonian incompressible laminar flow, while the effects of liquid thermal expansion and density on the results are neglected.
  3. It is assumed that the surface tension can be simplified to an equivalent pressure acting on the free surface of the molten pool, and the effect of chemical composition on the results is negligible.
  4. Neglecting the effect of the gas flow field on the molten pool.
  5. The mass loss due to evaporation of the liquid metal is not considered.
  6. The influence of the plasma effect of the molten metal on the calculation results is neglected.

It is worth noting that the formulation of assumptions requires a trade-off between accuracy and computational efficiency. In the above models, some physical phenomena that have a small effect or high difficulty on the calculation results are simplified or ignored. Such simplifications make numerical simulations more efficient and computationally tractable, while still yielding accurate results.

4. Initial conditions

The preheating temperature of the substrate was set to 393 K, at which time all materials were in the solid state and the flow rate was zero.

5. Material parameters

The material used is SS316L and the relevant parameters required for numerical simulations are shown in Table I.46,77,78

TABLE I.

SS316L-related parameters.

PropertySymbolValue
Density of solid metal (kg/m3�metal 7980 
Solid phase line temperature (K) �� 1658 
Liquid phase line temperature (K) �� 1723 
Vaporization temperature (K) �� 3090 
Latent heat of melting (⁠ J/kg⁠) �� 2.60×105 
Latent heat of evaporation (⁠ J/kg⁠) �� 7.45×106 
Surface tension of liquid phase (N /m⁠) � 1.60 
Liquid metal viscosity (kg/m s) �� 6×10−3 
Gaseous metal viscosity (kg/m s) �gas 1.85×10−5 
Temperature coefficient of surface tension (N/m K) ��/�T 0.80×10−3 
Molar mass (⁠ kg/mol⁠) 0.05 593 
Emissivity � 0.26 
Laser absorption �0 0.35 
Ambient pressure (kPa) �� 101 325 
Ambient temperature (K) �0 300 
Stefan–Boltzmann constant (W/m2 K4� 5.67×10−8 
Thermal conductivity of metals (⁠ W/m K⁠) � 24.55 
Density of protective gas (kg/m3�gas 1.25 
Coefficient of thermal expansion (/K) �� 16×10−6 
Generalized gas constant (⁠ J/mol K⁠) 8.314 

III. RESULTS AND DISCUSSION

With the objective of studying in depth the evolutionary patterns of single-track and double-track molten pool development, detailed observations were made for certain specific locations in the model, as shown in Fig. 6. In this figure, P1 and P2 represent the longitudinal tangents to the centers of the two melt tracks in the XZ plane, while L1 is the transverse profile in the YZ plane. The scanning direction is positive and negative along the X axis. Points A and B are the locations of the centers of the molten pool of the first and second melt tracks, respectively (x = 1.995 × 10−4, y = 5 × 10−7, and z = −4.85 × 10−5).

FIG. 6.

VIEW LARGEDOWNLOAD SLIDE

Schematic diagram of observation position.

A. Single-track simulation

A series of single-track molten pool simulation experiments were carried out in order to investigate the influence law of laser power as well as scanning speed on the HP-LPBF process. Figure 7 demonstrates the evolution of the 3D morphology and temperature field of the single-track molten pool in the time period of 50–500 μs under a laser power of 100 W and a scanning speed of 800 mm/s. The powder bed is in the natural cooling state. When t = 50 μs, the powder is heated by the laser heat and rapidly melts and settles to form the initial molten pool. This process is accompanied by partial melting of the substrate and solidification together with the melted powder. The molten pool rapidly expands with increasing width, depth, length, and temperature, as shown in Fig. 7(a). When t = 150 μs, the molten pool expands more obviously, and the temperature starts to transfer to the surrounding area, forming a heat-affected zone. At this point, the width of the molten pool tends to stabilize, and the temperature in the center of the molten pool has reached its peak and remains largely stable. However, the phenomenon of molten pool spatter was also observed in this process, as shown in Fig. 7(b). As time advances, when t = 300 μs, solidification begins to occur at the tail of the molten pool, and tiny ripples are produced on the solidified surface. This is due to the fact that the melt flows toward the region with large temperature gradient under the influence of Marangoni convection and solidifies together with the melt at the end of the bath. At this point, the temperature gradient at the front of the bath is significantly larger than at the end. While the width of the molten pool was gradually reduced, the shape of the molten pool was gradually changed to a “comet” shape. In addition, a slight depression was observed at the top of the bath because the peak temperature at the surface of the bath reached the evaporation temperature, which resulted in a recoil pressure perpendicular to the surface of the bath downward, creating a depressed region. As the laser focal spot moves and is paired with the Marangoni convection of the melt, these recessed areas will be filled in as shown in Fig. 7(c). It has been shown that the depressed regions are the result of the coupled effect of Marangoni convection, recoil pressure, and surface tension.79 By t = 500 μs, the width and height of the molten pool stabilize and show a “comet” shape in Fig. 7(d).

FIG. 7.

VIEW LARGEDOWNLOAD SLIDE

Single-track molten pool process: (a) t = 50  ��⁠, (b) t = 150  ��⁠, (c) t = 300  ��⁠, (d) t = 500  ��⁠.

Figure 8 depicts the velocity vector diagram of the P1 profile in a single-track molten pool, the length of the arrows represents the magnitude of the velocity, and the maximum velocity is about 2.36 m/s. When t = 50 μs, the molten pool takes shape, and the velocities at the two ends of the pool are the largest. The variation of the velocities at the front end is especially more significant in Fig. 8(a). As the time advances to t = 150 μs, the molten pool expands rapidly, in which the velocity at the tail increases and changes more significantly, while the velocity at the front is relatively small. At this stage, the melt moves backward from the center of the molten pool, which in turn expands the molten pool area. The melt at the back end of the molten pool center flows backward along the edge of the molten pool surface and then converges along the edge of the molten pool to the bottom center, rising to form a closed loop. Similarly, a similar closed loop is formed at the front end of the center of the bath, but with a shorter path. However, a large portion of the melt in the center of the closed loop formed at the front end of the bath is in a nearly stationary state. The main cause of this melt flow phenomenon is the effect of temperature gradient and surface tension (the Marangoni effect), as shown in Figs. 8(b) and 8(e). This dynamic behavior of the melt tends to form an “elliptical” pool. At t = 300 μs, the tendency of the above two melt flows to close the loop is more prominent and faster in Fig. 8(c). When t = 500 μs, the velocity vector of the molten pool shows a stable trend, and the closed loop of melt flow also remains stable. With the gradual laser focal spot movement, the melt is gradually solidified at its tail, and finally, a continuous and stable single track is formed in Fig. 8(d).

FIG. 8.

VIEW LARGEDOWNLOAD SLIDE

Vector plot of single-track molten pool velocity in XZ longitudinal section: (a) t = 50  ��⁠, (b) t = 150  ��⁠, (c) t = 300  ��⁠, (d) t = 500  ��⁠, (e) molten pool flow.

In order to explore in depth the transient evolution of the molten pool, the evolution of the single-track temperature field and the melt flow was monitored in the YZ cross section. Figure 9(a) shows the state of the powder bed at the initial moment. When t = 250 μs, the laser focal spot acts on the powder bed and the powder starts to melt and gradually collects in the molten pool. At this time, the substrate will also start to melt, and the melt flow mainly moves in the downward and outward directions and the velocity is maximum at the edges in Fig. 9(b). When t = 300 μs, the width and depth of the molten pool increase due to the recoil pressure. At this time, the melt flows more slowly at the center, but the direction of motion is still downward in Fig. 9(c). When t = 350 μs, the width and depth of the molten pool further increase, at which time the intensity of the melt flow reaches its peak and the direction of motion remains the same in Fig. 9(d). When t = 400 μs, the melt starts to move upward, and the surrounding powder or molten material gradually fills up, causing the surface of the molten pool to begin to flatten. At this time, the maximum velocity of the melt is at the center of the bath, while the velocity at the edge is close to zero, and the edge of the melt starts to solidify in Fig. 9(e). When t = 450 μs, the melt continues to move upward, forming a convex surface of the melt track. However, the melt movement slows down, as shown in Fig. 9(f). When t = 500 μs, the melt further moves upward and its speed gradually becomes smaller. At the same time, the melt solidifies further, as shown in Fig. 9(g). When t = 550 μs, the melt track is basically formed into a single track with a similar “mountain” shape. At this stage, the velocity is close to zero only at the center of the molten pool, and the flow behavior of the melt is poor in Fig. 9(h). At t = 600 μs, the melt stops moving and solidification is rapidly completed. Up to this point, a single track is formed in Fig. 9(i). During the laser action on the powder bed, the substrate melts and combines with the molten state powder. The powder-to-powder fusion is like the convergence of water droplets, which are rapidly fused by surface tension. However, the fusion between the molten state powder and the substrate occurs driven by surface tension, and the molten powder around the molten pool is pulled toward the substrate (a wetting effect occurs), which ultimately results in the formation of a monolithic whole.38,80,81

FIG. 9.

VIEW LARGEDOWNLOAD SLIDE

Evolution of single-track molten pool temperature and melt flow in the YZ cross section: (a) t = 0  ��⁠, (b) t = 250  ��⁠, (c) t = 300  ��⁠, (d) t = 350  ��⁠, (e) t = 400  ��⁠, (f) t = 450  ��⁠, (g) t = 500  ��⁠, (h) t = 550  ��⁠, (i) t = 600  ��⁠.

The wetting ability between the liquid metal and the solid substrate in the molten pool directly affects the degree of balling of the melt,82,83 and the wetting ability can be measured by the contact angle of a single track in Fig. 10. A smaller value of contact angle represents better wettability. The contact angle α can be calculated by�=�1−�22,

(17)

where �1 and �2 are the contact angles of the left and right regions, respectively.

FIG. 10.

VIEW LARGEDOWNLOAD SLIDE

Schematic of contact angle.

Relevant studies have confirmed that the wettability is better at a contact angle α around or below 40°.84 After measurement, a single-track contact angle α of about 33° was obtained under this process parameter, which further confirms the good wettability.

B. Double-track simulation

In order to deeply investigate the influence of hatch spacing on the characteristics of the HP-LPBF process, a series of double-track molten pool simulation experiments were systematically carried out. Figure 11 shows in detail the dynamic changes of the 3D morphology and temperature field of the double-track molten pool in the time period of 2050–2500 μs under the conditions of laser power of 100 W, scanning speed of 800 mm/s, and hatch spacing of 0.06 mm. By comparing the study with Fig. 7, it is observed that the basic characteristics of the 3D morphology and temperature field of the second track are similar to those of the first track. However, there are subtle differences between them. The first track exhibits a basically symmetric shape, but the second track morphology shows a slight deviation influenced by the difference in thermal diffusion rate between the solidified metal and the powder. Otherwise, the other characteristic information is almost the same as that of the first track. Figure 12 shows the velocity vector plot of the P2 profile in the double-track molten pool, with a maximum velocity of about 2.63 m/s. The melt dynamics at both ends of the pool are more stable at t = 2050 μs, where the maximum rate of the second track is only 1/3 of that of the first one. Other than that, the rest of the information is almost no significant difference from the characteristic information of the first track. Figure 13 demonstrates a detailed observation of the double-track temperature field and melts flow in the YZ cross section, and a comparative study with Fig. 9 reveals that the width of the second track is slightly wider. In addition, after the melt direction shifts from bottom to top, the first track undergoes four time periods (50 μs) to reach full solidification, while the second track takes five time periods. This is due to the presence of significant heat buildup in the powder bed after the forming of the first track, resulting in a longer dynamic time of the melt and an increased molten pool lifetime. In conclusion, the level of specimen forming can be significantly optimized by adjusting the laser power and hatch spacing.

FIG. 11.

VIEW LARGEDOWNLOAD SLIDE

Double-track molten pool process: (a) t = 2050  ��⁠, (b) t = 2150  ��⁠, (c) t = 2300  ��⁠, (d) t = 2500  ��⁠.

FIG. 12.

VIEW LARGEDOWNLOAD SLIDE

Vector plot of double-track molten pool velocity in XZ longitudinal section: (a) t = 2050  ��⁠, (b) t = 2150  ��⁠, (c) t = 2300  ��⁠, (d) t = 2500  ��⁠.

FIG. 13.

VIEW LARGEDOWNLOAD SLIDE

Evolution of double-track molten pool temperature and melt flow in the YZ cross section: (a) t = 2250  ��⁠, (b) t = 2300  ��⁠, (c) t = 2350  ��⁠, (d) t = 2400  ��⁠, (e) t = 2450  ��⁠, (f) t = 2500  ��⁠, (g) t = 2550  ��⁠, (h) t = 2600  ��⁠, (i) t = 2650  ��⁠.

In order to quantitatively detect the molten pool dimensions as well as the remolten region dimensions, the molten pool characterization information in Fig. 14 is constructed by drawing the boundary on the YZ cross section based on the isothermal surface of the liquid phase line. It can be observed that the heights of the first track and second track are basically the same, but the depth of the second track increases relative to the first track. The molten pool width is mainly positively correlated with the laser power as well as the scanning speed (the laser line energy density �⁠). However, the remelted zone width is negatively correlated with the hatch spacing (the overlapping ratio). Overall, the forming quality of the specimens can be directly influenced by adjusting the laser power, scanning speed, and hatch spacing.

FIG. 14.

VIEW LARGEDOWNLOAD SLIDE

Double-track molten pool characterization information on YZ cross section.

In order to study the variation rule of the temperature in the center of the molten pool with time, Fig. 15 demonstrates the temperature variation curves with time for two reference points, A and B. Among them, the red dotted line indicates the liquid phase line temperature of SS316L. From the figure, it can be seen that the maximum temperature at the center of the molten pool in the first track is lower than that in the second track, which is mainly due to the heat accumulation generated after passing through the first track. The maximum temperature gradient was calculated to be 1.69 × 108 K/s. When the laser scanned the first track, the temperature in the center of the molten pool of the second track increased slightly. Similarly, when the laser scanned the second track, a similar situation existed in the first track. Since the temperature gradient in the second track is larger than that in the first track, the residence time of the liquid phase in the molten pool of the first track is longer than that of the second track.

FIG. 15.

VIEW LARGEDOWNLOAD SLIDE

Temperature profiles as a function of time for two reference points A and B.

C. Simulation analysis of molten pool under different process parameters

In order to deeply investigate the effects of various process parameters on the mesoscopic-scale temperature field, molten pool characteristic information and defects of HP-LPBF, numerical simulation experiments on mesoscopic-scale laser power, scanning speed, and hatch spacing of double-track molten pools were carried out.

1. Laser power

Figure 16 shows the effects of different laser power on the morphology and temperature field of the double-track molten pool at a scanning speed of 800 mm/s and a hatch spacing of 0.06 mm. When P = 50 W, a smaller molten pool is formed due to the lower heat generated by the Gaussian light source per unit time. This leads to a smaller track width, which results in adjacent track not lapping properly and the presence of a large number of unmelted powder particles, resulting in an increase in the number of defects, such as pores in the specimen. The surface of the track is relatively flat, and the depth is small. In addition, the temperature gradient before and after the molten pool was large, and the depression location appeared at the biased front end in Fig. 16(a). When P = 100 W, the surface of the track is flat and smooth with excellent lap. Due to the Marangoni effect, the velocity field of the molten pool is in the form of “vortex,” and the melt has good fluidity, and the maximum velocity reaches 2.15 m/s in Fig. 16(b). When P = 200 W, the heat generated by the Gaussian light source per unit time is too large, resulting in the melt rapidly reaching the evaporation temperature, generating a huge recoil pressure, forming a large molten pool, and the surface of the track is obviously raised. The melt movement is intense, especially the closed loop at the center end of the molten pool. At this time, the depth and width of the molten pool are large, leading to the expansion of the remolten region and the increased chance of the appearance of porosity defects in Fig. 16(c). The results show that at low laser power, the surface tension in the molten pool is dominant. At high laser power, recoil pressure is its main role.

FIG. 16.

VIEW LARGEDOWNLOAD SLIDE

Simulation results of double-track molten pool under different laser powers: (a) P = 50 W, (b) P = 100 W, (c) P = 200 W.

Table II shows the effect of different laser powers on the characteristic information of the double-track molten pool at a scanning speed of 800 mm/s and a hatch spacing of 0.06 mm. The negative overlapping ratio in the table indicates that the melt tracks are not lapped, and 26/29 indicates the melt depth of the first track/second track. It can be seen that with the increase in laser power, the melt depth, melt width, melt height, and remelted zone show a gradual increase. At the same time, the overlapping ratio also increases. Especially in the process of laser power from 50 to 200 W, the melting depth and melting width increased the most, which increased nearly 2 and 1.5 times, respectively. Meanwhile, the overlapping ratio also increases with the increase in laser power, which indicates that the melting and fusion of materials are better at high laser power. On the other hand, the dimensions of the molten pool did not change uniformly with the change of laser power. Specifically, the depth-to-width ratio of the molten pool increased from about 0.30 to 0.39 during the increase from 50 to 120 W, which further indicates that the effective heat transfer in the vertical direction is greater than that in the horizontal direction with the increase in laser power. This dimensional response to laser power is mainly affected by the recoil pressure and also by the difference in the densification degree between the powder layer and the metal substrate. In addition, according to the experimental results, the contact angle shows a tendency to increase and then decrease during the process of laser power increase, and always stays within the range of less than 33°. Therefore, in practical applications, it is necessary to select the appropriate laser power according to the specific needs in order to achieve the best processing results.

TABLE II.

Double-track molten pool characterization information at different laser powers.

Laser power (W)Depth (μm)Width (μm)Height (μm)Remolten region (μm)Overlapping ratio (%)Contact angle (°)
50 16 54 11 −10 23 
100 26/29 74 14 18 23.33 33 
200 37/45 116 21 52 93.33 28 

2. Scanning speed

Figure 17 demonstrates the effect of different scanning speeds on the morphology and temperature field of the double-track molten pool at a laser power of 100 W and a hatch spacing of 0.06 mm. With the gradual increase in scanning speed, the surface morphology of the molten pool evolves from circular to elliptical. When � = 200 mm/s, the slow scanning speed causes the material to absorb too much heat, which is very easy to trigger the overburning phenomenon. At this point, the molten pool is larger and the surface morphology is uneven. This situation is consistent with the previously discussed scenario with high laser power in Fig. 17(a). However, when � = 1600 mm/s, the scanning speed is too fast, resulting in the material not being able to absorb sufficient heat, which triggers the powder particles that fail to melt completely to have a direct effect on the bonding of the melt to the substrate. At this time, the molten pool volume is relatively small and the neighboring melt track cannot lap properly. This result is consistent with the previously discussed case of low laser power in Fig. 17(b). Overall, the ratio of the laser power to the scanning speed (the line energy density �⁠) has a direct effect on the temperature field and surface morphology of the molten pool.

FIG. 17.

VIEW LARGEDOWNLOAD SLIDE

Simulation results of double-track molten pool under different scanning speed: (a)  � = 200 mm/s, (b)  � = 1600 mm/s.

Table III shows the effects of different scanning speed on the characteristic information of the double-track molten pool under the condition of laser power of 100 W and hatch spacing of 0.06 mm. It can be seen that the scanning speed has a significant effect on the melt depth, melt width, melt height, remolten region, and overlapping ratio. With the increase in scanning speed, the melt depth, melt width, melt height, remelted zone, and overlapping ratio show a gradual decreasing trend. Among them, the melt depth and melt width decreased faster, while the melt height and remolten region decreased relatively slowly. In addition, when the scanning speed was increased from 200 to 800 mm/s, the decreasing speeds of melt depth and melt width were significantly accelerated, while the decreasing speeds of overlapping ratio were relatively slow. When the scanning speed was further increased to 1600 mm/s, the decreasing speeds of melt depth and melt width were further accelerated, and the un-lapped condition of the melt channel also appeared. In addition, the contact angle increases and then decreases with the scanning speed, and both are lower than 33°. Therefore, when selecting the scanning speed, it is necessary to make reasonable trade-offs according to the specific situation, and take into account the factors of melt depth, melt width, melt height, remolten region, and overlapping ratio, in order to achieve the best processing results.

TABLE III.

Double-track molten pool characterization information at different scanning speeds.

Scanning speed (mm/s)Depth (μm)Width (μm)Height (μm)Remolten region (μm)Overlapping ratio (%)Contact angle (°)
200 55/68 182 19/32 124 203.33 22 
1600 13 50 11 −16.67 31 

3. Hatch spacing

Figure 18 shows the effect of different hatch spacing on the morphology and temperature field of the double-track molten pool under the condition of laser power of 100 W and scanning speed of 800 mm/s. The surface morphology and temperature field of the first track and second track are basically the same, but slightly different. The first track shows a basically symmetric morphology along the scanning direction, while the second track shows a slight offset due to the difference in the heat transfer rate between the solidified material and the powder particles. When the hatch spacing is too small, the overlapping ratio increases and the probability of defects caused by remelting phenomenon grows. When the hatch spacing is too large, the neighboring melt track cannot overlap properly, and the powder particles are not completely melted, leading to an increase in the number of holes. In conclusion, the ratio of the line energy density � to the hatch spacing (the volume energy density E) has a significant effect on the temperature field and surface morphology of the molten pool.

FIG. 18.

VIEW LARGEDOWNLOAD SLIDE

Simulation results of double-track molten pool under different hatch spacings: (a) H = 0.03 mm, (b) H = 0.12 mm.

Table IV shows the effects of different hatch spacing on the characteristic information of the double-track molten pool under the condition of laser power of 100 W and scanning speed of 800 mm/s. It can be seen that the hatch spacing has little effect on the melt depth, melt width, and melt height, but has some effect on the remolten region. With the gradual expansion of hatch spacing, the remolten region shows a gradual decrease. At the same time, the overlapping ratio also decreased with the increase in hatch spacing. In addition, it is observed that the contact angle shows a tendency to increase and then remain stable when the hatch spacing increases, which has a more limited effect on it. Therefore, trade-offs and decisions need to be made on a case-by-case basis when selecting the hatch spacing.

TABLE IV.

Double-track molten pool characterization information at different hatch spacings.

Hatch spacing (mm)Depth (μm)Width (μm)Height (μm)Remolten region (μm)Overlapping ratio (%)Contact angle (°)
0.03 25/27 82 14 59 173.33 30 
0.12 26 78 14 −35 33 

In summary, the laser power, scanning speed, and hatch spacing have a significant effect on the formation of the molten pool, and the correct selection of these three process parameters is crucial to ensure the forming quality. In addition, the melt depth of the second track is slightly larger than that of the first track at higher line energy density � and volume energy density E. This is mainly due to the fact that a large amount of heat accumulation is generated after the first track, forming a larger molten pool volume, which leads to an increase in the melt depth.

D. Simulation analysis of molten pool with powder particle size and laser focal spot diameter

Figure 19 demonstrates the effect of different powder particle sizes and laser focal spot diameters on the morphology and temperature field of the double-track molten pool under a laser power of 100 W, a scanning speed of 800 mm/s, and a hatch spacing of 0.06 mm. In the process of melting coarse powder with small laser focal spot diameter, the laser energy cannot completely melt the larger powder particles, resulting in their partial melting and further generating excessive pore defects. The larger powder particles tend to generate zigzag molten pool edges, which cause an increase in the roughness of the melt track surface. In addition, the molten pool is also prone to generate the present spatter phenomenon, which can directly affect the quality of forming. The volume of the formed molten pool is relatively small, while the melt depth, melt width, and melt height are all smaller relative to the fine powder in Fig. 19(a). In the process of melting fine powders with a large laser focal spot diameter, the laser energy is able to melt the fine powder particles sufficiently, even to the point of overmelting. This results in a large number of fine spatters being generated at the edge of the molten pool, which causes porosity defects in the melt track in Fig. 19(b). In addition, the maximum velocity of the molten pool is larger for large powder particle sizes compared to small powder particle sizes, which indicates that the temperature gradient in the molten pool is larger for large powder particle sizes and the melt motion is more intense. However, the size of the laser focal spot diameter has a relatively small effect on the melt motion. However, a larger focal spot diameter induces a larger melt volume with greater depth, width, and height. In conclusion, a small powder size helps to reduce the surface roughness of the specimen, and a small laser spot diameter reduces the minimum forming size of a single track.

FIG. 19.

VIEW LARGEDOWNLOAD SLIDE

Simulation results of double-track molten pool with different powder particle size and laser focal spot diameter: (a) focal spot = 25 μm, coarse powder, (b) focal spot = 80 μm, fine powder.

Table V shows the maximum temperature gradient at the reference point for different powder sizes and laser focal spot diameters. As can be seen from the table, the maximum temperature gradient is lower than that of HP-LPBF for both coarse powders with a small laser spot diameter and fine powders with a large spot diameter, a phenomenon that leads to an increase in the heat transfer rate of HP-LPBF, which in turn leads to a corresponding increase in the cooling rate and, ultimately, to the formation of finer microstructures.

TABLE V.

Maximum temperature gradient at the reference point for different powder particle sizes and laser focal spot diameters.

Laser power (W)Scanning speed (mm/s)Hatch spacing (mm)Average powder size (μm)Laser focal spot diameter (μm)Maximum temperature gradient (×107 K/s)
100 800 0.06 31.7 25 7.89 
11.5 80 7.11 

IV. CONCLUSIONS

In this study, the geometrical characteristics of 3D coarse and fine powder particles were first calculated using DEM and then numerical simulations of single track and double track in the process of forming SS316L from monolayer HP-LPBF at mesoscopic scale were developed using CFD method. The effects of Marangoni convection, surface tension, recoil pressure, gravity, thermal convection, thermal radiation, and evaporative heat dissipation on the heat and mass transfer in the molten pool were considered in this model. The effects of laser power, scanning speed, and hatch spacing on the dynamics of the single-track and double-track molten pools, as well as on other characteristic information, were investigated. The effects of the powder particle size on the molten pool were investigated comparatively with the laser focal spot diameter. The main conclusions are as follows:

  1. The results show that the temperature gradient at the front of the molten pool is significantly larger than that at the tail, and the molten pool exhibits a “comet” morphology. At the top of the molten pool, there is a slightly concave region, which is the result of the coupling of Marangoni convection, recoil pressure, and surface tension. The melt flow forms two closed loops, which are mainly influenced by temperature gradients and surface tension. This special dynamic behavior of the melt tends to form an “elliptical” molten pool and an almost “mountain” shape in single-track forming.
  2. The basic characteristics of the three-dimensional morphology and temperature field of the second track are similar to those of the first track, but there are subtle differences. The first track exhibits a basically symmetrical shape; however, due to the difference in thermal diffusion rates between the solidified metal and the powder, a slight asymmetry in the molten pool morphology of the second track occurs. After forming through the first track, there is a significant heat buildup in the powder bed, resulting in a longer dynamic time of the melt, which increases the life of the molten pool. The heights of the first track and second track remained essentially the same, but the depth of the second track was greater relative to the first track. In addition, the maximum temperature gradient was 1.69 × 108 K/s during HP-LPBF forming.
  3. At low laser power, the surface tension in the molten pool plays a dominant role. At high laser power, recoil pressure becomes the main influencing factor. With the increase of laser power, the effective heat transfer in the vertical direction is superior to that in the horizontal direction. With the gradual increase of scanning speed, the surface morphology of the molten pool evolves from circular to elliptical. In addition, the scanning speed has a significant effect on the melt depth, melt width, melt height, remolten region, and overlapping ratio. Too large or too small hatch spacing will lead to remelting or non-lap phenomenon, which in turn causes the formation of defects.
  4. When using a small laser focal spot diameter, it is difficult to completely melt large powder particle sizes, resulting in partial melting and excessive porosity generation. At the same time, large powder particles produce curved edges of the molten pool, resulting in increased surface roughness of the melt track. In addition, spatter occurs, which directly affects the forming quality. At small focal spot diameters, the molten pool volume is relatively small, and the melt depth, the melt width, and the melt height are correspondingly small. Taken together, the small powder particle size helps to reduce surface roughness, while the small spot diameter reduces the forming size.

REFERENCES

  1. S. L. Sing and W. Y. Yeong , “ Laser powder bed fusion for metal additive manufacturing: Perspectives on recent developments,” Virtual Phys. Prototyping. 15, 359–370 (2020).https://doi.org/10.1080/17452759.2020.1779999
    Google ScholarCrossref
  2. A. M. Khorasani , I. G. Jithin , J. K. Veetil , and A. H. Ghasemi , “ A review of technological improvements in laser-based powder bed fusion of metal printers,” Int. J. Adv. Manuf. Technol. 108, 191–209 (2020).https://doi.org/10.1007/s00170-020-05361-3
    Google ScholarCrossref
  3. Y. Qin , A. Brockett , Y. Ma , A. Razali , J. Zhao , C. Harrison , W. Pan , X. Dai , and D. Loziak , “ Micro-manufacturing: Research, technology outcomes and development issues,” Int. J. Adv. Manuf. Technol. 47, 821–837 (2010).https://doi.org/10.1007/s00170-009-2411-2
    Google ScholarCrossref
  4. B. Nagarajan , Z. Hu , X. Song , W. Zhai , and J. Wei , “ Development of micro selective laser melting: The state of the art and future perspectives,” Engineering. 5, 702–720 (2019).https://doi.org/10.1016/j.eng.2019.07.002
    Google ScholarCrossref
  5. Y. Wei , G. Chen , W. Li , Y. Zhou , Z. Nie , J. Xu , and W. Zhou , “ Micro selective laser melting of SS316L: Single tracks, defects, microstructures and thermal/mechanical properties,” Opt. Laser Technol. 145, 107469 (2022).https://doi.org/10.1016/j.optlastec.2021.107469
    Google ScholarCrossref
  6. Y. Wei , G. Chen , W. Li , M. Li , Y. Zhou , Z. Nie , and J. Xu , “ Process optimization of micro selective laser melting and comparison of different laser diameter for forming different powder,” Opt. Laser Technol. 150, 107953 (2022).https://doi.org/10.1016/j.optlastec.2022.107953
    Google ScholarCrossref
  7. H. Zhiheng , B. Nagarajan , X. Song , R. Huang , W. Zhai , and J. Wei , “ Formation of SS316L single tracks in micro selective laser melting: Surface, geometry, and defects,” Adv. Mater. Sci. Eng. 2019, 9451406.https://doi.org/10.1155/2019/9451406
    Crossref
  8. B. Nagarajan , Z. Hu , S. Gao , X. Song , R. Huang , M. Seita , and J. Wei , “ Effect of in-situ laser remelting on the microstructure of SS316L fabricated by micro selective laser melting,” in Advanced Surface Enhancement, edited by Sho Itoh and Shashwat Shukla , Lecture Notes in Mechanical Engineering ( Springer Singapore, Singapore, 2020), pp. 330–336.
    Google ScholarCrossref
  9. H. Zhiheng , B. Nagarajan , X. Song , R. Huang , W. Zhai , and J. Wei , “ Tailoring surface roughness of micro selective laser melted SS316L by in-situ laser remelting,” in Advanced Surface Enhancement, edited by Sho Itoh and Shashwat Shukla , Lecture Notes in Mechanical Engineering ( Springer Singapore, Singapore, 2020), pp. 337–343.
    Google Scholar
  10. J. Fu , Z. Hu , X. Song , W. Zhai , Y. Long , H. Li , and M. Fu , “ Micro selective laser melting of NiTi shape memory alloy: Defects, microstructures and thermal/mechanical properties,” Opt. Laser Technol. 131, 106374 (2020).https://doi.org/10.1016/j.optlastec.2020.106374
    Google ScholarCrossref
  11. E. Abele and M. Kniepkamp , “ Analysis and optimisation of vertical surface roughness in micro selective laser melting,” Surf. Topogr.: Metrol. Prop. 3, 034007 (2015).https://doi.org/10.1088/2051-672X/3/3/034007
    Google ScholarCrossref
  12. S. Qu , J. Ding , J. Fu , M. Fu , B. Zhang , and X. Song , “ High-precision laser powder bed fusion processing of pure copper,” Addit. Manuf. 48, 102417 (2021).https://doi.org/10.1016/j.addma.2021.102417
    Google ScholarCrossref
  13. Y. Wei , G. Chen , M. Li , W. Li , Y. Zhou , J. Xu , and Z. wei , “ High-precision laser powder bed fusion of 18Ni300 maraging steel and its SiC reinforcement composite materials,” J. Manuf. Process. 84, 750–763 (2022).https://doi.org/10.1016/j.jmapro.2022.10.049
    Google ScholarCrossref
  14. B. Liu , R. Wildman , T. Christopher , I. Ashcroft , and H. Richard , “ Investigation the effect of particle size distribution on processing parameters optimisation in selective laser melting process,” in 2011 International Solid Freeform Fabrication Symposium ( University of Texas at Austin, 2011).
    Google Scholar
  15. T. D. McLouth , G. E. Bean , D. B. Witkin , S. D. Sitzman , P. M. Adams , D. N. Patel , W. Park , J.-M. Yang , and R. J. Zaldivar , “ The effect of laser focus shift on microstructural variation of Inconel 718 produced by selective laser melting,” Mater. Des. 149, 205–213 (2018).https://doi.org/10.1016/j.matdes.2018.04.019
    Google ScholarCrossref
  16. Y. Qian , Y. Wentao , and L. Feng , “ Mesoscopic simulations of powder bed fusion: Research progresses and conditions,” Electromachining Mould 06, 46–52 (2017).https://doi.org/10.3969/j.issn.1009-279X.2017.06.012
    Google Scholar
  17. J. Fu , S. Qu , J. Ding , X. Song , and M. W. Fu , “ Comparison of the microstructure, mechanical properties and distortion of stainless Steel 316L fabricated by micro and conventional laser powder bed fusion,” Addit. Manuf. 44, 102067 (2021).https://doi.org/10.1016/j.addma.2021.102067
    Google ScholarCrossref
  18. N. T. Aboulkhair , I. Maskery , C. Tuck , I. Ashcroft , and N. M. Everitt , “ The microstructure and mechanical properties of selectively laser Melted AlSi10Mg: The effect of a conventional T6-like heat treatment,” Mater. Sci. Eng. A 667, 139–146 (2016).https://doi.org/10.1016/j.msea.2016.04.092
    Google ScholarCrossref
  19. S. Y. Chen , J. C. Huang , C. T. Pan , C. H. Lin , T. L. Yang , Y. S. Huang , C. H. Ou , L. Y. Chen , D. Y. Lin , H. K. Lin , T. H. Li , J. S. C. Jang , and C. C. Yang , “ Microstructure and mechanical properties of open-cell porous Ti-6Al-4V fabricated by selective laser melting,” J. Alloys Compd. 713, 248–254 (2017).https://doi.org/10.1016/j.jallcom.2017.04.190
    Google ScholarCrossref
  20. Y. Bai , Y. Yang , D. Wang , and M. Zhang , “ Influence mechanism of parameters process and mechanical properties evolution mechanism of Maraging steel 300 by selective laser melting,” Mater. Sci. Eng. A 703, 116–123 (2017).https://doi.org/10.1016/j.msea.2017.06.033
    Google ScholarCrossref
  21. Y. Bai , Y. Yang , Z. Xiao , M. Zhang , and D. Wang , “ Process optimization and mechanical property evolution of AlSiMg0.75 by selective laser melting,” Mater. Des. 140, 257–266 (2018).https://doi.org/10.1016/j.matdes.2017.11.045
    Google ScholarCrossref
  22. Y. Liu , M. Zhang , W. Shi , Y. Ma , and J. Yang , “ Study on performance optimization of 316L stainless steel parts by high-efficiency selective laser melting,” Opt. Laser Technol. 138, 106872 (2021).https://doi.org/10.1016/j.optlastec.2020.106872
    Google ScholarCrossref
  23. D. Gu , Y.-C. Hagedorn , W. Meiners , G. Meng , R. J. S. Batista , K. Wissenbach , and R. Poprawe , “ Densification behavior, microstructure evolution, and wear performance of selective laser melting processed commercially pure titanium,” Acta Mater. 60, 3849–3860 (2012).https://doi.org/10.1016/j.actamat.2012.04.006
    Google ScholarCrossref
  24. N. Read , W. Wang , K. Essa , and M. M. Attallah , “ Selective laser melting of AlSi10Mg alloy: Process optimisation and mechanical properties development,” Mater. Des. 65, 417–424 (2015).https://doi.org/10.1016/j.matdes.2014.09.044
    Google ScholarCrossref
  25. I. A. Roberts , C. J. Wang , R. Esterlein , M. Stanford , and D. J. Mynors , “ A three-dimensional finite element analysis of the temperature field during laser melting of metal powders in additive layer manufacturing,” Int. J. Mach. Tools Manuf. 49(12–13), 916–923 (2009).https://doi.org/10.1016/j.ijmachtools.2009.07.004
    Google ScholarCrossref
  26. K. Dai and L. Shaw , “ Finite element analysis of the effect of volume shrinkage during laser densification,” Acta Mater. 53(18), 4743–4754 (2005).https://doi.org/10.1016/j.actamat.2005.06.014
    Google ScholarCrossref
  27. K. Carolin , E. Attar , and P. Heinl , “ Mesoscopic simulation of selective beam melting processes,” J. Mater. Process. Technol. 211(6), 978–987 (2011).https://doi.org/10.1016/j.jmatprotec.2010.12.016
    Google ScholarCrossref
  28. F.-J. Gürtler , M. Karg , K.-H. Leitz , and M. Schmidt , “ Simulation of laser beam melting of steel powders using the three-dimensional volume of fluid method,” Phys. Procedia 41, 881–886 (2013).https://doi.org/10.1016/j.phpro.2013.03.162
    Google ScholarCrossref
  29. P. Meakin and R. Jullien , “ Restructuring effects in the rain model for random deposition,” J. Phys. France 48(10), 1651–1662 (1987).https://doi.org/10.1051/jphys:0198700480100165100
    Google ScholarCrossref
  30. J-m Wang , G-h Liu , Y-l Fang , and W-k Li , “ Marangoni effect in nonequilibrium multiphase system of material processing,” Rev. Chem. Eng. 32(5), 551–585 (2016).https://doi.org/10.1515/revce-2015-0067
    Google ScholarCrossref
  31. W. Ye , S. Zhang , L. L. Mendez , M. Farias , J. Li , B. Xu , P. Li , and Y. Zhang , “ Numerical simulation of the melting and alloying processes of elemental titanium and boron powders using selective laser alloying,” J. Manuf. Process. 64, 1235–1247 (2021).https://doi.org/10.1016/j.jmapro.2021.02.044
    Google ScholarCrossref
  32. U. S. Bertoli , A. J. Wolfer , M. J. Matthews , J.-P. R. Delplanque , and J. M. Schoenung , “ On the limitations of volumetric energy density as a design parameter for selective laser melting,” Mater. Des. 113, 331–340 (2017).https://doi.org/10.1016/j.matdes.2016.10.037
    Google ScholarCrossref
  33. W. E. King , H. D. Barth , V. M. Castillo , G. F. Gallegos , J. W. Gibbs , D. E. Hahn , C. Kamath , and A. M. Rubenchik , “ Observation of keyhole-mode laser melting in laser powder-bed fusion additive manufacturing,” J. Mater. Process. Technol. 214(12), 2915–2925 (2014).https://doi.org/10.1016/j.jmatprotec.2014.06.005
    Google ScholarCrossref
  34. L. Cao , “ Numerical simulation of the impact of laying powder on selective laser melting single-pass formation,” Int. J. Heat Mass Transfer 141, 1036–1048 (2019).https://doi.org/10.1016/j.ijheatmasstransfer.2019.07.053
    Google ScholarCrossref
  35. L. Huang , X. Hua , D. Wu , and F. Li , “ Numerical study of keyhole instability and porosity formation mechanism in laser welding of aluminum alloy and steel,” J. Mater. Process. Technol. 252, 421–431 (2018).https://doi.org/10.1016/j.jmatprotec.2017.10.011
    Google ScholarCrossref
  36. K. Q. Le , C. Tang , and C. H. Wong , “ On the study of keyhole-mode melting in selective laser melting process,” Int. J. Therm. Sci. 145, 105992 (2019).https://doi.org/10.1016/j.ijthermalsci.2019.105992
    Google ScholarCrossref
  37. J.-H. Cho and S.-J. Na , “ Theoretical analysis of keyhole dynamics in polarized laser drilling,” J. Phys. D: Appl. Phys. 40(24), 7638 (2007).https://doi.org/10.1088/0022-3727/40/24/007
    Google ScholarCrossref
  38. W. Ye , “ Mechanism analysis of selective laser melting and metallurgy process based on base element powder of titanium and boron,” Ph.D. dissertation ( Nanchang University, 2021).
    Google Scholar
  39. R. Ammer , M. Markl , U. Ljungblad , C. Körner , and U. Rüde , “ Simulating fast electron beam melting with a parallel thermal free surface lattice Boltzmann method,” Comput. Math. Appl. 67(2), 318–330 (2014).https://doi.org/10.1016/j.camwa.2013.10.001
    Google ScholarCrossref
  40. H. Chen , Q. Wei , S. Wen , Z. Li , and Y. Shi , “ Flow behavior of powder particles in layering process of selective laser melting: Numerical modeling and experimental verification based on discrete element method,” Int. J. Mach. Tools Manuf. 123, 146–159 (2017).https://doi.org/10.1016/j.ijmachtools.2017.08.004
    Google ScholarCrossref
  41. F. Verhaeghe , T. Craeghs , J. Heulens , and L. Pandelaers , “ A pragmatic model for selective laser melting with evaporation,” Acta Mater. 57(20), 6006–6012 (2009).https://doi.org/10.1016/j.actamat.2009.08.027
    Google ScholarCrossref
  42. C. H. Fu and Y. B. Guo , “ Three-dimensional temperature gradient mechanism in selective laser melting of Ti-6Al-4V,” J. Manuf. Sci. Eng. 136(6), 061004 (2014).https://doi.org/10.1115/1.4028539
    Google ScholarCrossref
  43. Y. Xiang , Z. Shuzhe , L. Junfeng , W. Zhengying , Y. Lixiang , and J. Lihao , “ Numerical simulation and experimental verification for selective laser single track melting forming of Ti6Al4V,” J. Zhejiang Univ. (Eng. Sci.) 53(11), 2102–2109 + 2117 (2019).https://doi.org/10.3785/j.issn.1008-973X.2019.11.007
    Google Scholar
  44. Q. He , H. Xia , J. Liu , X. Ao , and S. Lin , “ Modeling and numerical studies of selective laser melting: Multiphase flow, solidification and heat transfer,” Mater. Des. 196, 109115 (2020).https://doi.org/10.1016/j.matdes.2020.109115
    Google ScholarCrossref
  45. L. Cao , “ Mesoscopic-scale numerical simulation including the influence of process parameters on SLM single-layer multi-pass formation,” Metall. Mater. Trans. A 51, 4130–4145 (2020).https://doi.org/10.1007/s11661-020-05831-z
    Google ScholarCrossref
  46. L. Cao , “ Mesoscopic-scale numerical investigation including the influence of process parameters on LPBF multi-layer multi-path formation,” Comput. Model. Eng. Sci. 126(1), 5–23 (2021).https://doi.org/10.32604/cmes.2021.014693
    Google ScholarCrossref
  47. H. Yin and S. D. Felicelli , “ Dendrite growth simulation during solidification in the LENS process,” Acta Mater. 58(4), 1455–1465 (2010).https://doi.org/10.1016/j.actamat.2009.10.053
    Google ScholarCrossref
  48. P. Nie , O. A. Ojo , and Z. Li , “ Numerical modeling of microstructure evolution during laser additive manufacturing of a nickel-based superalloy,” Acta Mater. 77, 85–95 (2014).https://doi.org/10.1016/j.actamat.2014.05.039
    Google ScholarCrossref
  49. Z. Liu and H. Qi , “ Effects of substrate crystallographic orientations on crystal growth and microstructure formation in laser powder deposition of nickel-based superalloy,” Acta Mater. 87, 248–258 (2015).https://doi.org/10.1016/j.actamat.2014.12.046
    Google ScholarCrossref
  50. L. Wei , L. Xin , W. Meng , and H. Weidong , “ Cellular automaton simulation of the molten pool of laser solid forming process,” Acta Phys. Sin. 64(01), 018103–018363 (2015).https://doi.org/10.7498/aps.64.018103
    Google ScholarCrossref
  51. R. Acharya , J. A. Sharon , and A. Staroselsky , “ Prediction of microstructure in laser powder bed fusion process,” Acta Mater. 124, 360–371 (2017).https://doi.org/10.1016/j.actamat.2016.11.018
    Google ScholarCrossref
  52. M. R. Rolchigo and R. LeSar , “ Modeling of binary alloy solidification under conditions representative of additive manufacturing,” Comput. Mater. Sci. 150, 535–545 (2018).https://doi.org/10.1016/j.commatsci.2018.04.004
    Google ScholarCrossref
  53. S. Geng , P. Jiang , L. Guo , X. Gao , and G. Mi , “ Multi-scale simulation of grain/sub-grain structure evolution during solidification in laser welding of aluminum alloys,” Int. J. Heat Mass Transfer 149, 119252 (2020).https://doi.org/10.1016/j.ijheatmasstransfer.2019.119252
    Google ScholarCrossref
  54. W. L. Wang , W. Q. Liu , X. Yang , R. R. Xu , and Q. Y. Dai , “ Multi-scale simulation of columnar-to-equiaxed transition during laser selective melting of rare earth magnesium alloy,” J. Mater. Sci. Technol. 119, 11–24 (2022).https://doi.org/10.1016/j.jmst.2021.12.029
    Google ScholarCrossref
  55. Q. Xia , J. Yang , and Y. Li , “ On the conservative phase-field method with the N-component incompressible flows,” Phys. Fluids 35, 012120 (2023).https://doi.org/10.1063/5.0135490
    Google ScholarCrossref
  56. Q. Xia , G. Sun , J. Kim , and Y. Li , “ Multi-scale modeling and simulation of additive manufacturing based on fused deposition technique,” Phys. Fluids 35, 034116 (2023).https://doi.org/10.1063/5.0141316
    Google ScholarCrossref
  57. A. Hussein , L. Hao , C. Yan , and R. Everson , “ Finite element simulation of the temperature and stress fields in single layers built without-support in selective laser melting,” Mater. Des. 52, 638–647 (2013).https://doi.org/10.1016/j.matdes.2013.05.070
    Google ScholarCrossref
  58. J. Ding , P. Colegrove , J. Mehnen , S. Ganguly , P. M. Sequeira Almeida , F. Wang , and S. Williams , “ Thermo-mechanical analysis of wire and arc additive layer manufacturing process on large multi-layer parts,” Comput. Mater. Sci. 50(12), 3315–3322 (2011).https://doi.org/10.1016/j.commatsci.2011.06.023
    Google ScholarCrossref
  59. Y. Du , X. You , F. Qiao , L. Guo , and Z. Liu , “ A model for predicting the temperature field during selective laser melting,” Results Phys. 12, 52–60 (2019).https://doi.org/10.1016/j.rinp.2018.11.031
    Google ScholarCrossref
  60. X. Luo , M. Liu , L. Zhenhua , H. Li , and J. Shen , “ Effect of different heat-source models on calculated temperature field of selective laser melted 18Ni300,” Chin. J. Lasers 48(14), 1402005–1402062 (2021).https://doi.org/10.3788/CJL202148.1402005
    Google ScholarCrossref
  61. J. F. Li , L. Li , and F. H. Stott , “ Thermal stresses and their implication on cracking during laser melting of ceramic materials,” Acta Mater. 52(14), 4385–4398 (2004).https://doi.org/10.1016/j.actamat.2004.06.005
    Google ScholarCrossref
  62. P. Aggarangsi and J. L. Beuth , “ Localized preheating approaches for reducing residual stress in additive manufacturing,” paper presented at the 2006 International Solid Freeform Fabrication Symposium, The University of Texas in Austin on August 14–16, 2006.
  63. K. Dai and L. Shaw , “ Thermal and mechanical finite element modeling of laser forming from metal and ceramic powders,” Acta Mater. 52(1), 69–80 (2004).https://doi.org/10.1016/j.actamat.2003.08.028
    Google ScholarCrossref
  64. A. H. Nickel , D. M. Barnett , and F. B. Prinz , “ Thermal stresses and deposition patterns in layered manufacturing,” Mater. Sci. Eng. A 317(1–2), 59–64 (2001).https://doi.org/10.1016/S0921-5093(01)01179-0
    Google ScholarCrossref
  65. M. F. Zaeh and G. Branner , “ Investigations on residual stresses and deformations in selective laser melting,” Prod. Eng. 4(1), 35–45 (2010).https://doi.org/10.1007/s11740-009-0192-y
    Google ScholarCrossref
  66. P. Bian , J. Shi , Y. Liu , and Y. Xie , “ Influence of laser power and scanning strategy on residual stress distribution in additively manufactured 316L steel,” Opt. Laser Technol. 132, 106477 (2020).https://doi.org/10.1016/j.optlastec.2020.106477
    Google ScholarCrossref
  67. B. M. Marques , C. M. Andrade , D. M. Neto , M. C. Oliveira , J. L. Alves , and L. F. Menezes , “ Numerical analysis of residual stresses in parts produced by selective laser melting process,” Procedia Manuf. 47, 1170–1177 (2020).https://doi.org/10.1016/j.promfg.2020.04.167
    Google ScholarCrossref
  68. W. Mu , “ Numerical simulation of SLM forming process and research and prediction of forming properties,” MA thesis ( Anhui Jianzhu University, 2022).
    Google Scholar
  69. Y. Zhang , “ Multi-scale multi-physics modeling of laser powder bed fusion process of metallic materials with experiment validation,” Ph.D. dissertation ( Purdue University, 2018).
    Google Scholar
  70. Y. Qian , “ Mesoscopic simulation studies of key processing issues for powder bed fusion technology,” Ph.D. dissertation ( Tsinghua University, 2019).
    Google Scholar
  71. N. V. Brilliantov , S. Frank , J.-M. Hertzsch , and T. Pöschel , “ Model for collisions in granular gases,” Phys. Rev. E 53(5), 5382–5392 (1996).https://doi.org/10.1103/PhysRevE.53.5382
    Google ScholarCrossref
  72. Z. Xiao , “ Research on microscale selective laser melting process of high strength pure copper specimens,” MA thesis ( Hunan University, 2022).
    Google Scholar
  73. Z. Li , K. Mukai , M. Zeze , and K. C. Mills , “ Determination of the surface tension of liquid stainless steel,” J. Mater. Sci. 40(9–10), 2191–2195 (2005).https://doi.org/10.1007/s10853-005-1931-x
    Google ScholarCrossref
  74. R. Scardovelli and S. Zaleski , “ Analytical relations connecting linear interfaces and volume fractions in rectangular grids,” J. Comput. Phys. 164(1), 228–237 (2000).https://doi.org/10.1006/jcph.2000.6567
    Google ScholarCrossref
  75. D.-W. Cho , W.-I. Cho , and S.-J. Na , “ Modeling and simulation of arc: Laser and hybrid welding process,” J. Manuf. Process. 16(1), 26–55 (2014).https://doi.org/10.1016/j.jmapro.2013.06.012
    Google ScholarCrossref
    76.Flow3D. Version 11.1.0: User Manual ( FlowScience, Santa Fe, NM, USA, 2015).
  76. Y. Tian , L. Yang , D. Zhao , Y. Huang , and J. Pan , “ Numerical analysis of powder bed generation and single track forming for selective laser melting of ss316l stainless steel,” J. Manuf. Process. 58, 964–974 (2020).https://doi.org/10.1016/j.jmapro.2020.09.002
    Google ScholarCrossref
  77. C. Tang , K. Q. Le , and C. H. Wong , “ Physics of humping formation in laser powder bed fusion,” Int. J. Heat Mass Transfer 149, 119172 (2020).https://doi.org/10.1016/j.ijheatmasstransfer.2019.119172
    Google ScholarCrossref
  78. L. Cao , “ Mesoscopic-scale simulation of pore evolution during laser powder bed fusion process,” Comput. Mater. Sci. 179, 109686 (2020).https://doi.org/10.1016/j.commatsci.2020.109686
    Google ScholarCrossref
  79. R. Li , J. Liu , Y. Shi , W. Li , and W. Jiang , “ Balling behavior of stainless steel and nickel powder during selective laser melting process,” Int. J. Adv. Manuf. Technol. 59(9–12), 1025–1035 (2012).https://doi.org/10.1007/s00170-011-3566-1
    Google ScholarCrossref
  80. S. A. Khairallah and A. Anderson , “ Mesoscopic simulation model of selective laser melting of stainless steel powder,” J. Mater. Process. Technol. 214(11), 2627–2636 (2014).https://doi.org/10.1016/j.jmatprotec.2014.06.001
    Google ScholarCrossref
  81. J. Liu , D. Gu , H. Chen , D. Dai , and H. Zhang , “ Influence of substrate surface morphology on wetting behavior of tracks during selective laser melting of aluminum-based alloys,” J. Zhejiang Univ. Sci. A 19(2), 111–121 (2018).https://doi.org/10.1631/jzus.A1700599
    Google ScholarCrossref
  82. L. Li , J. Li , and T. Fan , “ Phase-field modeling of wetting and balling dynamics in powder bed fusion process,” Phys. Fluids 33, 042116 (2021).https://doi.org/10.1063/5.0046771
    Google ScholarCrossref
  83. X. Nie , Z. Hu , H. Zhu , Z. Hu , L. Ke , and X. Zeng , “ Analysis of processing parameters and characteristics of selective laser melted high strength Al-Cu-Mg alloys: from single tracks to cubic samples,” J. Mater. Process. Technol. 256, 69–77 (2018).https://doi.org/10.1016/j.jmatprotec.2018.01.030
    Google ScholarCrossref
Predicting solid-state phase transformations during metal additive manufacturing: A case study on electron-beam powder bed fusion of Inconel-738

Predicting solid-state phase transformations during metal additive manufacturing: A case study on electron-beam powder bed fusion of Inconel-738

금속 적층 제조 중 고체 상 변형 예측: Inconel-738의 전자빔 분말층 융합에 대한 사례 연구

Nana Kwabena Adomako a, Nima Haghdadi a, James F.L. Dingle bc, Ernst Kozeschnik d, Xiaozhou Liao bc, Simon P. Ringer bc, Sophie Primig a

Abstract

Metal additive manufacturing (AM) has now become the perhaps most desirable technique for producing complex shaped engineering parts. However, to truly take advantage of its capabilities, advanced control of AM microstructures and properties is required, and this is often enabled via modeling. The current work presents a computational modeling approach to studying the solid-state phase transformation kinetics and the microstructural evolution during AM. Our approach combines thermal and thermo-kinetic modelling. A semi-analytical heat transfer model is employed to simulate the thermal history throughout AM builds. Thermal profiles of individual layers are then used as input for the MatCalc thermo-kinetic software. The microstructural evolution (e.g., fractions, morphology, and composition of individual phases) for any region of interest throughout the build is predicted by MatCalc. The simulation is applied to an IN738 part produced by electron beam powder bed fusion to provide insights into how γ′ precipitates evolve during thermal cycling. Our simulations show qualitative agreement with our experimental results in predicting the size distribution of γ′ along the build height, its multimodal size character, as well as the volume fraction of MC carbides. Our findings indicate that our method is suitable for a range of AM processes and alloys, to predict and engineer their microstructures and properties.

Graphical Abstract

ga1

Keywords

Additive manufacturing, Simulation, Thermal cycles, γ′ phase, IN738

1. Introduction

Additive manufacturing (AM) is an advanced manufacturing method that enables engineering parts with intricate shapes to be fabricated with high efficiency and minimal materials waste. AM involves building up 3D components layer-by-layer from feedstocks such as powder [1]. Various alloys, including steel, Ti, Al, and Ni-based superalloys, have been produced using different AM techniques. These techniques include directed energy deposition (DED), electron- and laser powder bed fusion (E-PBF and L-PBF), and have found applications in a variety of industries such as aerospace and power generation [2][3][4]. Despite the growing interest, certain challenges limit broader applications of AM fabricated components in these industries and others. One of such limitations is obtaining a suitable and reproducible microstructure that offers the desired mechanical properties consistently. In fact, the AM as-built microstructure is highly complex and considerably distinctive from its conventionally processed counterparts owing to the complicated thermal cycles arising from the deposition of several layers upon each other [5][6].

Several studies have reported that the solid-state phases and solidification microstructure of AM processed alloys such as CMSX-4, CoCr [7][8], Ti-6Al-4V [9][10][11]IN738 [6]304L stainless steel [12], and IN718 [13][14] exhibit considerable variations along the build direction. For instance, references [9][10] have reported that there is a variation in the distribution of α and β phases along the build direction in Ti-alloys. Similarly, the microstructure of an L-PBF fabricated martensitic steel exhibits variations in the fraction of martensite [15]. Furthermore, some of the present authors and others [6][16][17][18][19][20] have recently reviewed and reported that there is a difference in the morphology and fraction of nanoscale precipitates as a function of build height in Ni-based superalloys. These non-uniformities in the as-built microstructure result in an undesired heterogeneity in mechanical and other important properties such as corrosion and oxidation [19][21][22][23]. To obtain the desired microstructure and properties, additional processing treatments are utilized, but this incurs extra costs and may lead to precipitation of detrimental phases and grain coarsening. Therefore, a through-process understanding of the microstructure evolution under repeated heating and cooling is now needed to further advance 3D printed microstructure and property control.

It is now commonly understood that the microstructure evolution during printing is complex, and most AM studies concentrate on the microstructure and mechanical properties of the final build only. Post-printing studies of microstructure characteristics at room temperature miss crucial information on how they evolve. In-situ measurements and modelling approaches are required to better understand the complex microstructural evolution under repeated heating and cooling. Most in-situ measurements in AM focus on monitoring the microstructural changes, such as phase transformations and melt pool dynamics during fabrication using X-ray scattering and high-speed X-ray imaging [24][25][26][27]. For example, Zhao et al. [25] measured the rate of solidification and described the α/β phase transformation during L-PBF of Ti-6Al-4V in-situ. Also, Wahlmann et al. [21] recently used an L-PBF machine coupled with X-ray scattering to investigate the changes in CMSX-4 phase during successive melting processes. Although these techniques provide significant understanding of the basic principles of AM, they are not widely accessible. This is due to the great cost of the instrument, competitive application process, and complexities in terms of the experimental set-up, data collection, and analysis [26][28].

Computational modeling techniques are promising and more widely accessible tools that enable advanced understanding, prediction, and engineering of microstructures and properties during AM. So far, the majority of computational studies have concentrated on physics based process models for metal AM, with the goal of predicting the temperature profile, heat transfer, powder dynamics, and defect formation (e.g., porosity) [29][30]. In recent times, there have been efforts in modeling of the AM microstructure evolution using approaches such as phase-field [31], Monte Carlo (MC) [32], and cellular automata (CA) [33], coupled with finite element simulations for temperature profiles. However, these techniques are often restricted to simulating the evolution of solidification microstructures (e.g., grain and dendrite structure) and defects (e.g., porosity). For example, Zinovieva et al. [33] predicted the grain structure of L-PBF Ti-6Al-4V using finite difference and cellular automata methods. However, studies on the computational modelling of the solid-state phase transformations, which largely determine the resulting properties, remain limited. This can be attributed to the multi-component and multi-phase nature of most engineering alloys in AM, along with the complex transformation kinetics during thermal cycling. This kind of research involves predictions of the thermal cycle in AM builds, and connecting it to essential thermodynamic and kinetic data as inputs for the model. Based on the information provided, the thermokinetic model predicts the history of solid-state phase microstructure evolution during deposition as output. For example, a multi-phase, multi-component mean-field model has been developed to simulate the intermetallic precipitation kinetics in IN718 [34] and IN625 [35] during AM. Also, Basoalto et al. [36] employed a computational framework to examine the contrasting distributions of process-induced microvoids and precipitates in two Ni-based superalloys, namely IN718 and CM247LC. Furthermore, McNamara et al. [37] established a computational model based on the Johnson-Mehl-Avrami model for non-isothermal conditions to predict solid-state phase transformation kinetics in L-PBF IN718 and DED Ti-6Al-4V. These models successfully predicted the size and volume fraction of individual phases and captured the repeated nucleation and dissolution of precipitates that occur during AM.

In the current study, we propose a modeling approach with appreciably short computational time to investigate the detailed microstructural evolution during metal AM. This may include obtaining more detailed information on the morphologies of phases, such as size distribution, phase fraction, dissolution and nucleation kinetics, as well as chemistry during thermal cycling and final cooling to room temperature. We utilize the combination of the MatCalc thermo-kinetic simulator and a semi-analytical heat conduction model. MatCalc is a software suite for simulation of phase transformations, microstructure evolution and certain mechanical properties in engineering alloys. It has successfully been employed to simulate solid-state phase transformations in Ni-based superalloys [38][39], steels [40], and Al alloys [41] during complex thermo-mechanical processes. MatCalc uses the classical nucleation theory as well as the so-called Svoboda-Fischer-Fratzl-Kozeschnik (SFFK) growth model as the basis for simulating precipitation kinetics [42]. Although MatCalc was originally developed for conventional thermo-mechanical processes, we will show that it is also applicable for AM if the detailed time-temperature profile of the AM build is known. The semi-analytical heat transfer code developed by Stump and Plotkowski [43] is used to simulate these profile throughout the AM build.

1.1. Application to IN738

Inconel-738 (IN738) is a precipitation hardening Ni-based superalloy mainly employed in high-temperature components, e.g. in gas turbines and aero-engines owing to its exceptional mechanical properties at temperatures up to 980 °C, coupled with high resistance to oxidation and corrosion [44]. Its superior high-temperature strength (∼1090 MPa tensile strength) is provided by the L12 ordered Ni3(Al,Ti) γ′ phase that precipitates in a face-centered cubic (FCC) γ matrix [45][46]. Despite offering great properties, IN738, like most superalloys with high γ′ fractions, is challenging to process owing to its propensity to hot cracking [47][48]. Further, machining of such alloys is challenging because of their high strength and work-hardening rates. It is therefore difficult to fabricate complex INC738 parts using traditional manufacturing techniques like casting, welding, and forging.

The emergence of AM has now made it possible to fabricate such parts from IN738 and other superalloys. Some of the current authors’ recent research successfully applied E-PBF to fabricate defect-free IN738 containing γ′ throughout the build [16][17]. The precipitated γ′ were heterogeneously distributed. In particular, Haghdadi et al. [16] studied the origin of the multimodal size distribution of γ′, while Lim et al. [17] investigated the gradient in γ′ character with build height and its correlation to mechanical properties. Based on these results, the present study aims to extend the understanding of the complex and site-specific microstructural evolution in E-PBF IN738 by using a computational modelling approach. New experimental evidence (e.g., micrographs not published previously) is presented here to support the computational results.

2. Materials and Methods

2.1. Materials preparation

IN738 Ni-based superalloy (59.61Ni-8.48Co-7.00Al-17.47Cr-3.96Ti-1.01Mo-0.81W-0.56Ta-0.49Nb-0.47C-0.09Zr-0.05B, at%) gas-atomized powder was used as feedstock. The powders, with average size of 60 ± 7 µm, were manufactured by Praxair and distributed by Astro Alloys Inc. An Arcam Q10 machine by GE Additive with an acceleration voltage of 60 kV was used to fabricate a 15 × 15 × 25 mm3 block (XYZ, Z: build direction) on a 316 stainless steel substrate. The block was 3D-printed using a ‘random’ spot melt pattern. The random spot melt pattern involves randomly selecting points in any given layer, with an equal chance of each point being melted. Each spot melt experienced a dwell time of 0.3 ms, and the layer thickness was 50 µm. Some of the current authors have previously characterized the microstructure of the very same and similar builds in more detail [16][17]. A preheat temperature of ∼1000 °C was set and kept during printing to reduce temperature gradients and, in turn, thermal stresses [49][50][51]. Following printing, the build was separated from the substrate through electrical discharge machining. It should be noted that this sample was simultaneously printed with the one used in [17] during the same build process and on the same build plate, under identical conditions.

2.2. Microstructural characterization

The printed sample was longitudinally cut in the direction of the build using a Struers Accutom-50, ground, and then polished to 0.25 µm suspension via standard techniques. The polished x-z surface was electropolished and etched using Struers A2 solution (perchloric acid in ethanol). Specimens for image analysis were polished using a 0.06 µm colloidal silica. Microstructure analyses were carried out across the height of the build using optical microscopy (OM) and scanning electron microscopy (SEM) with focus on the microstructure evolution (γ′ precipitates) in individual layers. The position of each layer being analyzed was determined by multiplying the layer number by the layer thickness (50 µm). It should be noted that the position of the first layer starts where the thermal profile is tracked (in this case, 2 mm from the bottom). SEM images were acquired using a JEOL 7001 field emission microscope. The brightness and contrast settings, acceleration voltage of 15 kV, working distance of 10 mm, and other SEM imaging parameters were all held constant for analysis of the entire build. The ImageJ software was used for automated image analysis to determine the phase fraction and size of γ′ precipitates and carbides. A 2-pixel radius Gaussian blur, following a greyscale thresholding and watershed segmentation was used [52]. Primary γ′ sizes (>50 nm), were measured using equivalent spherical diameters. The phase fractions were considered equal to the measured area fraction. Secondary γ′ particles (<50 nm) were not considered here. The γ′ size in the following refers to the diameter of a precipitate.

2.3. Hardness testing

A Struers DuraScan tester was utilized for Vickers hardness mapping on a polished x-z surface, from top to bottom under a maximum load of 100 mN and 10 s dwell time. 30 micro-indentations were performed per row. According to the ASTM standard [53], the indentations were sufficiently distant (∼500 µm) to assure that strain-hardened areas did not interfere with one another.

2.4. Computational simulation of E-PBF IN738 build

2.4.1. Thermal profile modeling

The thermal history was generated using the semi-analytical heat transfer code (also known as the 3DThesis code) developed by Stump and Plotkowski [43]. This code is an open-source C++ program which provides a way to quickly simulate the conductive heat transfer found in welding and AM. The key use case for the code is the simulation of larger domains than is practicable with Computational Fluid Dynamics/Finite Element Analysis programs like FLOW-3D AM. Although simulating conductive heat transfer will not be an appropriate simplification for some investigations (for example the modelling of keyholding or pore formation), the 3DThesis code does provide fast estimates of temperature, thermal gradient, and solidification rate which can be useful for elucidating microstructure formation across entire layers of an AM build. The mathematics involved in the code is as follows:

In transient thermal conduction during welding and AM, with uniform and constant thermophysical properties and without considering fluid convection and latent heat effects, energy conservation can be expressed as:(1)��∂�∂�=�∇2�+�̇where � is density, � specific heat, � temperature, � time, � thermal conductivity, and �̇ a volumetric heat source. By assuming a semi-infinite domain, Eq. 1 can be analytically solved. The solution for temperature at a given time (t) using a volumetric Gaussian heat source is presented as:(2)��,�,�,�−�0=33�����32∫0�1������exp−3�′�′2��+�′�′2��+�′�′2����′(3)and��=12��−�′+��2for�=�,�,�(4)and�′�′=�−���′Where � is the vector �,�,� and �� is the location of the heat source.

The numerical integration scheme used is an adaptive Gaussian quadrature method based on the following nondimensionalization:(5)�=��xy2�,�′=��xy2�′,�=��xy,�=��xy,�=��xy,�=���xy

A more detailed explanation of the mathematics can be found in reference [43].

The main source of the thermal cycling present within a powder-bed fusion process is the fusion of subsequent layers. Therefore, regions near the top of a build are expected to undergo fewer thermal cycles than those closer to the bottom. For this purpose, data from the single scan’s thermal influence on multiple layers was spliced to represent the thermal cycles experienced at a single location caused by multiple subsequent layers being fused.

The cross-sectional area simulated by this model was kept constant at 1 × 1 mm2, and the depth was dependent on the build location modelled with MatCalc. For a build location 2 mm from the bottom, the maximum number of layers to simulate is 460. Fig. 1a shows a stitched overview OM image of the entire build indicating the region where this thermal cycle is simulated and tracked. To increase similarity with the conditions of the physical build, each thermal history was constructed from the results of two simulations generated with different versions of a random scan path. The parameters used for these thermal simulations can be found in Table 1. It should be noted that the main purpose of the thermal profile modelling was to demonstrate how the conditions at different locations of the build change relative to each other. Accurately predicting the absolute temperature during the build would require validation via a temperature sensor measurement during the build process which is beyond the scope of the study. Nonetheless, to establish the viability of the heat source as a suitable approximation for this study, an additional sensitivity analysis was conducted. This analysis focused on the influence of energy input on γ′ precipitation behavior, the central aim of this paper. This was achieved by employing varying beam absorption energies (0.76, 0.82 – the values utilized in the simulation, and 0.9). The direct impact of beam absorption efficiency on energy input into the material was investigated. Specifically, the initial 20 layers of the build were simulated and subsequently compared to experimental data derived from SEM. While phase fractions were found to be consistent across all conditions, disparities emerged in the mean size of γ′ precipitates. An absorption efficiency of 0.76 yielded a mean size of approximately 70 nm. Conversely, absorption efficiencies of 0.82 and 0.9 exhibited remarkably similar mean sizes of around 130 nm, aligning closely with the outcomes of the experiments.

Fig. 1

Table 1. A list of parameters used in thermal simulation of E-PBF.

ParameterValue
Spatial resolution5 µm
Time step0.5 s
Beam diameter200 µm
Beam penetration depth1 µm
Beam power1200 W
Beam absorption efficiency0.82
Thermal conductivity25.37 W/(m⋅K)
Chamber temperature1000 °C
Specific heat711.756 J/(kg⋅K)
Density8110 kg/m3

2.4.2. Thermo-kinetic simulation

The numerical analyses of the evolution of precipitates was performed using MatCalc version 6.04 (rel 0.011). The thermodynamic (‘mc_ni.tdb’, version 2.034) and diffusion (‘mc_ni.ddb’, version 2.007) databases were used. MatCalc’s basic principles are elaborated as follows:

The nucleation kinetics of precipitates are computed using a computational technique based on a classical nucleation theory [54] that has been modified for systems with multiple components [42][55]. Accordingly, the transient nucleation rate (�), which expresses the rate at which nuclei are formed per unit volume and time, is calculated as:(6)�=�0��*∙�xp−�*�∙�∙exp−��where �0 denotes the number of active nucleation sites, �* the rate of atomic attachment, � the Boltzmann constant, � the temperature, �* the critical energy for nucleus formation, τ the incubation time, and t the time. � (Zeldovich factor) takes into consideration that thermal excitation destabilizes the nucleus as opposed to its inactive state [54]. Z is defined as follows:(7)�=−12�kT∂2∆�∂�2�*12where ∆� is the overall change in free energy due to the formation of a nucleus and n is the nucleus’ number of atoms. ∆�’s derivative is evaluated at n* (critical nucleus size). �* accounts for the long-range diffusion of atoms required for nucleation, provided that the matrix’ and precipitates’ composition differ. Svoboda et al. [42] developed an appropriate multi-component equation for �*, which is given by:(8)�*=4��*2�4�∑�=1��ki−�0�2�0��0�−1where �* denotes the critical radius for nucleation, � represents atomic distance, and � is the molar volume. �ki and �0� represent the concentration of elements in the precipitate and matrix, respectively. The parameter �0� denotes the rate of diffusion of the ith element within the matrix. The expression for the incubation time � is expressed as [54]:(9)�=12�*�2

and �*, which represents the critical energy for nucleation:(10)�*=16�3�3∆�vol2where � is the interfacial energy, and ∆Gvol the change in the volume free energy. The critical nucleus’ composition is similar to the γ′ phase’s equilibrium composition at the same temperature. � is computed based on the precipitate and matrix compositions, using a generalized nearest neighbor broken bond model, with the assumption of interfaces being planar, sharp, and coherent [56][57][58].

In Eq. 7, it is worth noting that �* represents the fundamental variable in the nucleation theory. It contains �3/∆�vol2 and is in the exponent of the nucleation rate. Therefore, even small variations in γ and/or ∆�vol can result in notable changes in �, especially if �* is in the order of �∙�. This is demonstrated in [38] for UDIMET 720 Li during continuous cooling, where these quantities change steadily during precipitation due to their dependence on matrix’ and precipitate’s temperature and composition. In the current work, these changes will be even more significant as the system is exposed to multiple cycles of rapid cooling and heating.

Once nucleated, the growth of a precipitate is assessed using the radius and composition evolution equations developed by Svoboda et al. [42] with a mean-field method that employs the thermodynamic extremal principle. The expression for the total Gibbs free energy of a thermodynamic system G, which consists of n components and m precipitates, is given as follows:(11)�=∑���0��0�+∑�=1�4���33��+∑�=1��ki�ki+∑�=1�4���2��.

The chemical potential of component � in the matrix is denoted as �0�(�=1,…,�), while the chemical potential of component � in the precipitate is represented by �ki(�=1,…,�,�=1,…,�). These chemical potentials are defined as functions of the concentrations �ki(�=1,…,�,�=1,…,�). The interface energy density is denoted as �, and �� incorporates the effects of elastic energy and plastic work resulting from the volume change of each precipitate.

Eq. (12) establishes that the total free energy of the system in its current state relies on the independent state variables: the sizes (radii) of the precipitates �� and the concentrations of each component �ki. The remaining variables can be determined by applying the law of mass conservation to each component �. This can be represented by the equation:(12)��=�0�+∑�=1�4���33�ki,

Furthermore, the global mass conservation can be expressed by equation:(13)�=∑�=1���When a thermodynamic system transitions to a more stable state, the energy difference between the initial and final stages is dissipated. This model considers three distinct forms of dissipation effects [42]. These include dissipations caused by the movement of interfaces, diffusion within the precipitate and diffusion within the matrix.

Consequently, �̇� (growth rate) and �̇ki (chemical composition’s rate of change) of the precipitate with index � are derived from the linear system of equation system:(14)�ij��=��where �� symbolizes the rates �̇� and �̇ki [42]. Index i contains variables for precipitate radius, chemical composition, and stoichiometric boundary conditions suggested by the precipitate’s crystal structure. Eq. (10) is computed separately for every precipitate �. For a more detailed description of the formulae for the coefficients �ij and �� employed in this work please refer to [59].

The MatCalc software was used to perform the numerical time integration of �̇� and �̇ki of precipitates based on the classical numerical method by Kampmann and Wagner [60]. Detailed information on this method can be found in [61]. Using this computational method, calculations for E-PBF thermal cycles (cyclic heating and cooling) were computed and compared to experimental data. The simulation took approximately 2–4 hrs to complete on a standard laptop.

3. Results

3.1. Microstructure

Fig. 1 displays a stitched overview image and selected SEM micrographs of various γ′ morphologies and carbides after observations of the X-Z surface of the build from the top to 2 mm above the bottom. Fig. 2 depicts a graph that charts the average size and phase fraction of the primary γ′, as it changes with distance from the top to the bottom of the build. The SEM micrographs show widespread primary γ′ precipitation throughout the entire build, with the size increasing in the top to bottom direction. Particularly, at the topmost height, representing the 460th layer (Z = 22.95 mm), as seen in Fig. 1b, the average size of γ′ is 110 ± 4 nm, exhibiting spherical shapes. This is representative of the microstructure after it solidifies and cools to room temperature, without experiencing additional thermal cycles. The γ′ size slightly increases to 147 ± 6 nm below this layer and remains constant until 0.4 mm (∼453rd layer) from the top. At this position, the microstructure still closely resembles that of the 460th layer. After the 453rd layer, the γ′ size grows rapidly to ∼503 ± 19 nm until reaching the 437th layer (1.2 mm from top). The γ′ particles here have a cuboidal shape, and a small fraction is coarser than 600 nm. γ′ continue to grow steadily from this position to the bottom (23 mm from the top). A small fraction of γ′ is > 800 nm.

Fig. 2

Besides primary γ′, secondary γ′ with sizes ranging from 5 to 50 nm were also found. These secondary γ′ precipitates, as seen in Fig. 1f, were present only in the bottom and middle regions. A detailed analysis of the multimodal size distribution of γ′ can be found in [16]. There is no significant variation in the phase fraction of the γ′ along the build. The phase fraction is ∼ 52%, as displayed in Fig. 2. It is worth mentioning that the total phase fraction of γ′ was estimated based on the primary γ′ phase fraction because of the small size of secondary γ′. Spherical MC carbides with sizes ranging from 50 to 400 nm and a phase fraction of 0.8% were also observed throughout the build. The carbides are the light grey precipitates in Fig. 1g. The light grey shade of carbides in the SEM images is due to their composition and crystal structure [52]. These carbides are not visible in Fig. 1b-e because they were dissolved during electro-etching carried out after electropolishing. In Fig. 1g, however, the sample was examined directly after electropolishing, without electro-etching.

Table 2 shows the nominal and measured composition of γ′ precipitates throughout the build by atom probe microscopy as determined in our previous study [17]. No build height-dependent composition difference was observed in either of the γ′ precipitate populations. However, there was a slight disparity between the composition of primary and secondary γ′. Among the main γ′ forming elements, the primary γ′ has a high Ti concentration while secondary γ′ has a high Al concentration. A detailed description of the atom distribution maps and the proxigrams of the constituent elements of γ′ throughout the build can be found in [17].

Table 2. Bulk IN738 composition determined using inductively coupled plasma atomic emission spectroscopy (ICP-AES). Compositions of γ, primary γ′, and secondary γ′ at various locations in the build measured by APT. This information is reproduced from data in Ref. [17] with permission.

at%NiCrCoAlMoWTiNbCBZrTaOthers
Bulk59.1217.478.487.001.010.813.960.490.470.050.090.560.46
γ matrix
Top50.4832.9111.591.941.390.820.440.80.030.030.020.24
Mid50.3732.6111.931.791.540.890.440.10.030.020.020.010.23
Bot48.1034.5712.082.141.430.880.480.080.040.030.010.12
Primary γ′
Top72.172.513.4412.710.250.397.780.560.030.020.050.08
Mid71.602.573.2813.550.420.687.040.730.010.030.040.04
Bot72.342.473.8612.500.260.447.460.500.050.020.020.030.04
Secondary γ′
Mid70.424.203.2314.190.631.035.340.790.030.040.040.05
Bot69.914.063.6814.320.811.045.220.650.050.100.020.11

3.2. Hardness

Fig. 3a shows the Vickers hardness mapping performed along the entire X-Z surface, while Fig. 3b shows the plot of average hardness at different build heights. This hardness distribution is consistent with the γ′ precipitate size gradient across the build direction in Fig. 1Fig. 2. The maximum hardness of ∼530 HV1 is found at ∼0.5 mm away from the top surface (Z = 22.5), where γ′ particles exhibit the smallest observed size in Fig. 2b. Further down the build (∼ 2 mm from the top), the hardness drops to the 440–490 HV1 range. This represents the region where γ′ begins to coarsen. The hardness drops further to 380–430 HV1 at the bottom of the build.

Fig. 3

3.3. Modeling of the microstructural evolution during E-PBF

3.3.1. Thermal profile modeling

Fig. 4 shows the simulated thermal profile of the E-PBF build at a location of 23 mm from the top of the build, using a semi-analytical heat conduction model. This profile consists of the time taken to deposit 460 layers until final cooling, as shown in Fig. 4a. Fig. 4b-d show the magnified regions of Fig. 4a and reveal the first 20 layers from the top, a single layer (first layer from the top), and the time taken for the build to cool after the last layer deposition, respectively.

Fig. 4

The peak temperatures experienced by previous layers decrease progressively as the number of layers increases but never fall below the build preheat temperature (1000 °C). Our simulated thermal cycle may not completely capture the complexity of the actual thermal cycle utilized in the E-PBF build. For instance, the top layer (Fig. 4c), also representing the first deposit’s thermal profile without additional cycles (from powder heating, melting, to solidification), recorded the highest peak temperature of 1390 °C. Although this temperature is above the melting range of the alloy (1230–1360 °C) [62], we believe a much higher temperature was produced by the electron beam to melt the powder. Nevertheless, the solidification temperature and dynamics are outside the scope of this study as our focus is on the solid-state phase transformations during deposition. It takes ∼25 s for each layer to be deposited and cooled to the build temperature. The interlayer dwell time is 125 s. The time taken for the build to cool to room temperature (RT) after final layer deposition is ∼4.7 hrs (17,000 s).

3.3.2. MatCalc simulation

During the MatCalc simulation, the matrix phase is defined as γ. γ′, and MC carbide are included as possible precipitates. The domain of these precipitates is set to be the matrix (γ), and nucleation is assumed to be homogenous. In homogeneous nucleation, all atoms of the unit volume are assumed to be potential nucleation sitesTable 3 shows the computational parameters used in the simulation. All other parameters were set at default values as recommended in the version 6.04.0011 of MatCalc. The values for the interfacial energies are automatically calculated according to the generalized nearest neighbor broken bond model and is one of the most outstanding features in MatCalc [56][57][58]. It should be noted that the elastic misfit strain was not included in the calculation. The output of MatCalc includes phase fraction, size, nucleation rate, and composition of the precipitates. The phase fraction in MatCalc is the volume fraction. Although the experimental phase fraction is the measured area fraction, it is relatively similar to the volume fraction. This is because of the generally larger precipitate size and similar morphology at the various locations along the build [63]. A reliable phase fraction comparison between experiment and simulation can therefore be made.

Table 3. Computational parameters used in the simulation.

Precipitation domainγ
Nucleation site γ′Bulk (homogenous)
Nucleation site MC carbideBulk (Homogenous)
Precipitates class size250
Regular solution critical temperature γ′2500 K[64]
Calculated interfacial energyγ′ = 0.080–0.140 J/m2 and MC carbide = 0.410–0.430 J/m2
3.3.2.1. Precipitate phase fraction

Fig. 5a shows the simulated phase fraction of γ′ and MC carbide during thermal cycling. Fig. 5b is a magnified view of 5a showing the simulated phase fraction at the center points of the top 70 layers, whereas Fig. 5c corresponds to the first two layers from the top. As mentioned earlier, the top layer (460th layer) represents the microstructure after solidification. The microstructure of the layers below is determined by the number of thermal cycles, which increases with distance to the top. For example, layers 459, 458, 457, up to layer 1 (region of interest) experience 1, 2, 3 and 459 thermal cycles, respectively. In the top layer in Fig. 5c, the volume fraction of γ′ and carbides increases with temperature. For γ′, it decreases to zero when the temperature is above the solvus temperature after a few seconds. Carbides, however, remain constant in their volume fraction reaching equilibrium (phase fraction ∼ 0.9%) in a short time. The topmost layer can be compared to the first deposit, and the peak in temperature symbolizes the stage where the electron beam heats the powder until melting. This means γ′ and carbide precipitation might have started in the powder particles during heating from the build temperature and electron beam until the onset of melting, where γ′ dissolves, but carbides remain stable [28].

Fig. 5

During cooling after deposition, γ′ reprecipitates at a temperature of 1085 °C, which is below its solvus temperature. As cooling progresses, the phase fraction increases steadily to ∼27% and remains constant at 1000 °C (elevated build temperature). The calculated equilibrium fraction of phases by MatCalc is used to show the complex precipitation characteristics in this alloy. Fig. 6 shows that MC carbides form during solidification at 1320 °C, followed by γ′, which precipitate when the solidified layer cools to 1140 °C. This indicates that all deposited layers might contain a negligible amount of these precipitates before subsequent layer deposition, while being at the 1000 °C build temperature or during cooling to RT. The phase diagram also shows that the equilibrium fraction of the γ′ increases as temperature decreases. For instance, at 1000, 900, and 800 °C, the phase fractions are ∼30%, 38%, and 42%, respectively.

Fig. 6

Deposition of subsequent layers causes previous layers to undergo phase transformations as they are exposed to several thermal cycles with different peak temperatures. In Fig. 5c, as the subsequent layer is being deposited, γ′ in the previous layer (459th layer) begins to dissolve as the temperature crosses the solvus temperature. This is witnessed by the reduction of the γ′ phase fraction. This graph also shows how this phase dissolves during heating. However, the phase fraction of MC carbide remains stable at high temperatures and no dissolution is seen during thermal cycling. Upon cooling, the γ′ that was dissolved during heating reprecipitates with a surge in the phase fraction until 1000 °C, after which it remains constant. This microstructure is similar to the solidification microstructure (layer 460), with a similar γ′ phase fraction (∼27%).

The complete dissolution and reprecipitation of γ′ continue for several cycles until the 50th layer from the top (layer 411), where the phase fraction does not reach zero during heating to the peak temperature (see Fig. 5d). This indicates the ‘partial’ dissolution of γ′, which continues progressively with additional layers. It should be noted that the peak temperatures for layers that underwent complete dissolution were much higher (1170–1300 °C) than the γ′ solvus.

The dissolution and reprecipitation of γ′ during thermal cycling are further confirmed in Fig. 7, which summarizes the nucleation rate, phase fraction, and concentration of major elements that form γ′ in the matrix. Fig. 7b magnifies a single layer (3rd layer from top) within the full dissolution region in Fig. 7a to help identify the nucleation and growth mechanisms. From Fig. 7b, γ′ nucleation begins during cooling whereby the nucleation rate increases to reach a maximum value of approximately 1 × 1020 m−3s−1. This fast kinetics implies that some rearrangement of atoms is required for γ′ precipitates to form in the matrix [65][66]. The matrix at this stage is in a non-equilibrium condition. Its composition is similar to the nominal composition and remains unchanged. The phase fraction remains insignificant at this stage although nucleation has started. The nucleation rate starts declining upon reaching the peak value. Simultaneously, diffusion-controlled growth of existing nuclei occurs, depleting the matrix of γ′ forming elements (Al and Ti). Thus, from (7)(11), ∆�vol continuously decreases until nucleation ceases. The growth of nuclei is witnessed by the increase in phase fraction until a constant level is reached at 27% upon cooling to and holding at build temperature. This nucleation event is repeated several times.

Fig. 7

At the onset of partial dissolution, the nucleation rate jumps to 1 × 1021 m−3s−1, and then reduces sharply at the middle stage of partial dissolution. The nucleation rate reaches 0 at a later stage. Supplementary Fig. S1 shows a magnified view of the nucleation rate, phase fraction, and thermal profile, underpinning this trend. The jump in nucleation rate at the onset is followed by a progressive reduction in the solute content of the matrix. The peak temperatures (∼1130–1160 °C) are lower than those in complete dissolution regions but still above or close to the γ′ solvus. The maximum phase fraction (∼27%) is similar to that of the complete dissolution regions. At the middle stage, the reduction in nucleation rate is accompanied by a sharp drop in the matrix composition. The γ′ fraction drops to ∼24%, where the peak temperatures of the layers are just below or at γ′ solvus. The phase fraction then increases progressively through the later stage of partial dissolution to ∼30% towards the end of thermal cycling. The matrix solute content continues to drop although no nucleation event is seen. The peak temperatures are then far below the γ′ solvus. It should be noted that the matrix concentration after complete dissolution remains constant. Upon cooling to RT after final layer deposition, the nucleation rate increases again, indicating new nucleation events. The phase fraction reaches ∼40%, with a further depletion of the matrix in major γ′ forming elements.

3.3.2.2. γ′ size distribution

Fig. 8 shows histograms of the γ′ precipitate size distributions (PSD) along the build height during deposition. These PSDs are predicted at the end of each layer of interest just before final cooling to room temperature, to separate the role of thermal cycles from final cooling on the evolution of γ′. The PSD for the top layer (layer 460) is shown in Fig. 8a (last solidified region with solidification microstructure). The γ′ size ranges from 120 to 230 nm and is similar to the 44 layers below (2.2 mm from the top).

Fig. 8

Further down the build, γ′ begins to coarsen after layer 417 (44th layer from top). Fig. 8c shows the PSD after the 44th layer, where the γ′ size exhibits two peaks at ∼120–230 and ∼300 nm, with most of the population being in the former range. This is the onset of partial dissolution where simultaneously with the reprecipitation and growth of fresh γ′, the undissolved γ′ grows rapidly through diffusive transport of atoms to the precipitates. This is shown in Fig. 8c, where the precipitate class sizes between 250 and 350 represent the growth of undissolved γ′. Although this continues in the 416th layer, the phase fractions plot indicates that the onset of partial dissolution begins after the 411th layer. This implies that partial dissolution started early, but the fraction of undissolved γ′ was too low to impact the phase fraction. The reprecipitated γ′ are mostly in the 100–220 nm class range and similar to those observed during full dissolution.

As the number of layers increases, coarsening intensifies with continued growth of more undissolved γ′, and reprecipitation and growth of partially dissolved ones. Fig. 8d, e, and f show this sequence. Further down the build, coarsening progresses rapidly, as shown in Figs. 8d, 8e, and 8f. The γ′ size ranges from 120 to 1100 nm, with the peaks at 160, 180, and 220 nm in Figs. 8d, 8e, and 8f, respectively. Coarsening continues until nucleation ends during dissolution, where only the already formed γ′ precipitates continue to grow during further thermal cycling. The γ′ size at this point is much larger, as observed in layers 361 and 261, and continues to increase steadily towards the bottom (layer 1). Two populations in the ranges of ∼380–700 and ∼750–1100 nm, respectively, can be seen. The steady growth of γ′ towards the bottom is confirmed by the gradual decrease in the concentration of solute elements in the matrix (Fig. 7a). It should be noted that for each layer, the γ′ class with the largest size originates from continuous growth of the earliest set of the undissolved precipitates.

Fig. 9Fig. 10 and supplementary Figs. S2 and S3 show the γ′ size evolution during heating and cooling of a single layer in the full dissolution region, and early, middle stages, and later stages of partial dissolution, respectively. In all, the size of γ′ reduces during layer heating. Depending on the peak temperature of the layer which varies with build height, γ′ are either fully or partially dissolved as mentioned earlier. Upon cooling, the dissolved γ′ reprecipitate.

Fig. 9
Fig. 10

In Fig. 9, those layers that underwent complete dissolution (top layers) were held above γ′ solvus temperature for longer. In Fig. 10, layers at the early stage of partial dissolution spend less time in the γ′ solvus temperature region during heating, leading to incomplete dissolution. In such conditions, smaller precipitates are fully dissolved while larger ones shrink [67]. Layers in the middle stages of partial dissolution have peak temperatures just below or at γ′ solvus, not sufficient to achieve significant γ′ dissolution. As seen in supplementary Fig. S2, only a few smaller γ′ are dissolved back into the matrix during heating, i.e., growth of precipitates is more significant than dissolution. This explains the sharp decrease in concentration of Al and Ti in the matrix in this layer.

The previous sections indicate various phenomena such as an increase in phase fraction, further depletion of matrix composition, and new nucleation bursts during cooling. Analysis of the PSD after the final cooling of the build to room temperature allows a direct comparison to post-printing microstructural characterization. Fig. 11 shows the γ′ size distribution of layer 1 (460th layer from the top) after final cooling to room temperature. Precipitation of secondary γ′ is observed, leading to the multimodal size distribution of secondary and primary γ′. The secondary γ′ size falls within the 10–80 nm range. As expected, a further growth of the existing primary γ′ is also observed during cooling.

Fig. 11
3.3.2.3. γ′ chemistry after deposition

Fig. 12 shows the concentration of the major elements that form γ′ (Al, Ti, and Ni) in the primary and secondary γ′ at the bottom of the build, as calculated by MatCalc. The secondary γ′ has a higher Al content (13.5–14.5 at% Al), compared to 13 at% Al in the primary γ′. Additionally, within the secondary γ′, the smallest particles (∼10 nm) have higher Al contents than larger ones (∼70 nm). In contrast, for the primary γ′, there is no significant variation in the Al content as a function of their size. The Ni concentration in secondary γ′ (71.1–72 at%) is also higher in comparison to the primary γ′ (70 at%). The smallest secondary γ′ (∼10 nm) have higher Ni contents than larger ones (∼70 nm), whereas there is no substantial change in the Ni content of primary γ′, based on their size. As expected, Ti shows an opposite size-dependent variation. It ranges from ∼ 7.7–8.7 at% Ti in secondary γ′ to ∼9.2 at% in primary γ′. Similarly, within the secondary γ′, the smallest (∼10 nm) have lower Al contents than the larger ones (∼70 nm). No significant variation is observed for Ti content in primary γ′.

Fig. 12

4. Discussion

A combined modelling method is utilized to study the microstructural evolution during E-PBF of IN738. The presented results are discussed by examining the precipitation and dissolution mechanism of γ′ during thermal cycling. This is followed by a discussion on the phase fraction and size evolution of γ′ during thermal cycling and after final cooling. A brief discussion on carbide morphology is also made. Finally, a comparison is made between the simulation and experimental results to assess their agreement.

4.1. γ′ morphology as a function of build height

4.1.1. Nucleation of γ′

The fast precipitation kinetics of the γ′ phase enables formation of γ′ upon quenching from higher temperatures (above solvus) during thermal cycling [66]. In Fig. 7b, for a single layer in the full dissolution region, during cooling, the initial increase in nucleation rate signifies the first formation of nuclei. The slight increase in nucleation rate during partial dissolution, despite a decrease in the concentration of γ′ forming elements, may be explained by the nucleation kinetics. During partial dissolution and as the precipitates shrink, it is assumed that the regions at the vicinity of partially dissolved precipitates are enriched in γ′ forming elements [68][69]. This differs from the full dissolution region, in which case the chemical composition is evenly distributed in the matrix. Several authors have attributed the solute supersaturation of the matrix around primary γ′ to partial dissolution during isothermal ageing [69][70][71][72]. The enhanced supersaturation in the regions close to the precipitates results in a much higher driving force for nucleation, leading to a higher nucleation rate upon cooling. This phenomenon can be closely related to the several nucleation bursts upon continuous cooling of Ni-based superalloys, where second nucleation bursts exhibit higher nucleation rates [38][68][73][74].

At middle stages of partial dissolution, the reduction in the nucleation rate indicates that the existing composition and low supersaturation did not trigger nucleation as the matrix was closer to the equilibrium state. The end of a nucleation burst means that the supersaturation of Al and Ti has reached a low level, incapable of providing sufficient driving force during cooling to or holding at 1000 °C for further nucleation [73]. Earlier studies on Ni-based superalloys have reported the same phenomenon during ageing or continuous cooling from the solvus temperature to RT [38][73][74].

4.1.2. Dissolution of γ′ during thermal cycling

γ′ dissolution kinetics during heating are fast when compared to nucleation due to exponential increase in phase transformation and diffusion activities with temperature [65]. As shown in Fig. 9Fig. 10, and supplementary Figs. S2 and S3, the reduction in γ′ phase fraction and size during heating indicates γ′ dissolution. This is also revealed in Fig. 5 where phase fraction decreases upon heating. The extent of γ′ dissolution mostly depends on the temperature, time spent above γ′ solvus, and precipitate size [75][76][77]. Smaller γ′ precipitates are first to be dissolved [67][77][78]. This is mainly because more solute elements need to be transported away from large γ′ precipitates than from smaller ones [79]. Also, a high temperature above γ′ solvus temperature leads to a faster dissolution rate [80]. The equilibrium solvus temperature of γ′ in IN738 in our MatCalc simulation (Fig. 6) and as reported by Ojo et al. [47] is 1140 °C and 1130–1180 °C, respectively. This means the peak temperature experienced by previous layers decreases progressively from γ′ supersolvus to subsolvus, near-solvus, and far from solvus as the number of subsequent layers increases. Based on the above, it can be inferred that the degree of dissolution of γ′ contributes to the gradient in precipitate distribution.

Although the peak temperatures during later stages of partial dissolution are much lower than the equilibrium γ′ solvus, γ′ dissolution still occurs but at a significantly lower rate (supplementary Fig. S3). Wahlmann et al. [28] also reported a similar case where they observed the rapid dissolution of γ′ in CMSX-4 during fast heating and cooling cycles at temperatures below the γ′ solvus. They attributed this to the γ′ phase transformation process taking place in conditions far from the equilibrium. While the same reasoning may be valid for our study, we further believe that the greater surface area to volume ratio of the small γ′ precipitates contributed to this. This ratio means a larger area is available for solute atoms to diffuse into the matrix even at temperatures much below the solvus [81].

4.2. γ′ phase fraction and size evolution

4.2.1. During thermal cycling

In the first layer, the steep increase in γ′ phase fraction during heating (Fig. 5), which also represents γ′ precipitation in the powder before melting, has qualitatively been validated in [28]. The maximum phase fraction of 27% during the first few layers of thermal cycling indicates that IN738 theoretically could reach the equilibrium state (∼30%), but the short interlayer time at the build temperature counteracts this. The drop in phase fraction at middle stages of partial dissolution is due to the low number of γ′ nucleation sites [73]. It has been reported that a reduction of γ′ nucleation sites leads to a delay in obtaining the final volume fraction as more time is required for γ′ precipitates to grow and reach equilibrium [82]. This explains why even upon holding for 150 s before subsequent layer deposition, the phase fraction does not increase to those values that were observed in the previous full γ′ dissolution regions. Towards the end of deposition, the increase in phase fraction to the equilibrium value of 30% is as a result of the longer holding at build temperature or close to it [83].

During thermal cycling, γ′ particles begin to grow immediately after they first precipitate upon cooling. This is reflected in the rapid increase in phase fraction and size during cooling in Fig. 5 and supplementary Fig. S2, respectively. The rapid growth is due to the fast diffusion of solute elements at high temperatures [84]. The similar size of γ′ for the first 44 layers from the top can be attributed to the fact that all layers underwent complete dissolution and hence, experienced the same nucleation event and growth during deposition. This corresponds with the findings by Balikci et al. [85], who reported that the degree of γ′ precipitation in IN738LC does not change when a solution heat treatment is conducted above a certain critical temperature.

The increase in coarsening rate (Fig. 8) during thermal cycling can first be ascribed to the high peak temperature of the layers [86]. The coarsening rate of γ′ is known to increase rapidly with temperature due to the exponential growth of diffusion activity. Also, the simultaneous dissolution with coarsening could be another reason for the high coarsening rate, as γ′ coarsening is a diffusion-driven process where large particles grow by consuming smaller ones [78][84][86][87]. The steady growth of γ′ towards the bottom of the build is due to the much lower layer peak temperature, which is almost close to the build temperature, and reduced dissolution activity, as is seen in the much lower solute concentration in γ′ compared to those in the full and partial dissolution regions.

4.2.2. During cooling

The much higher phase fraction of ∼40% upon cooling signifies the tendency of γ′ to reach equilibrium at lower temperatures (Fig. 4). This is due to the precipitation of secondary γ′ and a further increase in the size of existing primary γ′, which leads to a multimodal size distribution of γ′ after cooling [38][73][88][89][90]. The reason for secondary γ′ formation during cooling is as follows: As cooling progresses, it becomes increasingly challenging to redistribute solute elements in the matrix owing to their lower mobility [38][73]. A higher supersaturation level in regions away from or free of the existing γ′ precipitates is achieved, making them suitable sites for additional nucleation bursts. More cooling leads to the growth of these secondary γ′ precipitates, but as the temperature and in turn, the solute diffusivity is low, growth remains slow.

4.3. Carbides

MC carbides in IN738 are known to have a significant impact on the high-temperature strength. They can also act as effective hardening particles and improve the creep resistance [91]. Precipitation of MC carbides in IN738 and several other superalloys is known to occur during solidification or thermal treatments (e.g., hot isostatic pressing) [92]. In our case, this means that the MC carbides within the E-PBF build formed because of the thermal exposure from the E-PBF thermal cycle in addition to initial solidification. Our simulation confirms this as MC carbides appear during layer heating (Fig. 5). The constant and stable phase fraction of MC carbides during thermal cycling can be attributed to their high melting point (∼1360 °C) and the short holding time at peak temperatures [75][93][94]. The solvus temperature for most MC carbides exceeds most of the peak temperatures observed in our simulation, and carbide dissolution kinetics at temperatures above the solvus are known to be comparably slow [95]. The stable phase fraction and random distribution of MC carbides signifies the slight influence on the gradient in hardness.

4.4. Comparison of simulations and experiments

4.4.1. Precipitate phase fraction and morphology as a function of build height

A qualitative agreement is observed for the phase fraction of carbides, i.e. ∼0.8% in the experiment and ∼0.9% in the simulation. The phase fraction of γ′ differs, with the experiment reporting a value of ∼51% and the simulation, 40%. Despite this, the size distribution of primary γ′ along the build shows remarkable consistency between experimental and computational analyses. It is worth noting that the primary γ′ morphology in the experimental analysis is observed in the as-fabricated state, whereas the simulation (Fig. 8) captures it during deposition process. The primary γ′ size in the experiment is expected to experience additional growth during the cooling phase. Regardless, both show similar trends in primary γ′ size increments from the top to the bottom of the build. The larger primary γ’ size in the simulation versus the experiment can be attributed to the fact that experimental and simulation results are based on 2D and 3D data, respectively. The absence of stereological considerations [96] in our analysis could have led to an underestimation of the precipitate sizes from SEM measurements. The early starts of coarsening (8th layer) in the experiment compared to the simulation (45th layer) can be attributed to a higher actual γ′ solvus temperature than considered in our simulation [47]. The solvus temperature of γ′ in a Ni-based superalloy is mainly determined by the detailed composition. A high amount of Cr and Co are known to reduce the solvus temperature, whereas Ta and Mo will increase it [97][98][99]. The elemental composition from our experimental work was used for the simulation except for Ta. It should be noted that Ta is not included in the thermodynamic database in MatCalc used, and this may have reduced the solvus temperature. This could also explain the relatively higher γ′ phase fraction in the experiment than in simulation, as a higher γ′ solvus temperature will cause more γ′ to precipitate and grow early during cooling [99][100].

Another possible cause of this deviation can be attributed to the extent of γ′ dissolution, which is mainly determined by the peak temperature. It can be speculated that individual peak temperatures at different layers in the simulation may have been over-predicted. However, one needs to consider that the true thermal profile is likely more complicated in the actual E-PBF process [101]. For example, the current model assumes that the thermophysical properties of the material are temperature-independent, which is not realistic. Many materials, including IN738, exhibit temperature-dependent properties such as thermal conductivityspecific heat capacity, and density [102]. This means that heat transfer simulations may underestimate or overestimate the temperature gradients and cooling rates within the powder bed and the solidified part. Additionally, the model does not account for the reduced thermal diffusivity through unmelted powder, where gas separating the powder acts as insulation, impeding the heat flow [1]. In E-PBF, the unmelted powder regions with trapped gas have lower thermal diffusivity compared to the fully melted regions, leading to localized temperature variations, and altered solidification behavior. These limitations can impact the predictions, particularly in relation to the carbide dissolution, as the peak temperatures may be underestimated.

While acknowledging these limitations, it is worth emphasizing that achieving a detailed and accurate representation of each layer’s heat source would impose tough computational challenges. Given the substantial layer count in E-PBF, our decision to employ a semi-analytical approximation strikes a balance between computational feasibility and the capture of essential trends in thermal profiles across diverse build layers. In future work, a dual-calibration strategy is proposed to further reduce simulation-experiment disparities. By refining temperature-independent thermophysical property approximations and absorptivity in the heat source model, and by optimizing interfacial energy descriptions in the kinetic model, the predictive precision could be enhanced. Further refining the simulation controls, such as adjusting the precipitate class size may enhance quantitative comparisons between modeling outcomes and experimental data in future work.

4.4.2. Multimodal size distribution of γ′ and concentration

Another interesting feature that sees qualitative agreement between the simulation and the experiment is the multimodal size distribution of γ′. The formation of secondary γ′ particles in the experiment and most E-PBF Ni-based superalloys is suggested to occur at low temperatures, during final cooling to RT [16][73][90]. However, so far, this conclusion has been based on findings from various continuous cooling experiments, as the study of the evolution during AM would require an in-situ approach. Our simulation unambiguously confirms this in an AM context by providing evidence for secondary γ′ precipitation during slow cooling to RT. Additionally, it is possible to speculate that the chemical segregation occurring during solidification, due to the preferential partitioning of certain elements between the solid and liquid phases, can contribute to the multimodal size distribution during deposition [51]. This is because chemical segregation can result in variations in the local composition of superalloys, which subsequently affects the nucleation and growth of γ′. Regions with higher concentrations of alloying elements will encourage the formation of larger γ′ particles, while regions with lower concentrations may favor the nucleation of smaller precipitates. However, it is important to acknowledge that the elevated temperature during the E-PBF process will largely homogenize these compositional differences [103][104].

A good correlation is also shown in the composition of major γ′ forming elements (Al and Ti) in primary and secondary γ′. Both experiment and simulation show an increasing trend for Al content and a decreasing trend for Ti content from primary to secondary γ′. The slight composition differences between primary and secondary γ′ particles are due to the different diffusivity of γ′ stabilizers at different thermal conditions [105][106]. As the formation of multimodal γ′ particles with different sizes occurs over a broad temperature range, the phase chemistry of γ′ will be highly size dependent. The changes in the chemistry of various γ′ (primary, secondary, and tertiary) have received significant attention since they have a direct influence on the performance [68][105][107][108][109]. Chen et al. [108][109], reported a high Al content in the smallest γ′ precipitates compared to the largest, while Ti showed an opposite trend during continuous cooling in a RR1000 Ni-based superalloy. This was attributed to the temperature and cooling rate at which the γ′ precipitates were formed. The smallest precipitates formed last, at the lowest temperature and cooling rate. A comparable observation is evident in the present investigation, where the secondary γ′ forms at a low temperature and cooling rate in comparison to the primary. The temperature dependence of γ′ chemical composition is further evidenced in supplementary Fig. S4, which shows the equilibrium chemical composition of γ′ as a function of temperature.

5. Conclusions

A correlative modelling approach capable of predicting solid-state phase transformations kinetics in metal AM was developed. This approach involves computational simulations with a semi-analytical heat transfer model and the MatCalc thermo-kinetic software. The method was used to predict the phase transformation kinetics and detailed morphology and chemistry of γ′ and MC during E-PBF of IN738 Ni-based superalloy. The main conclusions are:

  • 1.The computational simulations are in qualitative agreement with the experimental observations. This is particularly true for the γ′ size distribution along the build height, the multimodal size distribution of particles, and the phase fraction of MC carbides.
  • 2.The deviations between simulation and experiment in terms of γ′ phase fraction and location in the build are most likely attributed to a higher γ′ solvus temperature during the experiment than in the simulation, which is argued to be related to the absence of Ta in the MatCalc database.
  • 3.The dissolution and precipitation of γ′ occur fast and under non-equilibrium conditions. The level of γ′ dissolution determines the gradient in γ′ size distribution along the build. After thermal cycling, the final cooling to room temperature has further significant impacts on the final γ′ size, morphology, and distribution.
  • 4.A negligible amount of γ′ forms in the first deposited layer before subsequent layer deposition, and a small amount of γ′ may also form in the powder induced by the 1000 °C elevated build temperature before melting.

Our findings confirm the suitability of MatCalc to predict the microstructural evolution at various positions throughout a build in a Ni-based superalloy during E-PBF. It also showcases the suitability of a tool which was originally developed for traditional thermo-mechanical processing of alloys to the new additive manufacturing context. Our simulation capabilities are likely extendable to other alloy systems that undergo solid-state phase transformations implemented in MatCalc (various steels, Ni-based superalloys, and Al-alloys amongst others) as well as other AM processes such as L-DED and L-PBF which have different thermal cycle characteristics. New tools to predict the microstructural evolution and properties during metal AM are important as they provide new insights into the complexities of AM. This will enable control and design of AM microstructures towards advanced materials properties and performances.

CRediT authorship contribution statement

Primig Sophie: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization. Adomako Nana Kwabena: Writing – original draft, Writing – review & editing, Visualization, Software, Investigation, Formal analysis, Conceptualization. Haghdadi Nima: Writing – review & editing, Supervision, Project administration, Methodology, Conceptualization. Dingle James F.L.: Methodology, Conceptualization, Software, Writing – review & editing, Visualization. Kozeschnik Ernst: Writing – review & editing, Software, Methodology. Liao Xiaozhou: Writing – review & editing, Project administration, Funding acquisition. Ringer Simon P: Writing – review & editing, Project administration, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research was sponsored by the Department of Industry, Innovation, and Science under the auspices of the AUSMURI program – which is a part of the Commonwealth’s Next Generation Technologies Fund. The authors acknowledge the facilities and the scientific and technical assistance at the Electron Microscope Unit (EMU) within the Mark Wainwright Analytical Centre (MWAC) at UNSW Sydney and Microscopy Australia. Nana Adomako is supported by a UNSW Scientia PhD scholarship. Michael Haines’ (UNSW Sydney) contribution to the revised version of the original manuscript is thankfully acknowledged.

Appendix A. Supplementary material

Download : Download Word document (462KB)

Supplementary material.

Data Availability

Data will be made available on request.

References

Strain rate magnitude at the free surface, illustrating Kelvin-Helmoltz (KH) shear instabilities.

On the reef scale hydrodynamics at Sodwana Bay, South Africa

Environmental Fluid Mechanics (2022)Cite this article

Abstract

The hydrodynamics of coral reefs strongly influences their biological functioning, impacting processes such as nutrient availability and uptake, recruitment success and bleaching. For example, coral reefs located in oligotrophic regions depend on upwelling for nutrient supply. Coral reefs at Sodwana Bay, located on the east coast of South Africa, are an example of high latitude marginal reefs. These reefs are subjected to complex hydrodynamic forcings due to the interaction between the strong Agulhas current and the highly variable topography of the region. In this study, we explore the reef scale hydrodynamics resulting from the bathymetry for two steady current scenarios at Two-Mile Reef (TMR) using a combination of field data and numerical simulations. The influence of tides or waves was not considered for this study as well as reef-scale roughness. Tilt current meters with onboard temperature sensors were deployed at selected locations within TMR. We used field observations to identify the dominant flow conditions on the reef for numerical simulations that focused on the hydrodynamics driven by mean currents. During the field campaign, southerly currents were the predominant flow feature with occasional flow reversals to the north. Northerly currents were associated with greater variability towards the southern end of TMR. Numerical simulations showed that Jesser Point was central to the development of flow features for both the northerly and southerly current scenarios. High current variability in the south of TMR during reverse currents is related to the formation of Kelvin-Helmholtz type shear instabilities along the outer edge of an eddy formed north of Jesser Point. Furthermore, downward vertical velocities were computed along the offshore shelf at TMR during southerly currents. Current reversals caused a change in vertical velocities to an upward direction due to the orientation of the bathymetry relative to flow directions.

Highlights

  • A predominant southerly current was measured at Two-Mile Reef with occasional reversals towards the north.
  • Field observations indicated that northerly currents are spatially varied along Two-Mile Reef.
  • Simulation of reverse currents show the formation of a separated flow due to interaction with Jesser Point with Kelvin–Helmholtz type shear instabilities along the seaward edge.

지금까지 Sodwana Bay에서 자세한 암초 규모 유체 역학을 모델링하려는 시도는 없었습니다. 이러한 모델의 결과는 규모가 있는 산호초 사이의 흐름이 산호초 건강에 어떤 영향을 미치는지 탐색하는 데 사용할 수 있습니다. 이 연구에서는 Sodwana Bay의 유체역학을 탐색하는 데 사용할 수 있는 LES 모델을 개발하기 위한 단계별 접근 방식을 구현합니다. 여기서 우리는 이 초기 단계에서 파도와 조수의 영향을 배제하면서 Agulhas 해류의 유체역학에 초점을 맞춥니다. 이 접근법은 흐름의 첫 번째 LES를 제시하고 Sodwana Bay의 산호초에서 혼합함으로써 향후 연구의 기초를 제공합니다.

This is a preview of subscription content, access via your institution.

References

  1. Anarde K, Myres H, Figlus J (2016) Tilt current meter field validation in the surf zone. In: AGU fall meeting abstracts, vol 2016, pp EP23A—-0950
  2. Blocken B (2018) LES over RANS in building simulation for outdoor and indoor applications: A foregone conclusion? Build Simul 11(5):821–870. https://doi.org/10.1007/s12273-018-0459-3Article Google Scholar 
  3. Booij N, Ris RC, Holthuijsen LH (1999) A third-generation wave model for coastal regions: 1. Model description and validation. J Geophys Res Ocean 104(C4):7649–7666. https://doi.org/10.1029/98JC02622Article Google Scholar 
  4. Bouffanais R (2010) Advances and challenges of applied large-eddy simulation. Comput Fluids 39:735–738. https://doi.org/10.1016/j.compfluid.2009.12.003Article Google Scholar 
  5. Celliers L, Schleyer MH (2002) Coral bleaching on high-latitude marginal reefs at Sodwana Bay, South Africa. Mar Pollut Bull 44:1380–1387Article Google Scholar 
  6. Celliers L, Schleyer MH (2008) Coral community structure and risk assessment of high-latitude reefs at Sodwana Bay, South Africa. Biodivers Conserv 17(13):3097–3117. https://doi.org/10.1007/s10531-007-9271-6Article Google Scholar 
  7. Chen SC (2018) Performance assessment of FLOW-3D and XFlow in the numerical modelling of fish-bone type fishway hydraulics https://doi.org/10.15142/T3HH1J
  8. Corbella S, Pringle J, Stretch DD (2015) Assimilation of ocean wave spectra and atmospheric circulation patterns to improve wave modelling. Coast Eng 100:1–10. https://doi.org/10.1016/j.coastaleng.2015.03.003Article Google Scholar 
  9. Davis KA, Pawlak G, Monismith SG (2021) Turbulence and coral reefs. Ann Rev Mar Sci. https://doi.org/10.1146/annurev-marine-042120-071823Article Google Scholar 
  10. Flow Science Inc (2018) FLOW-3D, Version 12.0 Users Manual. Santa Fe, NM, https://www.flow3d.com/
  11. Flow Science Inc (2019) FLOW-3D, Version 12.0 [Computer Software]. Santa Fe, NM, https://www.flow3d.com/
  12. Franco A, Moernaut J, Schneider-Muntau B, Strasser M, Gems B (2020) The 1958 Lituya Bay tsunami – pre-event bathymetry reconstruction and 3D numerical modelling utilising the computational fluid dynamics software Flow-3D. Nat Hazards Earth Syst Sci 20(8):2255–2279Article Google Scholar 
  13. Fringer OB, Gerritsen M, Street RL (2006) An unstructured-grid, finite-volume, nonhydrostatic, parallel coastal ocean simulator. Ocean Model 14(3):139–173Article Google Scholar 
  14. Fringer OB, Dawson CN, He R, Ralston DK, Zhang YJ (2019) The future of coastal and estuarine modeling: findings from a workshop. Ocean Model 143(September):101458. https://doi.org/10.1016/j.ocemod.2019.101458Article Google Scholar 
  15. Glassom D, Celliers L, Schleyer MH (2006) Coral recruitment patterns at Sodwana Bay, South Africa. Coral Reefs 25(3):485–492. https://doi.org/10.1007/s00338-006-0117-6Article Google Scholar 
  16. Gomes A, Pinho JLS, Valente T, do Carmo JS, Hegde VA (2020) Performance assessment of a semi-circular breakwater through CFD modelling. J Mar Sci Eng. https://doi.org/10.3390/jmse8030226Article Google Scholar 
  17. Green RH, Lowe RJ, Buckley ML (2018) Hydrodynamics of a tidally forced coral reef atoll. J Geophys Res Oceans 123(10):7084–7101. https://doi.org/10.1029/2018JC013946Article Google Scholar 
  18. Hansen AB, Carstensen S, Christensen DF, Aagaard T (2017) Performance of a tilt current meter in the surf zone. Coastal dynamics
  19. Hench JL, Rosman JH (2013) Observations of spatial flow patterns at the coral colony scale on a shallow reef flat. J Geophys Res Ocean 118(3):1142–1156. https://doi.org/10.1002/jgrc.20105Article Google Scholar 
  20. Hirt CW (1993) Volume-fraction techniques: powerful tools for wind engineering. J Wind Eng Ind Aerodyn 46–47:327–338. https://doi.org/10.1016/0167-6105(93)90298-3Article Google Scholar 
  21. Hirt CW, Sicilian JM (1985) A porosity technique for the definition of obstacles in rectangular cell meshes. In: Proceedings of 4th International Conference on Ship Hydrodynamics https://ci.nii.ac.jp/naid/10009570543/en/
  22. Hocker LO, Hruska MA (2004) Interleaving synchronous data and asynchronous data in a single data storage file
  23. Hossain MM, Staples AE (2020) Effects of coral colony morphology on turbulent flow dynamics. PLoS ONE 15(10):e0225676. https://doi.org/10.1371/journal.pone.0225676Article Google Scholar 
  24. Jacob B, Stanev EV (2021) Understanding the impact of bathymetric changes in the german bight on coastal hydrodynamics: one step toward realistic morphodynamic modeling. Front Mar Sci. https://doi.org/10.3389/fmars.2021.640214Article Google Scholar 
  25. Koehl MAR, Hadfield MG (2010) Hydrodynamics of larval settlement from a larva’s point of view. Integr Comp Biol 50(4):539–551. https://doi.org/10.1093/icb/icq101Article Google Scholar 
  26. Lim A, Wheeler AJ, Price DM, O’Reilly L, Harris K, Conti L (2020) Influence of benthic currents on cold-water coral habitats: a combined benthic monitoring and 3D photogrammetric investigation. Sci Rep 10(1):19433. https://doi.org/10.1038/s41598-020-76446-yArticle Google Scholar 
  27. Limer BD, Bloomberg J, Holstein DM (2020) The influence of eddies on coral larval retention in the flower garden banks. Front Mar Sci 7:372. https://doi.org/10.3389/fmars.2020.00372Article Google Scholar 
  28. Monismith SG (2007) Hydrodynamics of coral reefs. Annu Rev Fluid Mech 39(1):37–55. https://doi.org/10.1146/annurev.fluid.38.050304.092125Article Google Scholar 
  29. Morris T (2009) Physical oceanography of Sodwana Bay and its effect on larval transport and coral bleaching. PhD thesis, Cape Peninsula University of Technology
  30. Morris T, Lamont T, Roberts MJ (2013) Effects of deep-sea eddies on the northern KwaZulu-Natal shelf, South Africa. Afr J Mar Sci 35(3):343–350. https://doi.org/10.2989/1814232X.2013.827991Article Google Scholar 
  31. Perry C, Larcombe P (2003) Marginal and non-reef-building coral environments. Coral Reefs 22:427–432. https://doi.org/10.1007/s00338-003-0330-5Article Google Scholar 
  32. Pope SB (2001) Turbulent flows. Cambridge University Press, CambridgeGoogle Scholar 
  33. Porter SN (2009) Biogeography and potential factors regulating shallow subtidal reef communities in the Western Indian Ocean. PhD thesis, University of Cape Town
  34. Porter SN, Schleyer MH (2017) Long-term dynamics of a high-latitude coral reef community at Sodwana Bay, South Africa. Coral Reefs 36(2):369–382. https://doi.org/10.1007/s00338-016-1531-zArticle Google Scholar 
  35. Porter SN, Schleyer MH (2019) Environmental variation and how its spatial structure influences the cross-shelf distribution of high-latitude coral communities in South Africa. Diversity. https://doi.org/10.3390/d11040057Article Google Scholar 
  36. Ramsay PJ (1994) Marine geology of the Sodwana Bay shelf, southeast Africa. Mar Geol 120(3–4):225–247. https://doi.org/10.1016/0025-3227(94)90060-4Article Google Scholar 
  37. Ramsay PJ, Mason TR (1990) Development of a type zoning model for Zululand coral reefs, Sodwana Bay, South Africa. J Coastal Res 6(4):829–852Google Scholar 
  38. Reguero BG, Beck MW, Agostini VN, Kramer P, Hancock B (2018) Coral reefs for coastal protection: a new methodological approach and engineering case study in Grenada. J Environ Manag 210:146–161. https://doi.org/10.1016/j.jenvman.2018.01.024Article Google Scholar 
  39. Reidenbach M, Stocking J, Szczyrba L, Wendelken C (2021) Hydrodynamic interactions with coral topography and its impact on larval settlement. Coral Reefs 40:1–15. https://doi.org/10.1007/s00338-021-02069-yArticle Google Scholar 
  40. Reidenbach MA, Koseff JR, Koehl MAR (2009) Hydrodynamic forces on larvae affect their settlement on coral reefs in turbulent, wave-driven flow. Limnol Oceanogr 54(1):318–330. https://doi.org/10.4319/lo.2009.54.1.0318Article Google Scholar 
  41. Roberts H, Richardson J, Lagumbay R, Meselhe E, Ma Y (2013) Hydrodynamic and sediment transport modeling using FLOW-3D for siting and optimization of the LCA medium diversion at white ditch hydrodynamic and sediment transport modeling using FLOW-3D for siting and optimization of the LCA medium diversion at white D (December)
  42. Roberts MJ, Ribbink AJ, Morris T, Berg MAVD, Engelbrecht DC, Harding RT (2006) Oceanographic environment of the Sodwana Bay coelacanths (Latimeria chalumnae), South Africa: coelacanth research. South Afr J Sci 102(9):435–443Google Scholar 
  43. Rogers JS, Monismith SG, Feddersen F, Storlazzi CD (2013) Hydrodynamics of spur and groove formations on a coral reef. J Geophys Res Ocean 118(6):3059–3073. https://doi.org/10.1002/jgrc.20225Article Google Scholar 
  44. Rogers JS, Monismith SG, Koweek DA, Torres WI, Dunbar RB (2016) Thermodynamics and hydrodynamics in an atoll reef system and their influence on coral cover. Limnol Oceanogr 61(6):2191–2206. https://doi.org/10.1002/lno.10365Article Google Scholar 
  45. Schleyer MH, Celliers L (2003) Coral dominance at the reef-sediment interface in marginal coral communities at Sodwana Bay, South Africa. Mar Freshw Res 54(8):967–972. https://doi.org/10.1071/MF02049Article Google Scholar 
  46. Schleyer MH, Porter SN (2018) Chapter One – drivers of soft and stony coral community distribution on the high-latitude coral reefs of South Africa. advances in marine biology, vol 80, Academic Press, pp 1–55, https://doi.org/10.1016/bs.amb.2018.09.001
  47. Scott F, Antolinez JAA, McCall R, Storlazzi C, Reniers A, Pearson S (2020) Hydro-morphological characterization of coral reefs for wave runup prediction. Front Mar Sci 7:361. https://doi.org/10.3389/fmars.2020.00361Article Google Scholar 
  48. Sebens KP, Grace SP, Helmuth B, Maney EJ Jr, Miles JS (1998) Water flow and prey capture by three scleractinian corals, Madracis mirabilis, Montastrea cavernosa and Porites porites, in a field enclosure. Mar Biol 131(2):347–360Article Google Scholar 
  49. Smagorinsky J (1963) General circulation experiments with the primitive equations. Mon Weather Rev 91(3):99–164Article Google Scholar 
  50. Stocking J, Laforsch C, Sigl R, Reidenbach M (2018) The role of turbulent hydrodynamics and surface morphology on heat and mass transfer in corals. J R Soc Interface 15:20180448. https://doi.org/10.1098/rsif.2018.0448Article Google Scholar 
  51. Van Leer B (1977) Towards the ultimate conservative difference scheme III. Upstream-centered finite-difference schemes for ideal compressible flow. J Comput Phys 23(3):263–275. https://doi.org/10.1016/0021-9991(77)90094-8Article Google Scholar 
  52. Wells C, Pringle J, Stretch D (2021) Cold water temperature anomalies on the Sodwana reefs and their driving mechanisms. South Afr J Sci. https://doi.org/10.17159/sajs.2021/9304Article Google Scholar 
  53. Wyatt ASJ, Lowe RJ, Humphries S, Waite AM (2010) Particulate nutrient fluxes over a fringing coral reef: relevant scales of phytoplankton production and mechanisms of supply. Mar Ecol Prog Ser 405:113–130Article Google Scholar 
  54. Yao Y, He T, Deng Z, Chen L, Guo H (2019) Large eddy simulation modeling of tsunami-like solitary wave processes over fringing reefs. Nat Hazards Earth Syst Sci 19(6):1281–1295. https://doi.org/10.5194/nhess-19-1281-2019Article Google Scholar 
  55. Zhao Q, Tanimoto K (1998) Numerical simulation of breaking waves by large eddy simulation and vof method. Coastal Engineering Proceedings 1(26), 10.9753/icce.v26.%p, https://journals.tdl.org/icce/index.php/icce/article/view/5656

Text and image taken from Deoraj, et al. (2022), On the reef scale hydrodynamics at Sodwana Bay, South Africa. Preprint courtesy the authors.

Figure 14. Defects: (a) Unmelt defects(Scheme NO.4);(b) Pores defects(Scheme NO.1); (c); Spattering defect (Scheme NO.3); (d) Low overlapping rate defects(Scheme NO.5).

Molten pool structure, temperature and velocity
flow in selective laser melting AlCu5MnCdVA alloy

용융 풀 구조, 선택적 온도 및 속도 흐름 레이저 용융 AlCu5MnCdVA 합금

Pan Lu1 , Zhang Cheng-Lin2,6,Wang Liang3, Liu Tong4 and Liu Jiang-lin5
1 Aviation and Materials College, Anhui Technical College of Mechanical and Electrical Engineering, Wuhu Anhui 241000, People’s
Republic of China 2 School of Engineering Science, University of Science and Technology of China, Hefei Anhui 230026, People’s Republic of China 3 Anhui Top Additive Manufacturing Technology Co., Ltd., Wuhu Anhui 241300, People’s Republic of China 4 Anhui Chungu 3D Printing Institute of Intelligent Equipment and Industrial Technology, Anhui 241300, People’s Republic of China 5 School of Mechanical and Transportation Engineering, Taiyuan University of Technology, Taiyuan Shanxi 030024, People’s Republic of
China 6 Author to whom any correspondence should be addressed.
E-mail: ahjdpanlu@126.com, jiao__zg@126.com, ahjdjxx001@126.com,tongliu1988@126.com and liujianglin@tyut.edu.cn

Keywords

SLM, molten pool, AlCu5MnCdVA alloy, heat flow, velocity flow, numerical simulation

Abstract

선택적 레이저 용융(SLM)은 열 전달, 용융, 상전이, 기화 및 물질 전달을 포함하는 복잡한 동적 비평형 프로세스인 금속 적층 제조(MAM)에서 가장 유망한 기술 중 하나가 되었습니다. 용융 풀의 특성(구조, 온도 흐름 및 속도 흐름)은 SLM의 최종 성형 품질에 결정적인 영향을 미칩니다. 이 연구에서는 선택적 레이저 용융 AlCu5MnCdVA 합금의 용융 풀 구조, 온도 흐름 및 속도장을 연구하기 위해 수치 시뮬레이션과 실험을 모두 사용했습니다.

그 결과 용융풀의 구조는 다양한 형태(깊은 오목 구조, 이중 오목 구조, 평면 구조, 돌출 구조 및 이상적인 평면 구조)를 나타냈으며, 용융 풀의 크기는 약 132 μm × 107 μm × 50 μm였습니다. : 용융풀은 초기에는 여러 구동력에 의해 깊이 15μm의 깊은 오목형상이었으나, 성형 후기에는 장력구배에 의해 높이 10μm의 돌출형상이 되었다. 용융 풀 내부의 금속 흐름은 주로 레이저 충격력, 금속 액체 중력, 표면 장력 및 반동 압력에 의해 구동되었습니다.

AlCu5MnCdVA 합금의 경우, 금속 액체 응고 속도가 매우 빠르며(3.5 × 10-4 S), 가열 속도 및 냉각 속도는 각각 6.5 × 107 K S-1 및 1.6 × 106 K S-1 에 도달했습니다. 시각적 표준으로 표면 거칠기를 선택하고, 낮은 레이저 에너지 AlCu5MnCdVA 합금 최적 공정 매개변수 창을 수치 시뮬레이션으로 얻었습니다: 레이저 출력 250W, 부화 공간 0.11mm, 층 두께 0.03mm, 레이저 스캔 속도 1.5m s-1 .

또한, 실험 프린팅과 수치 시뮬레이션과 비교할 때, 용융 풀의 폭은 각각 약 205um 및 약 210um이었고, 인접한 두 용융 트랙 사이의 중첩은 모두 약 65um이었다. 결과는 수치 시뮬레이션 결과가 실험 인쇄 결과와 기본적으로 일치함을 보여 수치 시뮬레이션 모델의 정확성을 입증했습니다.

Selective Laser Melting (SLM) has become one of the most promising technologies in Metal Additive Manufacturing (MAM), which is a complex dynamic non-equilibrium process involving heat transfer, melting, phase transition, vaporization and mass transfer. The characteristics of the molten pool (structure, temperature flow and velocity flow) have a decisive influence on the final forming quality of SLM. In this study, both numerical simulation and experiments were employed to study molten pool structure, temperature flow and velocity field in Selective Laser Melting AlCu5MnCdVA alloy. The results showed the structure of molten pool showed different forms(deep-concave structure, double-concave structure, plane structure, protruding structure and ideal planar structure), and the size of the molten pool was approximately 132 μm × 107 μm × 50 μm: in the early stage, molten pool was in a state of deep-concave shape with a depth of 15 μm due to multiple driving forces, while a protruding shape with a height of 10 μm duo to tension gradient in the later stages of forming. The metal flow inside the molten pool was mainly driven by laser impact force, metal liquid gravity, surface tension and recoil pressure. For AlCu5MnCdVA alloy, metal liquid solidification speed was extremely fast(3.5 × 10−4 S), the heating rate and cooling rate reached 6.5 × 107 K S−1 and 1.6 × 106 K S−1 , respectively. Choosing surface roughness as a visual standard, low-laser energy AlCu5MnCdVA alloy optimum process parameters window was obtained by numerical simulation: laser power 250 W, hatching space 0.11 mm, layer thickness 0.03 mm, laser scanning velocity 1.5 m s−1 . In addition, compared with experimental printing and numerical simulation, the width of the molten pool was about 205 um and about 210 um, respectively, and overlapping between two adjacent molten tracks was all about 65 um. The results showed that the numerical simulation results were basically consistent with the experimental print results, which proved the correctness of the numerical simulation model.

Figure 1. AlCu5MnCdVA powder particle size distribution.
Figure 1. AlCu5MnCdVA powder particle size distribution.
Figure 2. AlCu5MnCdVA powder
Figure 2. AlCu5MnCdVA powder
Figure 3. Finite element model and calculation domains of SLM.
Figure 3. Finite element model and calculation domains of SLM.
Figure 4. SLM heat transfer process.
Figure 4. SLM heat transfer process.
Figure 14. Defects: (a) Unmelt defects(Scheme NO.4);(b) Pores defects(Scheme NO.1); (c); Spattering defect (Scheme NO.3); (d) Low
overlapping rate defects(Scheme NO.5).
Figure 17. Two-pass molten tracks overlapping for Scheme NO.2.
Figure 17. Two-pass molten tracks overlapping for Scheme NO.2.

References

[1] Cuiyun H 2008 Phase diagram determination and thermodynamic study of Al–Cu–Mn, Al–Cu–Si, Al–Mg–Ni and Ni–Ti–Si systems Central South University
[2] Zhanfei Z 2017 Study on theta phase segregation and room temperature properties of high strength cast Al–Cu–Mn alloy Lanzhou University of Technology
[3] Nie X et al 2018 Analysis of processing parameters and characteristics of selective laser melted high strength Al–Cu–Mg alloys: from single tracks to cubic samplesJ. Mater. Process. Technol. 256 69–77
[4] Shenping Y et al 2017 Laser absorptance measurement of commonly used metal materials in laser additive manufacturing technology Aviation Manufacturing Technology 12 23–9
[5] Wenqing W 2007 Relationship between cooling rate and grain size of AlCu5MnCdVA alloy Harbin University of Technology
[6] Majeed M, Vural M, Raja S and Bilal Naim Shaikh M 2019 Finite element analysis of thermal behavior in maraging steel during SLM process Optik 208 113–24
[7] Khairallah S A, Anderson A T, Rubenchik A and King W E 2016 Laser powder-bed fusion additive manufacturing: physics of complex melt flow and formation mechanisms of pores, spatter, and denudation zones Acta Mater. 108 36–45
[8] Bo C, Zhiyu X, Quanquan Z, Yuanbiao W, Liping W and Jin C 2020 Process optimization and microstructure and properties of SLM forming Cu6AlNiSnInCe imitation gold alloy Chin. J. Nonferr. Met. 30 372–82
[9] Li W 2012 Research on performance of metal parts formed by selective laser melting Huazhong University of Science and Technology
[10] Yu Q 2013 The influence of different laser heat sources on the surface shape of the molten pool in laser cladding Surf. Technol. 42 40–3

[11] Xianfeng J, Xiangchen M, Rongwei S, Xigen Y and Ming Y 2015 Research on the influence of material state change on temperature field
in SLM processing Applied Laser 35 155–9
[12] Körner C, Attar E and Heinl P 2011 Mesoscopic simulation of selective beam melting processesJ. Mater. Process. Technol. 211 978–87
[13] Yadroitsev I, Gusarov A, Yadroitsava I and Smurov I 2010 Single track formation in selective laser melting of metal powdersJ. Mater.
Process. Technol. 210 1624–31
[14] King W, Anderson A T, Ferencz R M, Hodge N E, Kamath C and Khairallah S A 2014 Overview of modelling and simulation of metal
powder bed fusion process at Lawrence Livermore National Laboratory Mater. Sci. Technol. 31 957–68
[15] Hussein A, Hao L, Yan C and Everson R 2013 Finite element simulation of the temperature and stress fields in single layers built
without-support in selective laser melting Materials & Design (1980–2015) 52 638–47
[16] Qiu C, Panwisawas C, Ward M, Basoalto H C, Brooks J W and Attallah M M 2015 On the role of melt flow into the surface structure and
porosity development during selective laser melting Acta Mater. 96 72–9
[17] Weihao Y, Hui C and Qingsong W 2020 Thermodynamic behavior of laser selective melting molten pool under the action of recoil
pressure Journal of Mechanical Engineering 56 213–9
[18] Weijuan Y 2019 Numerical simulation of melt pool temperature field and morphology evolution during laser selective melting process
Xi’an University of Technology
[19] Genwang W 2017 Research on the establishment of laser heat source model based on energy distribution and its simulation application
Harbin Institute of Technology
[20] FLOW-3D 2017 User Manual (USA: FLOW SCIENCE)
[21] Hirt C and Nichols B 1981 Volume of fluid (VOF) method for the dynamics of free boundariesJ. Comput. Phys. 39 201–25
[22] Hu Z, Zhang H, Zhu H, Xiao Z, Nie X and Zeng X 2019 Microstructure, mechanical properties and strengthening mechanisms of
AlCu5MnCdVA aluminum alloy fabricated by selective laser melting Materials Science and Engineering: A 759 154–66
[23] Ketai H, Liu Z and Lechang Y 2020 Simulation of temperature field, microstructure and mechanical properties of 316L stainless steel in
selected laser melting Progress in Laser and Optoelectronics 9 1–18
[24] Cao L 2020 Workpiece-scale numerical simulations of SLM molten pool dynamic behavior of 316L stainless steel Comput. Math. Appl.
4 22–34
[25] Dening Z, Yongping L, Tinglu H and Junyi S 2000 Numerical study of fluid flow and heat transfer in molten pool under the condition of
moving heat source J. Met. 4 387–90
[26] Chengyun C, Cui F and Wenlong Z 2018 The effect of Marangoni flow on the thermal behavior and melt flow behavior of laser cladding
Applied Laser 38 409–16
[27] Peiying B and Enhuai Y 2020 The effect of laser power on the morphology and residual stress of the molten pool of metal laser selective
melting Progress in Laser and Optoelectronics 7 1–12 http://kns.cnki.net/kcms/detail/31.1690.TN.20190717.0933.032.html
[28] Zhen L, Dongyun Z, Zhe F and Chengjie W 2017 Numerical simulation of the influence of overlap rate on the forming quality of
Inconel 718 alloy by selective laser melting processing Applied Laser 37 187–93
[29] Wei W, Qi L, Guang Y, Lanyun Q and Xiong X 2015 Numerical simulation of electromagnetic field, temperature field and flowfield of
laser melting pool under the action of electromagnetic stirring China Laser 42 48–55
[30] Hu Y, He X, Yu G and Zhao S 2016 Capillary convection in pulsed—butt welding of miscible dissimilar couple Proc. Inst. Mech. Eng.
Part C J. Mech. Eng. Sci. 231 2429–40
[31] Li R 2010 Research on the key basic problems of selective laser melting forming of metal powder Huazhong University of Science and
Technology
[32] Zijue T, Weiwei L, Zhaorui Y, Hao W and Hongchao Z 2019 Study on the shape evolution behavior of metal laser melting deposition
based on molten pool dynamic characteristicsJournal of Mechanical Engineering 55 39–47
[33] Pan L, Cheng-Lin Z, Hai-Yi L, Liang W and Tong L 2020 A new two-step selective laser remelting of 316L stainless steel: process,
density, surface roughness, mechanical properties, microstructure Mater. Res. Express 7 056503
[34] Pan L, Cheng-Lin Z, Hai-Yi L, Jiang H, Tong L and Liang W 2019 The influence and optimization of forming process parameters of
316L stainless steel prepared by laser melting on the density Forging Technology 44 103–9

Figure 1.| Physical models of the vertical drop, backdrop and stepped drop developed in the Technical University of Lisbon.

Numerical modelling of air-water flows in sewer drops

하수구 방울의 공기-물 흐름 수치 모델링

Paula Beceiro (corresponding author)
Maria do Céu Almeida
Hydraulic and Environment Department (DHA), National Laboratory for Civil Engineering, Avenida do Brasil 101, 1700-066 Lisbon, Portugal
E-mail: pbeceiro@lnec.pt
Jorge Matos
Department of Civil Engineering, Arquitecture and Geosources,
Technical University of Lisbon (IST), Avenida Rovisco Pais 1, 1049-001 Lisbon, Portugal

ABSTRACT

물 흐름에 용존 산소(DO)의 존재는 해로운 영향의 발생을 방지하는 데 유익한 것으로 인식되는 호기성 조건을 보장하는 중요한 요소입니다.

하수도 시스템에서 흐르는 폐수에 DO를 통합하는 것은 공기-액체 경계면 또는 방울이나 접합부와 같은 특이점의 존재로 인해 혼입된 공기를 통한 연속 재방출의 영향을 정량화하기 위해 광범위하게 조사된 프로세스입니다. 공기 혼입 및 후속 환기를 향상시키기 위한 하수구 드롭의 위치는 하수구의 호기성 조건을 촉진하는 효과적인 방법입니다.

본 논문에서는 수직 낙하, 배경 및 계단식 낙하를 CFD(전산유체역학) 코드 FLOW-3D®를 사용하여 모델링하여 이러한 유형의 구조물의 존재로 인해 발생하는 난류로 인한 공기-물 흐름을 평가했습니다. 이용 가능한 실험적 연구에 기초한 수력학적 변수의 평가와 공기 혼입의 분석이 수행되었습니다.

이러한 구조물에 대한 CFD 모델의 결과는 Soares(2003), Afonso(2004) 및 Azevedo(2006)가 개발한 해당 물리적 모델에서 얻은 방류, 압력 헤드 및 수심의 측정을 사용하여 검증되었습니다.

유압 거동에 대해 매우 잘 맞았습니다. 수치 모델을 검증한 후 공기 연행 분석을 수행했습니다.

The presence of dissolved oxygen (DO) in water flows is an important factor to ensure the aerobic conditions recognised as beneficial to prevent the occurrence of detrimental effects. The incorporation of DO in wastewater flowing in sewer systems is a process widely investigated in order to quantify the effect of continuous reaeration through the air-liquid interface or air entrained due the presence of singularities such as drops or junctions. The location of sewer drops to enhance air entrainment and subsequently reaeration is an effective practice to promote aerobic conditions in sewers. In the present paper, vertical drops, backdrops and stepped drop was modelled using the computational fluid dynamics (CFD) code FLOW-3D® to evaluate the air-water flows due to the turbulence induced by the presence of this type of structures. The assessment of the hydraulic variables and an analysis of the air entrainment based in the available experimental studies were carried out. The results of the CFD models for these structures were validated using measurements of discharge, pressure head and water depth obtained in the corresponding physical models developed by Soares (2003), Afonso (2004) and Azevedo (2006). A very good fit was obtained for the hydraulic behaviour. After validation of numerical models, analysis of the air entrainment was carried out.

Key words | air entrainment, computational fluid dynamics (CFD), sewer drops

Figure 1.| Physical models of the vertical drop, backdrop and stepped drop developed in the Technical University of Lisbon.
Figure 1.| Physical models of the vertical drop, backdrop and stepped drop developed in the Technical University of Lisbon.
Figure 3. Comparison between the experimental and numerical pressure head along of the invert of the outlet pipe.
Figure 3. Comparison between the experimental and numerical pressure head along of the invert of the outlet pipe.
Figure 4. Average void fraction along the longitudinal axis of the outlet pipe for the lower discharges in the vertical drop and backdrop.
Figure 4. Average void fraction along the longitudinal axis of the outlet pipe for the lower discharges in the vertical drop and backdrop.

REFERENCES

Afonso, J. Dissipação de energia e rearejamento em quedas em colectores. M.Sc. Thesis, UTL/IST, Lisboa, Portugal.
Almeida, M. C., Butler, D. & Matos, J. S. Reaeration by sewer drops. In: 8th Int. Conf. on Urban Storm Drainage, Sydney, Australia.
Azevedo, R. I. Transferência de oxigénio em quedas guiadas em colectores. M.Sc. Thesis, IST, Lisboa, Portugal.
Beceiro, P., Almeida, M. C. & Matos, J. Numerical Modelling of air-water flows in a vertical drop and a backdrop. In: 3rd IAHR Europe Congress, Porto, Portugal.
Bombardelli, F. A., Meireles, I. & Matos, J. S. Laboratory measurements and multi-block numerical simulations of the mean flow and turbulence in the non-aerated skimming flow region of step stepped spillways. Environ. Fluid Mech. 11 (3), 263–288.
Brethour, J. M. & Hirt, C. W. Drift Model for TwoComponent Flows. Flow Science, Inc., Los Alamos, NM, USA.
Chamani, M. R. Jet Flow on Stepped Spillways and Drops. M.Sc. Thesis, University of Alberta, Alberta, Canada.
Chanson, H. Air Bubble Entrainment in Free-Surface Turbulent Shear Flow. Academic Press Inc., California, USA.
Chanson, H. Air bubble entrainment in open channels: flow structure and bubble size distribution. Int. J. Multiphase 23 (1), 193–203.
Chanson, H. Hydraulics of aerated flows: qui pro quo? Journal of Hydraulic Research 51 (3), 223–243.
Dufresne, M., Vazques, J., Terfous, A., Ghenaim, A. & Poulet, J. Experimental investigation and CFD modelling of flow, sedimentation, and solids separation in a combined sewer detention tank. Computer and Fluids 38, 1042–1049.
Durve, A. P. & Patwardhan, A. W. Numerical and experimental investigation of onset of gas entrainment phenomenon. Chemical Engineering Science 73, 140–150.
Felder, S. & Chanson, H. Air–water flows and free-surface profiles on a non-uniform stepped chute. Journal of Hydraulic Research 52 (2), 253–263.
Flow Science FLOW-3D User’s Manuals Version 10.0. Vol.1/2. Flow Science Inc., Los Alamos, NM, USA.
Granata, F., Marinis, G., Gargano, R. & Hager, W. H. Energy loss in circular drop manholes. In: 33rd IAHR Congress: Water Engineering for Sustainable Environment, British
Columbia, Vancouver, Canada. Hirt, C. W. Modeling Turbulent Entrainment of air at A Free Surface. Flow Science Inc., Los Alamos, NM, USA.
Hirt, C. W. & Nichols, B. D. Volume of fluid (VOF) method for the dynamics of free boundaries. Journal of Computational Physics 39, 201–225.
Hirt, C. W. & Sicilian, J. M. A porosity technique for the definition of obstacles in rectangular cell meshes. In: Proc. 4th Int, Conf. Ship Hydro., National Academy of Science, Washington, DC, USA.
Isfahani, A. H. G. & Brethour, J. On the Implementation of Two-Equation Turbulence Models in FLOW-3D. Flow Science Inc., Los Alamos, NM, USA.
Kouyi, G. L., Bret, P., Didier, J. M., Chocat, B. & Billat, C. The use of CFD modelling to optimise measurement of overflow rates in a downstream-controlled dual-overflow structure. Water Science and Technology 64 (2), 521–527.
Lopes, P., Leandro, J., Carvalho, R. F., Páscoa, P. & Martins, R. Numerical and experimental investigation of a gully under surcharge conditions. Urban Water Journal 12 (6), 468–476.
Martins, R., Leandro, J. & Carvalho, R. F. Characterization of the hydraulic performance of a gully under drainage conditions. Water Science and Technology 69 (12), 2423–2430.
Matias, N., Nielsel, A. H., Vollertsen, J., Ferreira, F. & Matos, J. S. Reaeration and hydrogen sulfide release at drop structures. In: 8th International Conference on Sewer Processes and Networks (SPN8), Rotterdam, Netherlands.
Matos, J. S. & Sousa, E. R. Prediction of dissolved oxygen concentration along sanitary sewers. Water Science and Technology 34 (5–6), 525–532.
Mignot, E., Bonakdari, H., Knothe, P., Lipeme Kouyi, G., Bessette, A., Rivière, N. & Bertrand-Krajewski, J. L. Experiments and 3D simulations of flow structures in junctions and of their influence on location of flowmeters. In: 12th International Conference on Urban Drainage, Porto Alegre, Brazil.
Ozmen-Cagatay, H. & Kocaman, S. Dam-break flow in the presence of obstacle: experiment and CFD Simulation. Engineering Applications of Computational Fluid Mechanics 5 (4), 541–552.
Shojaee Fard, M. H. & Boyaghchi, F. A. Studies of the influence of various blade outlet angles in a centrifugal pump when handling viscous fluids. American Journal of Applied Sciences 4 (9), 718–724.
Soares, A. Rearejamento em Quedas em Colectores de Águas Residuais. M.Sc. Thesis, FCTUC, Coimbra, Portugal.
Sousa, C. M. & Lopes, R. R. Hidráulica e rearejamento em quedas verticais em colectores. Estudo Experimental. Research Report, UTL/IST, Lisboa, Portugal.
Sousa, V., Meireles, I., Matos, J. & Almeida, M. C. Numerical modelling of air-water flow in a vertical drop manhole. In: 7th International Conference on Sewer Processes and Networks (SPN7), Shefield, UK.
Stovin, V., Guymer, I. & Lau, S. D. Approaches to validating a 3D CFD manhole model. In: 11th International Conference on Urban Drainage, Edinburgh, Scotland, UK.
Tota, P. V. Turbulent Flow Over A Backward-Facing Step Using the RNG Model. Flow Science Inc., Los Alamos, NM, USA.
Valero, D. & García-Bartual, R. Calibration of an air entrainment model for CFD spillway applications. In: Advances in Hydroinformatics. Springer, Singapore, pp. 571–582.
Versteeg, H. K. & Malalasekera, W. An Introduction to Computational Fluid Dynamics. The Finite Volume Method. Longman Group limited, England.
Yang, Y., Yang, J., Zuo, J., Li, Y., He, S., Yang, X. & Zhang, K. Study on two operating conditions of a full-scale oxidation ditch for optimization of energy consumption and effluent quality by using CFD model. Water Research 45 (11), 3439–3452.
Zhai, A. J., Zhang, Z., Zhang, W. & Chen, Q. Y. Evaluation of various turbulence models in predicting airflow and turbulence in enclosed environments by CFD: part 1— summary of prevalent Turbulence models. HVAC&R Research 13 (6), 853–870.
Zhao, C., Zhu, D. Z. & Rajaratnam, N. Computational and experimental study of surcharged flow at a 90W combining sewer junction. Journal of Hydraulic Engineering 134 (6), 688–700.

Development of macro-defect-free PBF-EB-processed Ti–6Al–4V alloys with superior plasticity using PREP-synthesized powder and machine learning-assisted process optimization

Development of macro-defect-free PBF-EB-processed Ti–6Al–4V alloys with superior plasticity using PREP-synthesized powder and machine learning-assisted process optimization

Yunwei GuiabKenta Aoyagib Akihiko Chibab
aDepartment of Materials Processing, Graduate School of Engineering, Tohoku University, 6-6 Aramaki Aza Aoba, Aoba-ku, Sendai, 980-8579, Japan
bInstitute for Materials Research, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai, 980-8577, Japan

Received 14 October 2022, Revised 23 December 2022, Accepted 3 January 2023, Available online 5 January 2023.Show lessAdd to MendeleyShareCite

https://doi.org/10.1016/j.msea.2023.144595Get rights and content

Abstract

The elimination of internal macro-defects is a key issue in Ti–6Al–4V alloys fabricated via powder bed fusion using electron beams (PBF-EB), wherein internal macro-defects mainly originate from the virgin powder and inappropriate printing parameters. This study compares different types powders by combining support vector machine techniques to determine the most suitable powder for PBF-EB and to predict the processing window for the printing parameters without internal macro-defects. The results show that powders fabricated via plasma rotating electrode process have the best sphericity, flowability, and minimal porosity and are most suitable for printing. Surface roughness criterion was also applied to determine the quality of the even surfaces, and support vector machine was used to construct processing maps capable of predicting a wide range of four-dimensional printing parameters to obtain macro-defect-free samples, offering the possibility of subsequent development of Ti–6Al–4V alloys with excellent properties. The macro-defect-free samples exhibited good elongation, with the best overall mechanical properties being the ultimate tensile strength and elongation of 934.7 MPa and 24.3%, respectively. The elongation of the three macro-defect-free samples was much higher than that previously reported for additively manufactured Ti–6Al–4V alloys. The high elongation of the samples in this work is mainly attributed to the elimination of internal macro-defects.

Introduction

Additive manufacturing (AM) technologies can rapidly manufacture complex or custom parts, reducing process steps and saving manufacturing time [[1], [2], [3], [4]], and are widely used in the aerospace, automotive, and other precision industries [5,6]. Powder bed fusion using an electron beam (PBF-EB) is an additive manufacturing method that uses a high-energy electron beam to melt metal powders layer by layer to produce parts. In contrast to selective laser melting, PBF-EB involves the preparation of samples in a high vacuum environment, which effectively prevents the introduction of impurities such as O and N. It also involves a preheating process for the print substrate and powder, which reduces residual thermal stress on the sample and subsequent heat treatment processes [[2], [3], [4],7]. Due to these features and advantages, PBF-EB technology is a very important AM technology with great potential in metallic materials. Moreover, PBF-EB is the ideal AM technology for the manufacture of complex components made of many alloys, such as titanium alloys, nickel-based superalloys, aluminum alloys and stainless steels [[2], [3], [4],8].

Ti–6Al–4V alloy is one of the prevalent commercial titanium alloys possessing high specific strength, excellent mechanical properties, excellent corrosion resistance, and good biocompatibility [9,10]. It is widely used in applications requiring low density and excellent corrosion resistance, such as the aerospace industry and biomechanical applications [11,12]. The mechanical properties of PBF-EB-processed Ti–6Al–4V alloys are superior to those fabricated by casting or forging, because the rapid cooling rate in PBF-EB results in finer grains [[12], [13], [14], [15], [16], [17], [18]]. However, the PBF-EB-fabricated parts often include internal macro-defects, which compromises their mechanical properties [[19], [20], [21], [22]]. This study focused on the elimination of macro-defects, such as porosity, lack of fusion, incomplete penetration and unmelted powders, which distinguishes them from micro-defects such as vacancies, dislocations, grain boundaries and secondary phases, etc. Large-sized fusion defects cause a severe reduction in mechanical strength. Smaller defects, such as pores and cracks, lead to the initiation of fatigue cracking and rapidly accelerate the cracking process [23]. The issue of internal macro-defects must be addressed to expand the application of the PBF-EB technology. The main studies for controlling internal macro-defects are online monitoring of defects, remelting and hot isostatic pressing (HIP). The literatures [24,25] report the use of infrared imaging or other imaging techniques to identify defects, but the monitoring of smaller sized defects is still not adequate. And in some cases remelting does not reduce the internal macro-defects of the part, but instead causes coarsening of the macrostructure and volatilization of some metal elements [23]. The HIP treatment does not completely eliminate the internal macro-defects, the original defect location may still act as a point of origin of the crack, and the subsequent treatment will consume more time and economic costs [23]. Therefore, optimizing suitable printing parameters to avoid internal macro-defects in printed parts at source is of great industrial value and research significance, and is an urgent issue in PBF-EB related technology.

There are two causes of internal macro-defects in the AM process: gas pores trapped in the virgin powder and the inappropriate printing parameters [7,23]. Gui et al. [26] classify internal macro-defects during PBF-EB process according to their shape, such as spherical defects, elongated shape defects, flat shape defects and other irregular shape defects. Of these, spherical defects mainly originate from raw material powders. Other shape defects mainly originate from lack of fusion or unmelted powders caused by unsuitable printing parameters, etc. The PBF-EB process requires powders with good flowability, and spherical powders are typically chosen as raw materials. The prevalent techniques for the fabrication of pre-alloyed powders are gas atomization (GA), plasma atomization (PA), and the plasma rotating electrode process (PREP) [27,28]. These methods yield powders with different characteristics that affect the subsequent fabrication. The selection of a suitable powder for PBF-EB is particularly important to produce Ti–6Al–4V alloys without internal macro-defects. The need to optimize several printing parameters such as beam current, scan speed, line offset, and focus offset make it difficult to eliminate internal macro-defects that occur during printing [23]. Most of the studies [11,12,22,[29], [30], [31], [32], [33]] on the optimization of AM processes for Ti–6Al–4V alloys have focused on samples with a limited set of parameters (e.g., power–scan speed) and do not allow for the guidance and development of unknown process windows for macro-defect-free samples. In addition, process optimization remains a time-consuming problem, with the traditional ‘trial and error’ method demanding considerable time and economic costs. The development of a simple and efficient method to predict the processing window for alloys without internal macro-defects is a key issue. In recent years, machine learning techniques have increasingly been used in the field of additive manufacturing and materials development [[34], [35], [36], [37]]. Aoyagi et al. [38] recently proposed a novel and efficient method based on a support vector machine (SVM) to optimize the two-dimensional process parameters (current and scan speed) and obtain PBF-EB-processed CoCr alloys without internal macro-defects. The method is one of the potential approaches toward effective optimization of more than two process parameters and makes it possible for the machine learning techniques to accelerate the development of alloys without internal macro-defects.

Herein, we focus on the elimination of internal macro-defects, such as pores, lack of fusion, etc., caused by raw powders and printing parameters. The Ti–6Al–4V powders produced by three different methods were compared, and the powder with the best sphericity, flowability, and minimal porosity was selected as the feedstock for subsequent printing. The relationship between the surface roughness and internal macro-defects in the Ti–6Al–4V components was also investigated. The combination of SVM and surface roughness indices (Sdr) predicted a wider four-dimensional processing window for obtaining Ti–6Al–4V alloys without internal macro-defects. Finally, we investigated the tensile properties of Ti–6Al–4V alloys at room temperature with different printing parameters, as well as the corresponding microstructures and fracture types.

Section snippets

Starting materials

Three types of Ti–6Al–4V alloy powders, produced by GA, PA, and PREP, were compared. The particle size distribution of the powders was determined using a laser particle size analyzer (LS230, Beckman Coulter, USA), and the flowability was measured using a Hall flowmeter (JIS-Z2502, Tsutsui Scientific Instruments Co., Ltd., Japan), according to the ASTM B213 standard. The powder morphology and internal macro-defects were determined using scanning electron microscopy (SEM, JEOL JCM-6000) and X-ray 

Comparison of the characteristics of GA, PA, and PREP Ti–6Al–4V powders

The particle size distributions (PSDs) and flowability of the three types of Ti–6Al–4V alloy powders produced by GA, PA, and PREP are shown in Fig. 2. Although the average particle sizes are similar (89.4 μm for GA, 82.5 μm for PA, and 86.1μm for PREP), the particle size range is different for the three types of powder (6.2–174.8 μm for GA, 27.3–139.2 μm for PA, and 39.4–133.9 μm for PREP). The flowability of the GA, PA, and PREP powders was 30.25 ± 0.98, 26.54 ± 0.37, and 25.03 ± 0.22 (s/50

Conclusions

The characteristics of the three types of Ti–6Al–4V alloy powders produced via GA, PA, and PREP were compared. The PREP powder with the best sphericity, flowability, and low porosity was found to be the most favorable powder for subsequent printing of Ti–6Al–4V alloys without internal macro-defects. The quantitative criterion of Sdr <0.015 for even surfaces was also found to be applicable to Ti–6Al–4V alloys. The process maps of Ti–6Al–4V alloys include two regions, high beam current/scan speed 

Uncited references

[55]; [56]; [57]; [58]; [59]; [60]; [61]; [62]; [63]; [64]; [65].

CRediT authorship contribution statement

Yunwei Gui: Writing – original draft, Visualization, Validation, Investigation. Kenta Aoyagi: Writing – review & editing, Supervision, Resources, Methodology, Funding acquisition, Conceptualization. Akihiko Chiba: Supervision, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was based on the results obtained from project JPNP19007, commissioned by the New Energy and Industrial Technology Development Organization (NEDO). This work was also supported by JSPS KAKENHI (Proposal No. 21K03801) and the Inter-University Cooperative Research Program (Proposal nos. 18G0418, 19G0411, and 20G0418) of the Cooperative Research and Development Center for Advanced Materials, Institute for Materials Research, Tohoku University. It was also supported by the Council for

References (65)

View more references

Cited by (0)

Recommended articles (6)

Fig. 2 Modeling of bubble point test apparatus (left) and computational grid (righ

Flow-3d를 이용한 표면장력 탱크용메시스크린모델링

Modeling of Mesh Screen for Use in Surface TensionTankUsing Flow-3d Software

Hyuntak Kim․ Sang Hyuk Lim․Hosung Yoon․Jeong-Bae Park*․Sejin Kwon

ABSTRACT

Mesh screen modeling and liquid propellant discharge simulation of surface tension tank wereperformed using commercial CFD software Flow-3d. 350 × 2600, 400 × 3000 and 510 × 3600 DTW mesh screen were modeled using macroscopic porous media model. Porosity, capillary pressure, and drag
coefficient were assigned for each mesh screen model, and bubble point simulations were performed. The
mesh screen model was validated with the experimental data. Based on the screen modeling, liquidpropellant discharge simulation from PMD tank was performed. NTO was assigned as the liquidpropellant, and void was set to flow into the tank inlet to achieve an initial volume flowrate of
liquid propellant in 3 × 10-3 g acceleration condition. The intial flow pressure drop through the meshscreen was approximately 270 Pa, and the pressure drop increased with time. Liquid propellant
discharge was sustained until the flow pressure drop reached approximately 630 Pa, which was near
the estimated bubble point value of the screen model.

초 록

상용 CFD 프로그램 Flow-3d를 활용하여, 표면 장력 탱크 적용을 위한 메시 스크린의 모델링 및 추진제 배출 해석을 수행하였다. Flow-3d 내 거시적 다공성 매체 모델을 사용하였으며, 350 × 2600, 400× 3000, 510 × 3600 DTW 메시 스크린에 대한 공극률, 모세관압, 항력계수를 스크린 모델에 대입 후, 기포점 측정 시뮬레이션을 수행하였다.

시뮬레이션 결과를 실험 데이터와 비교하였으며, 메시 스크린 모델링의 적절성을 검증하였다. 이를 기반으로 스크린 모델을 포함한 PMD 구조체에 대한 추진제 배출 해석을 수행하였다. 추진제는 액상의 NTO를 가정하였으며, 3 × 10-3 g 가속 조건에서 초기 유량을만족하도록 void를 유입시켰다. 메시 스크린을 통한 차압은 초기 약 270 Pa에서 시간에 따라 증가하였으며, 스크린 모델의 예상 기포점과 유사한 630 Pa에 이르기까지 액상 추진제 배출을 지속하였다.

Key Words

Surface Tension Tank(표면장력 탱크), Propellant Management Device(추진제 관리 장치),
Mesh Screen(메시 스크린), Porous Media Model(다공성 매체 모델), Bubble Point(기포점)

서론

    우주비행체를 미소 중력 조건 내에서 운용하 는 경우, 가압 기체가 액상의 추진제와 혼합되어 엔진으로 공급될 우려가 있으므로 이를 방지하 기 위한 탱크의 설계가 필요하다.

    다이어프램 (Diaphragm), 피스톤(Piston) 등 다양한 장치들 이 활용되고 있으며, 이 중 표면 장력 탱크는 내 부의 메시 스크린(Mesh screen), 베인(Vane) 등 의 구조체에서 추진제의 표면장력을 활용함으로 써 액상 추진제의 이송 및 배출을 유도하는 방 식이다.

    표면 장력 탱크는 구동부가 없는 구조로 신뢰성이 높고, 전 부분을 티타늄 등의 금속 재 질로 구성함으로써 부식성 추진제의 사용 조건 에서도 장기 운용이 가능한 장점이 있다. 위에서 언급한 메시 스크린(Mesh screen)은 수 십 마이크로미터 두께의 금속 와이어를 직조한 다공성 재질로 표면 장력 탱크의 핵심 구성 요소 중 하나이다.

    미세 공극 상 추진제의 표면장력에 의해 기체와 액체 간 계면을 일정 차압 내에서 유지시킬 수 있다. 이러한 성질로 인해 일정 조 건에서 가압 기체가 메시 스크린을 통과하지 못 하게 되고, 스크린을 탱크 유로에 설치함으로써 액상의 추진제 배출을 유도할 수 있다.

    메시 스크린이 가압 기체를 통과시키기 직전 의 기체-액체 계면에 형성되는 최대 차압을 기포 점 (Bubble point) 이라 칭하며, 메시 스크린의 주 요 성능 지표 중 하나이다. IPA, 물, LH2, LCH4 등 다양한 기준 유체 및 추진제, 다양한 메시 스 크린 사양에 대해 기포점 측정 관련 실험적 연 구가 이루어져 왔다 [1-3].

    위 메시 스크린을 포함하여 표면 장력 탱크 내 액상의 추진제 배출을 유도하는 구조물 일체 를 PMD(Propellant management device)라 칭하 며, 갤러리(Gallery), 베인(Vane), 스펀지(Sponge), 트랩(Trap) 등 여러 종류의 구조물에 대해 각종 형상 변수를 내포한다[4, 5].

    따라서 다양한 파라미터를 고려한 실험적 연구는 제약이 따를 수 있으며, 베인 등 상대적으로 작은 미소 중력 조건에서 개방형 유로를 활용하는 경우 지상 추진제 배출 실험이 불가능하다[6]. 그러므로 CFD를 통한 표면장력 탱크 추진제 배출 해석은 다양한 작동 조건 및 PMD 형상 변수에 따른 추진제 거동을 이해하고, 탱크를 설계하는 데 유용하게 활용될 수 있다.

    상기 추진제 배출 해석을 수행하기 위해서는 핵심 요소 중 하나인 메시 스크린에 대한 모델링이 필수적이다. Chato, McQuillen 등은 상용 CFD 프로그램인 Fluent를 통해, 갤러리 내 유동 시뮬레이션을 수행하였으며, 이 때 메시 스크린에 ‘porous jump’ 경계 조건을 적용함으로써 액상의 추진제가 스크린을 통과할 때 생기는 압력 강하를 모델링하였다[7, 8].

    그러나 앞서 언급한 메시 스크린의 기포점 특성을 모델링한 사례는 찾아보기 힘들다. 이는 스크린을 활용하는 표면 장력 탱크 내 액상 추진제 배출 현상을 해석적으로 구현하기 위해 반드시 필요한 부분이다. 본 연구에서는 자유표면 해석에 상대적으로 강점을 지닌 상용 CFD 프로그램 Flow-3d를 사용하여, 메시 스크린을 모델링하였다.

    거시적 다공성 매체 모델(Macroscopic porous mediamodel)을 활용하여 메시 스크린 모델 영역에 공극률(Porosity), 모세관압(Capillary pressure), 항력 계수(Drag coefficient)를 지정하고, 이를 기반으로 기포점 측정 시뮬레이션을 수행, 해석 결과와 실험 데이터 간 비교 및 검증을 수행하였다.

    이를 기반으로 메시 스크린 및 PMD구조체를 포함한 탱크의 추진제 배출 해석을 수행하고, 기포점 특성의 반영 여부를 확인하였다.

    Fig. 1 Real geometry-based mesh screen model (left)
and mesh screen model based on macroscopic
porous media model in Flow-3d (righ
    Fig. 1 Real geometry-based mesh screen model (left) and mesh screen model based on macroscopic porous media model in Flow-3d (righ
    Fig. 2 Modeling of bubble point test apparatus (left)
and computational grid (righ
    Fig. 2 Modeling of bubble point test apparatus (left) and computational grid (righ)
    Fig. 3 Modeling of sump in a tank (left) and lower part
of the sump structure (right)
    Fig. 3 Modeling of sump in a tank (left) and lower part of the sump structure (right)

    참 고 문 헌

    1. David J. C and Maureen T. K, ScreenChannel Liquid Aquisition Devices for Cryogenic Propellants” NASA-TM-2005- 213638, 2005
    2. Hartwig, J., Mann, J. A. Jr., Darr, S. R., “Parametric Analysis of the LiquidHydrogen and Nitrogen Bubble Point Pressure for Cryogenic Liquid AcquisitionDevices”, Cryogenics, Vol. 63, 2014, pp. 25-36
    3. Jurns, J. M., McQuillen, J. B.,BubblePoint Measurement with Liquid Methane of a Screen Capillary Liquid AcquisitionDevice”, NASA-TM-2009-215496, 2009
    4. Jaekle, D. E. Jr., “Propellant Management Device: Conceptual Design and Analysis: Galleries”, AIAA 29th Joint PropulsionConference, AIAA-97-2811, 1997
    5. Jaekle, D. E. Jr., “Propellant Management Device: Conceptual Design and Analysis: Traps and Troughs”, AIAA 31th Joint Propulsion Conference, AIAA-95-2531, 1995
    6. Yu, A., Ji, B., Zhuang, B. T., Hu, Q., Luo, X. W., Xu, H. Y., “Flow Analysis inaVane-type Surface Tension Propellant Tank”, IOP Conference Series: MaterialsScience and Engineering, Vol. 52, No. 7, – 990 – 2013, Article number: 072018
    7. Chato, D. J., McQuillen, J. B., Motil, B. J., Chao, D. F., Zhang, N., CFD simulation of Pressure Drops in Liquid Acquisition Device Channel with Sub-Cooled Oxygen”, World Academy of Science, Engineering and Technology, Vol. 3, 2009, pp. 144-149
    8. McQuillen, J. B., Chao, D. F., Hall, N. R., Motil, B. J., Zhang, N., CFD simulation of Flow in Capillary Flow Liquid Acquisition Device Channel”, World Academy of Science, Engineering and Technology, Vol. 6, 2012, pp. 640-646
    9. Hartwig, J., Chato, D., McQuillen, J.,  Screen Channel LAD Bubble Point Tests in Liquid Hydrogen”, International Journal of Hydrogen Energy, Vol. 39, No. 2, 2014, pp. 853-861
    10. Fischer, A., Gerstmann, J., “Flow Resistance of Metallic Screens in Liquid, Gaseous and Cryogenic Flow”, 5th European Conferencefor Aeronautics and Space Sciences, Munich, Germany, 2013
    11. Fries, N., Odic, K., Dreyer, M., Wickingof Perfectly Wetting Liquids into a MetallicMesh”, 2nd International Conference onPorous Media and its Applications inScience and Engineering, 2007
    12. Seo, M, K., Kim, D, H., Seo, C, W., Lee, S, Y., Jang, S, P., Koo, J., “Experimental Study of Pressure Drop in CompressibleFluid through Porous Media”, Transactionsof the Korean Society of Mechanical Engineers – B, Vol. 37, No. 8, pp. 759-765, 2013.
    13. Hartwig, J., Mann, J. A., “Bubble Point Pressures of Binary Methanol/Water Mixtures in Fine-Mesh Screens”, AlChEJournal, Vol. 60, No. 2, 2014, pp. 730-739
    Serife Yurdagul Kumcu−2−KSCE Journal of Civil Engineeringthe use of CFD for the assessment of a design, as well as screeningand optimizing of hydraulic structures and cofferdam layouts. Theyconclude that CFD has been successful in optimizing the finalconceptual configuration for the hydraulics design of the project,but recommend that physical modeling still be used as a finalconfirmation.This paper provides experimental studies performed on Kav akDam and analyses the stability of spillway design by usingFLOW-3D model. It compares the hydraulic model tests withFLOW-3D simulation results and gives information on howaccurately a commercially available Computational Fluid Dynamic(CFD) model can predict the spillway discharge capacity andpressure distribution along the spillway bottom surface. 2. Physical ModelA 1/50-scaled undistorted physical model of the Kavsak Damspillway and stilling basin was built and tested at the HydraulicModel Laboratory of State Hydraulic Works of Turkey (DSI).The model was constructed of plexiglas and was fabricated toconform to the distinctive shape of an ogee crest. The spillwayhas 45.8 m in width and 57 m long with a bottom slope of 125%.The length of the stilling basin is about 90 m. During model tests,flow velocities were measured with an ultrasonic flow meter.Pressures on the spillway were measured using a piezometerssçTable 1. Upstream and Downstream Operating Conditions of theKavsak DamRun Upstream reservoir elevation (m)Downstream tailwater elevation (m)1 306.55 168.002 311.35 174.503 314.00 178.904 316.50 182.55Fig. 1. (a) Original Project Design and Final Project Design after Experimental Investigations and Flow Measurement Sections at theApproach, (b) Top View Experimentally Modified Approach in the Laboratory, (c) Side View of the Experimentally Modified Approachin the Laboratory

    Investigation of flow over spillway modeling and comparison between experimental data and CFD analysis

    여수로 모델링 및 실험 데이터와 CFD 해석의 비교에 대한 조사

    DOI:10.1007/s12205-016-1257-z

    Authors:

    Serife Yurdagul Kumcu at Necmettin Erbakan Üniversitesi

    Serife Yurdagul Kumcu

    Abstract and Figures

    As a part of design process for hydro-electric generating stations, hydraulic engineers typically conduct some form of model testing. The desired outcome from the testing can vary considerably depending on the specific situation, but often characteristics such as velocity patterns, discharge rating curves, water surface profiles, and pressures at various locations are measured. Due to recent advances in computational power and numerical techniques, it is now also possible to obtain much of this information through numerical modeling. In this paper, hydraulic characteristics of Kavsak Dam and Hydroelectric Power Plant (HEPP), which are under construction and built for producing energy in Turkey, were investigated experimentally by physical model studies. The 1/50-scaled physical model was used in conducting experiments. Flow depth, discharge and pressure data were recorded for different flow conditions. Serious modification was made on the original project with the experimental study. In order to evaluate the capability of the computational fluid dynamics on modeling spillway flow a comparative study was made by using results obtained from physical modeling and Computational Fluid Dynamics (CFD) simulation. A commercially available CFD program, which solves the Reynolds-averaged Navier-Stokes (RANS) equations, was used to model the numerical model setup by defining cells where the flow is partially or completely restricted in the computational space. Discharge rating curves, velocity patterns and pressures were used to compare the results of the physical model and the numerical model. It was shown that there is reasonably good agreement between the physical and numerical models in flow characteristics.

    수력 발전소 설계 프로세스의 일부로 수력 엔지니어는 일반적으로 어떤 형태의 모델 테스트를 수행합니다. 테스트에서 원하는 결과는 특정 상황에 따라 상당히 다를 수 있지만 속도 패턴, 방전 등급 곡선, 수면 프로파일 및 다양한 위치에서의 압력과 같은 특성이 측정되는 경우가 많습니다. 최근 계산 능력과 수치 기법의 발전으로 인해 이제는 수치 모델링을 통해 이러한 정보의 대부분을 얻을 수도 있습니다.

    본 논문에서는 터키에서 에너지 생산을 위해 건설 중인 Kavsak 댐과 수력발전소(HEPP)의 수력학적 특성을 물리적 모델 연구를 통해 실험적으로 조사하였다. 1/50 스케일의 물리적 모델이 실험 수행에 사용되었습니다. 다양한 흐름 조건에 대해 흐름 깊이, 배출 및 압력 데이터가 기록되었습니다. 실험 연구를 통해 원래 프로젝트에 대대적인 수정이 이루어졌습니다.

    배수로 흐름 모델링에 대한 전산유체역학의 능력을 평가하기 위해 물리적 모델링과 전산유체역학(CFD) 시뮬레이션 결과를 이용하여 비교 연구를 수행하였습니다. RANS(Reynolds-averaged Navier-Stokes) 방정식을 푸는 상업적으로 이용 가능한 CFD 프로그램은 흐름이 계산 공간에서 부분적으로 또는 완전히 제한되는 셀을 정의하여 수치 모델 설정을 모델링하는 데 사용되었습니다.

    물리적 모델과 수치 모델의 결과를 비교하기 위해 배출 등급 곡선, 속도 패턴 및 압력을 사용했습니다. 유동 특성에서 물리적 모델과 수치 모델 간에 상당히 좋은 일치가 있는 것으로 나타났습니다.

    Serife Yurdagul Kumcu−2−KSCE Journal of Civil Engineeringthe use of CFD for the assessment of a design, as well as screeningand optimizing of hydraulic structures and cofferdam layouts. Theyconclude that CFD has been successful in optimizing the finalconceptual configuration for the hydraulics design of the project,but recommend that physical modeling still be used as a finalconfirmation.This paper provides experimental studies performed on Kav akDam and analyses the stability of spillway design by usingFLOW-3D model. It compares the hydraulic model tests withFLOW-3D simulation results and gives information on howaccurately a commercially available Computational Fluid Dynamic(CFD) model can predict the spillway discharge capacity andpressure distribution along the spillway bottom surface. 2. Physical ModelA 1/50-scaled undistorted physical model of the Kavsak Damspillway and stilling basin was built and tested at the HydraulicModel Laboratory of State Hydraulic Works of Turkey (DSI).The model was constructed of plexiglas and was fabricated toconform to the distinctive shape of an ogee crest. The spillwayhas 45.8 m in width and 57 m long with a bottom slope of 125%.The length of the stilling basin is about 90 m. During model tests,flow velocities were measured with an ultrasonic flow meter.Pressures on the spillway were measured using a piezometerssçTable 1. Upstream and Downstream Operating Conditions of theKavsak DamRun Upstream reservoir elevation (m)Downstream tailwater elevation (m)1 306.55 168.002 311.35 174.503 314.00 178.904 316.50 182.55Fig. 1. (a) Original Project Design and Final Project Design after Experimental Investigations and Flow Measurement Sections at theApproach, (b) Top View Experimentally Modified Approach in the Laboratory, (c) Side View of the Experimentally Modified Approachin the Laboratory
    Serife Yurdagul Kumcu−2−KSCE Journal of Civil Engineeringthe use of CFD for the assessment of a design, as well as screeningand optimizing of hydraulic structures and cofferdam layouts. Theyconclude that CFD has been successful in optimizing the finalconceptual configuration for the hydraulics design of the project,but recommend that physical modeling still be used as a finalconfirmation.This paper provides experimental studies performed on Kav akDam and analyses the stability of spillway design by usingFLOW-3D model. It compares the hydraulic model tests withFLOW-3D simulation results and gives information on howaccurately a commercially available Computational Fluid Dynamic(CFD) model can predict the spillway discharge capacity andpressure distribution along the spillway bottom surface. 2. Physical ModelA 1/50-scaled undistorted physical model of the Kavsak Damspillway and stilling basin was built and tested at the HydraulicModel Laboratory of State Hydraulic Works of Turkey (DSI).The model was constructed of plexiglas and was fabricated toconform to the distinctive shape of an ogee crest. The spillwayhas 45.8 m in width and 57 m long with a bottom slope of 125%.The length of the stilling basin is about 90 m. During model tests,flow velocities were measured with an ultrasonic flow meter.Pressures on the spillway were measured using a piezometerssçTable 1. Upstream and Downstream Operating Conditions of theKavsak DamRun Upstream reservoir elevation (m)Downstream tailwater elevation (m)1 306.55 168.002 311.35 174.503 314.00 178.904 316.50 182.55Fig. 1. (a) Original Project Design and Final Project Design after Experimental Investigations and Flow Measurement Sections at theApproach, (b) Top View Experimentally Modified Approach in the Laboratory, (c) Side View of the Experimentally Modified Approachin the Laboratory

    References

    Bureau of Reclamation (1977). Design of small dams, U.S. Government Printing Office, Washington, D.C., U.S.

    Bureau of Reclamation (1990). Cavitation in chute and spillways, Engineering Monograph, No.42, U.S. Chanel, P. G. (2008). An evaluation of computational fluid dynamics for

    spillway modeling, MSc Thesis, University of Manitoba Winnipeg, Manitoba, Canada.

    Chanson, H. (2002). The hydraulics of stepped chutes and spillways,Balkema, Lisse, The Netherlands.

    Chanson, H. and Gonzalez, C. A. (2005). “Physical modeling and scale effects of air-water flows on stepped spillways.” Journal of Zhejiang University Science, Vol. 6A, No. 3, pp. 243-250.

    Demiroz, E. (1986). “Specifications of aeration structures which are added to the spillways.” DSI Report, HI-754, DSI-TAKK Publications, Ankara, Turkey.

    Erfanain-Azmoudeh, M. H. and Kamanbedast, A. A. (2013). “Determine the appropriate location of aerator system on gotvandoliadam’s spillway using Flow 3D.” American-Eurasian J. Agric. & Environ. Sci., Vol. 13, No. 3, pp. 378-383, DOI: 10.5829/idosi.aejaes.2013. 13.03. 458.

    Falvey, H. T. (1990). Cavitation in chutes and spillways, Engineering Monograph 42 Water Resources Technical Publication US Printing Office, Bureau of Reclamation, Denver.

    Flow-3D User ’s Manual (2012). Flow science, Inc., Santa Fe, N.M.

    Hirt, C. W. (1992). “Volume-fraction techniques: Powerful tools for flow

    modeling.” Flow Science Report, No. FSI-92-00-02, Flow Science, Inc., Santa Fe, N.M.

    Hirt C. W. and Nichols B. D. (1981). “Volume of Fluid (VOF) method for the dynamics of free boundaries.”Jornal of Computational Physics, Vol. 39, pp. 201-225, DOI: 10.1016/0021-9991(81)90145-5.

    Hirt, C. W. and Sicilian, J. M. (1985). “A Porosity technique for the definition of obstacles in rectangular cell meshes.” Proceedings of the 4th International Conference on Ship Hydro-dynamics, 24-27 September 1985, National Academic of Sciences, Washington DC.

    Ho, D., Boyes, K., Donohoo, S., and Cooper, B. (2003). “Numerical flow analysis for spillways.” 43rd ANCOLD Conference, Hobart, Tas m a nia .

    Johnson, M. C. and Savage, B. M. (2006). “Physical and numerical comparison of flow over ogee spillway in the presence of tailwater.”

    Journal of Hydraulic Engineering, Vol. 132, No. 12, pp. 1353-135, DOI: 10.1061/(ASCE)0733-9429.

    Kim, S. D., Lee, H. J., and An, S. D. (2010). “Improvement of hydraulic stability for spillway using CFD model.” Int. Journal of the Physical Sciences, Vol. 5, No. 6, pp. 774-780.

    Kokpinar, M. A. and Gogus, M. (2002). “High speed jet flows over spillway aerators.” Canadian Journal of Civil Engineering, Vol. 29, No. 6, pp. 885-898, DOI: 10.1139/l02-088.

    Kumcu, S. Y. (2010). Hydraulic model studies of Kavsak Dam and HEPP, DSI Report, HI-1005, DSI-TAKK Publications, Ankara, Turkey.

    Margeirsson, B. (2007). Computational modeling of flow over a spillway, MSc Thesis, Chalmers University of Technology, Gothenburg, Sweden.

    Nichols, B. D. and Hirt, C. W. (1975). “Methods for calculating multi-dimensional, transient free surface flows past bodies.” Proc. First Intern. Conf. Num., Ship Hydrodynamics, Gaithersburg, ML.

    Savage, B. M. and Johnson, M. C. (2001). “Flow over ogee spillway: Physical and numerical model case study.” Journal of Hydraulic Engineering, ASCE, Vol. 127, No. 8, pp. 640-649, DOI: 10.1061/(ASCE)0733-9429.

    Souders, D. T. and Hirt, C. W. (2004). “Modeling entrainment of air at turbulent free surfaces.” Critical Transitions in Water and Environmental resources Management, pp. 1-10.

    entürk, F. (1994). Hydraulics of dams and reservoirs, Water Resources Publication Colorado, USA.

    Teklemariam, E., Korbaylo, B, Groeneveld, J., Sydor, K., and Fuchs, D. (2001). Optimization of hydraulic design using computational fluid dynamics, Waterpower XII, Salt Lake City, Utah.

    Teklemariam, E., Shumilak, B., Sydor, K., Murray, D., Fuchs, D., and Holder, G. (2008). “An integral approach using both physical and computational modeling can be beneficial in addressing the full range of hydraulic design issues.” CDA Annual Conference, Winnipeg, Canada.

    Usta, E. (2014). Numerical investigation of hydraulic characteristics of Laleli Dam spillway and comparison with physical model study, Master Thesis, Middle East Technical University, Ankara, Turkey.

    Versteeg, H. K. and Malalasekera, W. (1996). An introduction to computational fluid dynamics, Longman Scientific and Technical, Longman Group Limited, Harlow, England.

    Vischer, D. L. and Hager, W. H. (1997). Dam hydraulics, J. Wiley & Sons Ltd., England.

    Wagner, W. E. (1967). “Glen Canyon diversion tunnel outlets.” J. Hydraulic Division, ASCE, Vol. 93, No. HY6, pp. 113-134.

    Willey, J., Ewing, T., Wark, B., and Lesleighter, E. (2012). Comple-mentary use of physical and numerical modeling techniques in spillway design refinement, Commission Internationale Des Grands Barrages, Kyoto, June 2012.

    Fig. 1. Averaged error trend.

    Assessment of spillway modeling using computational fluid dynamics

    전산유체역학을 이용한 여수로 모델링 평가

    Authors: Paul G. Chanel and John C. Doering AUTHORS INFO & AFFILIATIONS

    Publication: Canadian Journal of Civil Engineering

    3 December 2008

    Abstract

    Throughout the design and planning period for future hydroelectric generating stations, hydraulic engineers are increasingly integrating computational fluid dynamics (CFD) into the process. As a result, hydraulic engineers are interested in the reliability of CFD software to provide accurate flow data for a wide range of structures, including a variety of different spillways. In the literature, CFD results have generally been in agreement with physical model experimental data. Despite past success, there has not been a comprehensive assessment that looks at the ability of CFD to model a range of different spillway configurations, including flows with various gate openings. In this article, Flow-3D is used to model the discharge over ogee-crested spillways. The numerical model results are compared with physical model studies for three case study evaluations. The comparison indicates that the accuracy of Flow-3D is related to the parameter P/Hd.

    미래의 수력 발전소를 위한 설계 및 계획 기간 동안 유압 엔지니어는 전산유체역학(CFD)을 프로세스에 점점 더 많이 통합하고 있습니다. 결과적으로 유압 엔지니어는 다양한 여수로를 포함하여 광범위한 구조에 대한 정확한 흐름 데이터를 제공하는 CFD 소프트웨어의 신뢰성에 관심을 갖고 있습니다. 문헌에서 CFD 결과는 일반적으로 물리적 모델 실험 데이터와 일치했습니다. 과거의 성공에도 불구하고 다양한 게이트 개구부가 있는 흐름을 포함하여 다양한 여수로 구성을 모델링하는 CFD의 기능을 살펴보는 포괄적인 평가는 없었습니다. 이 기사에서는 Flow-3D를 사용하여 ogee-crested 방수로의 배출을 모델링합니다. 세 가지 사례 연구 평가를 위해 수치 모델 결과를 물리적 모델 연구와 비교합니다. 비교는 Flow-3D의 정확도가 매개변수 P/Hd와 관련되어 있음을 나타냅니다.

    Résumé

    Les ingénieurs en hydraulique intègrent de plus en plus la dynamique des fluides numérique (« CFD ») dans le processus de conception et de planification des futures centrales. Ainsi, les ingénieurs en hydraulique s’intéressent à la fiabilité du logiciel de « CFD » afin de fournir des données précises sur le débit pour une large gamme de structures, incluant différents types d’évacuateurs. Les résultats de « CFD » dans la littérature ont été globalement sont généralement en accord avec les données expérimentales des essais physiques. Malgré les succès antérieurs, il n’y avait aucune évaluation complète de la capacité des « CFD » à modéliser une plage de configuration des évacuateurs, incluant les débits à diverses ouvertures de vannes. Dans le présent article, le logiciel Flow-3D est utilisé pour modéliser le débit par des évacuateurs en doucine. Les résultats du modèle de calcul sont comparés à ceux des essais physiques pour trois études de cas. La comparaison montre que la précision du logiciel Flow-3D est associée au paramètre P/Hd.

    Fig. 1. Averaged error trend.
    Fig. 1. Averaged error trend.

    Get full access to this article

    View all available purchase options and get full access to this article.

    GET ACCESSALREADY A SUBSCRIBER? SIGN IN AS AN INDIVIDUAL OR VIA YOUR INSTITUTION

    References

    Chanel, P.G., and Doering, J.C. 2007. An evaluation of computational fluid dynamics for spillway modelling. In Proceedings of the 16th Australasian Fluid Mechanics Conference (AFMC), Gold Coast, Queensland, Australia, 3–7 December 2007. pp. 1201–1206.

    Google Scholar

    Flow Science, Inc. 2007. Flow-3D user’s manuals. Version 9.2. Flow Science, Inc., Santa Fe, N.M.

    Google Scholar

    Gessler, D. 2005. CFD modeling of spillway performance, EWRI 2005: Impacts of global climate change. In Proceedings of the World Water and Environmental Resources Congress, Anchorage, Alaska, 15–19 May 2005. Edited by R. Walton. American Society of Civil Engineers, Reston, Va.

    Google Scholar

    Hirt, C.W., and Nichols, B.D. 1981. Volume of fluid (VOF) method for the dynamics of free boundaries. Journal of Computational Physics, 39(1): 201–225.

    Crossref

    ISI

    Google Scholar

    Hirt, C.W., and Sicilian, J.M. 1985. A porosity technique for the definition of obstacles in rectangular cell meshes. In Proceedings of the 4th International Conference on Ship Hydro-dynamics, Washington, D.C., 24–27 September 1985. National Academy of Sciences, Washington, D.C.

    Google Scholar

    Ho, D., Cooper, B., Riddette, K., and Donohoo, S. 2006. Application of numerical modelling to spillways in Australia. In Dams and Reservoirs, Societies and Environment in the 21st Century. Edited by Berga et al. Taylor and Francis Group, London.

    Google Scholar

    LaSalle Consulting Group Inc. 1992. Conawapa generating station. Sectional model study of the spillway. LaSalle Consulting Group Inc., Montréal, Que.

    Google Scholar

    Lemke, D.E. 1989. A comparison of the hydraulic performance of an orifice and an overflow spillway in a northern application using physical modeling. M.Sc. thesis, University of Manitoba, Winnipeg, Man.

    Google Scholar

    Savage, B.M., and Johnson, M.C. 2001. Flow over ogee spillway: Physical and numerical model case study. Journal of Hydraulic Engineering, 127(8): 640–649.

    Crossref

    ISI

    Google Scholar

    Teklemariam, E., Korbaylo, B., Groeneveld, J., Sydor, K., and Fuchs, D. 2001. Optimization of hydraulic design using computational fluid dynamics. In Proceedings of Waterpower XII, Salt Lake City, Utah, 9–11 July 2001.

    Google Scholar

    Teklemariam, E., Korbaylo, B., Groeneveld, J., and Fuchs, D. 2002. Computational fluid dynamics: Diverse applications in hydropower project’s design and analysis. In Proceedings of the CWRA 55th Annual Conference, Winnipeg, Man., 11–14 June 2002. Canadian Water Resources Association, Cambridge, Ontario.

    Google Scholar

    Western Canadian Hydraulic Laboratories Inc. 1980. Hydraulics model studies limestone generating station spillway/diversion structure flume study. Final report. Western Canadian Hydraulic Laboratories Inc., Port Coquitlam, B.C.

    Google Scholar

    Fig. 1. (a) Dimensions of the casting with runners (unit: mm), (b) a melt flow simulation using Flow-3D software together with Reilly's model[44], predicted that a large amount of bifilms (denoted by the black particles) would be contained in the final casting. (c) A solidification simulation using Pro-cast software showed that no shrinkage defect was contained in the final casting.

    AZ91 합금 주물 내 연행 결함에 대한 캐리어 가스의 영향

    TianLiabJ.M.T.DaviesaXiangzhenZhuc
    aUniversity of Birmingham, Birmingham B15 2TT, United Kingdom
    bGrainger and Worrall Ltd, Bridgnorth WV15 5HP, United Kingdom
    cBrunel Centre for Advanced Solidification Technology, Brunel University London, Kingston Ln, London, Uxbridge UB8 3PH, United Kingdom

    Abstract

    An entrainment defect (also known as a double oxide film defect or bifilm) acts a void containing an entrapped gas when submerged into a light-alloy melt, thus reducing the quality and reproducibility of the final castings. Previous publications, carried out with Al-alloy castings, reported that this trapped gas could be subsequently consumed by the reaction with the surrounding melt, thus reducing the void volume and negative effect of entrainment defects. Compared with Al-alloys, the entrapped gas within Mg-alloy might be more efficiently consumed due to the relatively high reactivity of magnesium. However, research into the entrainment defects within Mg alloys has been significantly limited. In the present work, AZ91 alloy castings were produced under different carrier gas atmospheres (i.e., SF6/CO2, SF6/air). The evolution processes of the entrainment defects contained in AZ91 alloy were suggested according to the microstructure inspections and thermodynamic calculations. The defects formed in the different atmospheres have a similar sandwich-like structure, but their oxide films contained different combinations of compounds. The use of carrier gases, which were associated with different entrained-gas consumption rates, affected the reproducibility of AZ91 castings.

    연행 결함(이중 산화막 결함 또는 이중막이라고도 함)은 경합금 용융물에 잠길 때 갇힌 가스를 포함하는 공극으로 작용하여 최종 주물의 품질과 재현성을 저하시킵니다. Al-합금 주물을 사용하여 수행된 이전 간행물에서는 이 갇힌 가스가 주변 용융물과의 반응에 의해 후속적으로 소모되어 공극 부피와 연행 결함의 부정적인 영향을 줄일 수 있다고 보고했습니다. Al-합금에 비해 마그네슘의 상대적으로 높은 반응성으로 인해 Mg-합금 내에 포집된 가스가 더 효율적으로 소모될 수 있습니다. 그러나 Mg 합금 내 연행 결함에 대한 연구는 상당히 제한적이었습니다. 현재 작업에서 AZ91 합금 주물은 다양한 캐리어 가스 분위기(즉, SF6/CO2, SF6/공기)에서 생산되었습니다. AZ91 합금에 포함된 연행 결함의 진화 과정은 미세 조직 검사 및 열역학 계산에 따라 제안되었습니다. 서로 다른 분위기에서 형성된 결함은 유사한 샌드위치 구조를 갖지만 산화막에는 서로 다른 화합물 조합이 포함되어 있습니다. 다른 동반 가스 소비율과 관련된 운반 가스의 사용은 AZ91 주물의 재현성에 영향을 미쳤습니다.

    Keywords

    Magnesium alloy, Casting, Oxide film, Bifilm, Entrainment defect, Reproducibility

    1. Introduction

    As the lightest structural metal available on Earth, magnesium became one of the most attractive light metals over the last few decades. The magnesium industry has consequently experienced a rapid development in the last 20 years [1,2], indicating a large growth in demand for Mg alloys all over the world. Nowadays, the use of Mg alloys can be found in the fields of automobiles, aerospace, electronics and etc.[3,4]. It has been predicted that the global consumption of Mg metals will further increase in the future, especially in the automotive industry, as the energy efficiency requirement of both traditional and electric vehicles further push manufactures lightweight their design [3,5,6].

    The sustained growth in demand for Mg alloys motivated a wide interest in the improvement of the quality and mechanical properties of Mg-alloy castings. During a Mg-alloy casting process, surface turbulence of the melt can lead to the entrapment of a doubled-over surface film containing a small quantity of the surrounding atmosphere, thus forming an entrainment defect (also known as a double oxide film defect or bifilm) [7][8][9][10]. The random size, quantity, orientation, and placement of entrainment defects are widely accepted to be significant factors linked to the variation of casting properties [7]. In addition, Peng et al. [11] found that entrained oxides films in AZ91 alloy melt acted as filters to Al8Mn5 particles, trapping them as they settle. Mackie et al. [12] further suggested that entrained oxide films can act to trawl the intermetallic particles, causing them to cluster and form extremely large defects. The clustering of intermetallic compounds made the entrainment defects more detrimental for the casting properties.

    Most of the previous studies regarding entrainment defects were carried out on Al-alloys [7,[13][14][15][16][17][18], and a few potential methods have been suggested for diminishing their negative effect on the quality of Al-alloy castings. Nyahumwa et al.,[16] shows that the void volume within entrainment defects could be reduced by a hot isostatic pressing (HIP) process. Campbell [7] suggested the entrained gas within the defects could be consumed due to reaction with the surrounding melt, which was further verified by Raiszedeh and Griffiths [19].The effect of the entrained gas consumption on the mechanical properties of Al-alloy castings has been investigated by [8,9], suggesting that the consumption of the entrained gas promoted the improvement of the casting reproducibility.

    Compared with the investigation concerning the defects within Al-alloys, research into the entrainment defects within Mg-alloys has been significantly limited. The existence of entrainment defects has been demonstrated in Mg-alloy castings [20,21], but their behaviour, evolution, as well as entrained gas consumption are still not clear.

    In a Mg-alloy casting process, the melt is usually protected by a cover gas to avoid magnesium ignition. The cavities of sand or investment moulds are accordingly required to be flushed with the cover gas prior to the melt pouring [22]. Therefore, the entrained gas within Mg-alloy castings should contain the cover gas used in the casting process, rather than air only, which may complicate the structure and evolution of the corresponding entrainment defects.

    SF6 is a typical cover gas widely used for Mg-alloy casting processes [23][24][25]. Although this cover gas has been restricted to use in European Mg-alloy foundries, a commercial report has pointed out that this cover is still popular in global Mg-alloy industry, especially in the countries which dominated the global Mg-alloy production, such as China, Brazil, India, etc. [26]. In addition, a survey in academic publications also showed that this cover gas was widely used in recent Mg-alloy studies [27]. The protective mechanism of SF6 cover gas (i.e., the reaction between liquid Mg-alloy and SF6 cover gas) has been investigated by several previous researchers, but the formation process of the surface oxide film is still not clearly understood, and even some published results are conflicting with each other. In early 1970s, Fruehling [28] found that the surface film formed under SF6 was MgO mainly with traces of fluorides, and suggested that SF6 was absorbed in the Mg-alloy surface film. Couling [29] further noticed that the absorbed SF6 reacted with the Mg-alloy melt to form MgF2. In last 20 years, different structures of the Mg-alloy surface films have been reported, as detailed below.(1)

    Single-layered film. Cashion [30,31] used X-ray Photoelectron Spectroscopy (XPS) and Auger Spectroscopy (AES) to identify the surface film as MgO and MgF2. He also found that composition of the film was constant throughout the thickness and the whole experimental holding time. The film observed by Cashion had a single-layered structure created from a holding time from 10 min to 100 min.(2)

    Double-layered film. Aarstad et. al [32] reported a doubled-layered surface oxide film in 2003. They observed several well-distributed MgF2 particles attached to the preliminary MgO film and grew until they covered 25–50% of the total surface area. The inward diffusion of F through the outer MgO film was the driving force for the evolution process. This double-layered structure was also supported by Xiong’s group [25,33] and Shih et al. [34].(3)

    Triple-layered film. The triple-layered film and its evolution process were reported in 2002 by Pettersen [35]. Pettersen found that the initial surface film was a MgO phase and then gradually evolved to the stable MgF2 phase by the inward diffusion of F. In the final stage, the film has a triple-layered structure with a thin O-rich interlayer between the thick top and bottom MgF2 layers.(4)

    Oxide film consisted of discrete particles. Wang et al [36] stirred the Mg-alloy surface film into the melt under a SF6 cover gas, and then inspect the entrained surface film after the solidification. They found that the entrained surface films were not continues as the protective surface films reported by other researchers but composed of discrete particles. The young oxide film was composed of MgO nano-sized oxide particles, while the old oxide films consist of coarse particles (about 1  µm in average size) on one side that contained fluorides and nitrides.

    The oxide films of a Mg-alloy melt surface or an entrained gas are both formed due to the reaction between liquid Mg-alloy and the cover gas, thus the above-mentioned research regarding the Mg-alloy surface film gives valuable insights into the evolution of entrainment defects. The protective mechanism of SF6 cover gas (i.e., formation of a Mg-alloy surface film) therefore indicated a potential complicated evolution process of the corresponding entrainment defects.

    However, it should be noted that the formation of a surface film on a Mg-alloy melt is in a different situation to the consumption of an entrained gas that is submerged into the melt. For example, a sufficient amount of cover gas was supported during the surface film formation in the studies previously mentioned, which suppressed the depletion of the cover gas. In contrast, the amount of entrained gas within a Mg-alloy melt is finite, and the entrained gas may become fully depleted. Mirak [37] introduced 3.5%SF6/air bubbles into a pure Mg-alloy melt solidifying in a specially designed permanent mould. It was found that the gas bubbles were entirely consumed, and the corresponding oxide film was a mixture of MgO and MgF2. However, the nucleation sites (such as the MgF2 spots observed by Aarstad [32] and Xiong [25,33]) were not observed. Mirak also speculated that the MgF2 formed prior to MgO in the oxide film based on the composition analysis, which was opposite to the surface film formation process reported in previous literatures (i.e., MgO formed prior to MgF2). Mirak’s work indicated that the oxide-film formation of an entrained gas may be quite different from that of surface films, but he did not reveal the structure and evolution of the oxide films.

    In addition, the use of carrier gas in the cover gases also influenced the reaction between the cover gas and the liquid Mg-alloy. SF6/air required a higher content of SF6 than did a SF6/CO2 carrier gas [38], to avoid the ignition of molten magnesium, revealing different gas-consumption rates. Liang et.al [39] suggested that carbon was formed in the surface film when CO2 was used as a carrier gas, which was different from the films formed in SF6/air. An investigation into Mg combustion [40] reported a detection of Mg2C3 in the Mg-alloy sample after burning in CO2, which not only supported Liang’s results, but also indicated a potential formation of Mg carbides in double oxide film defects.

    The work reported here is an investigation into the behaviour and evolution of entrainment defects formed in AZ91 Mg-alloy castings, protected by different cover gases (i.e., SF6/air and SF6/CO2). These carrier gases have different protectability for liquid Mg alloy, which may be therefore associated with different consumption rates and evolution processes of the corresponding entrained gases. The effect of the entrained-gas consumption on the reproducibility of AZ91 castings was also studied.

    2. Experiment

    2.1. Melting and casting

    Three kilograms AZ91 alloy was melted in a mild steel crucible at 700 ± 5 °C. The composition of the AZ91 alloy has been shown in Table 1. Prior to heating, all oxide scale on the ingot surface was removed by machining. The cover gases used were 0.5%SF6/air or 0.5%SF6/CO2 (vol.%) at a flow rate of 6 L/min for different castings. The melt was degassed by argon with a flow rate of 0.3 L/min for 15 min [41,42], and then poured into sand moulds. Prior to pouring, the sand mould cavity was flushed with the cover gas for 20 min [22]. The residual melt (around 1 kg) was solidified in the crucible.

    Table 1. Composition (wt.%) of the AZ91 alloy used in this study.

    AlZnMnSiFeNiMg
    9.40.610.150.020.0050.0017Residual

    Fig. 1(a) shows the dimensions of the casting with runners. A top-filling system was deliberately used to generate entrainment defects in the final castings. Green and Campbell [7,43] suggested that a top-filling system caused more entrainment events (i.e., bifilms) during a casting process, compared with a bottom-filling system. A melt flow simulation (Flow-3D software) of this mould, using Reilly’s model [44] regarding the entrainment events, also predicted that a large amount of bifilms would be contained in the final casting (denoted by the black particles in Fig. 1b).

    Fig. 1. (a) Dimensions of the casting with runners (unit: mm), (b) a melt flow simulation using Flow-3D software together with Reilly's model[44], predicted that a large amount of bifilms (denoted by the black particles) would be contained in the final casting. (c) A solidification simulation using Pro-cast software showed that no shrinkage defect was contained in the final casting.

    Shrinkage defects also affect the mechanical properties and reproducibility of castings. Since this study focused on the effect of bifilms on the casting quality, the mould has been deliberately designed to avoid generating shrinkage defects. A solidification simulation using ProCAST software showed that no shrinkage defect would be contained in the final casting, as shown in Fig. 1c. The casting soundness has also been confirmed using a real time X-ray prior to the test bar machining.

    The sand moulds were made from resin-bonded silica sand, containing 1wt. % PEPSET 5230 resin and 1wt. % PEPSET 5112 catalyst. The sand also contained 2 wt.% Na2SiF6 to act as an inhibitor [45]. The pouring temperature was 700 ± 5 °C. After the solidification, a section of the runner bars was sent to the Sci-Lab Analytical Ltd for a H-content analysis (LECO analysis), and all the H-content measurements were carried out on the 5th day after the casting process. Each of the castings was machined into 40 test bars for a tensile strength test, using a Zwick 1484 tensile test machine with a clip extensometer. The fracture surfaces of the broken test bars were examined using Scanning Electron Microscope (SEM, Philips JEOL7000) with an accelerating voltage of 5–15 kV. The fractured test bars, residual Mg-alloy solidified in the crucible, and the casting runners were then sectioned, polished and also inspected using the same SEM. The cross-section of the oxide film found on the test-bar fracture surface was exposed by the Focused Ion Beam milling technique (FIB), using a CFEI Quanta 3D FEG FIB-SEM. The oxide film required to be analysed was coated with a platinum layer. Then, a gallium ion beam, accelerated to 30 kV, milled the material substrate surrounding the platinum coated area to expose the cross section of the oxide film. EDS analysis of the oxide film’s cross section was carried out using the FIB equipment at accelerating voltage of 30 kV.

    2.2. Oxidation cell

    As previously mentioned, several past researchers investigated the protective film formed on a Mg-alloy melt surface [38,39,[46][47][48][49][50][51][52]. During these experiments, the amount of cover gas used was sufficient, thus suppressing the depletion of fluorides in the cover gas. The experiment described in this section used a sealed oxidation cell, which limited the supply of cover gas, to study the evolution of the oxide films of entrainment defects. The cover gas contained in the oxidation cell was regarded as large-size “entrained bubble”.

    As shown in Fig. 2, the main body of the oxidation cell was a closed-end mild steel tube which had an inner length of 400 mm, and an inner diameter of 32 mm. A water-cooled copper tube was wrapped around the upper section of the cell. When the tube was heated, the cooling system created a temperature difference between the upper and lower sections, causing the interior gas to convect within the tube. The temperature was monitored by a type-K thermocouple located at the top of the crucible. Nie et al. [53] suggested that the SF6 cover gas would react with the steel wall of the holding furnace when they investigated the surface film of a Mg-alloy melt. To avoid this reaction, the interior surface of the steel oxidation cell (shown in Fig. 2) and the upper half section of the thermocouple were coated with boron nitride (the Mg-alloy was not in contact with boron nitride).

    Fig. 2. Schematic of the oxidation cell used to study the evolution of the oxide films of the entrainment defects (unit mm).

    During the experiment, a block of solid AZ91 alloy was placed in a magnesia crucible located at the bottom of the oxidation cell. The cell was heated to 100 °C in an electric resistance furnace under a gas flow rate of 1 L/min. The cell was held at this temperature for 20 min, to replace the original trapped atmosphere (i.e. air). Then, the oxidation cell was further heated to 700 °C, melting the AZ91 sample. The gas inlet and exit valves were then closed, creating a sealed environment for oxidation under a limited supply of cover gas. The oxidation cell was then held at 700 ± 10 °C for periods of time from 5 min to 30 min in 5-min intervals. At the end of each holding time, the cell was quenched in water. After cooling to room temperature, the oxidised sample was sectioned, polished, and subsequently examined by SEM.

    3. Results

    3.1. Structure and composition of the entrainment defects formed in SF6/air

    The structure and composition of the entrainment defect formed in the AZ91 castings under a cover gas of 0.5%SF6/air was observed by SEM and EDS. The results indicate that there exist two types of entrainment defects which are sketched in Fig. 3: (1) Type A defect whose oxide film has a traditional single-layered structure and (2) Type B defect, whose oxide film has two layers. The details of these defects were introduced in the following. Here it should be noticed that, as the entrainment defects are also known as biofilms or double oxide film, the oxide films of Type B defect were referred to as “multi-layered oxide film” or “multi-layered structure” in the present work to avoid a confusing description such as “the double-layered oxide film of a double oxide film defect”.

    Fig. 3. Schematic of the different types of entrainment defects found in AZ91 castings. (a) Type A defect with a single-layered oxide film and (b) Type B defect with two-layered oxide film.

    Fig. 4(a-b) shows a Type A defect having a compact single-layered oxide film with about 0.4 µm thickness. Oxygen, fluorine, magnesium and aluminium were detected in this film (Fig. 4c). It is speculated that oxide film is the mixture of fluoride and oxide of magnesium and aluminium. The detection of fluorine revealed that an entrained cover gas was contained in the formation of this defect. That is to say that the pores shown in Fig. 4(a) were not shrinkage defects or hydrogen porosity, but entrainment defects. The detection of aluminium was different with Xiong and Wang’s previous study [47,48], which showed that no aluminium was contained in their surface film of an AZ91 melt protected by a SF6 cover gas. Sulphur could not be clearly recognized in the element map, but there was a S-peak in the corresponding ESD spectrum.

    Fig. 4. (a) A Type A entrainment defect formed in SF6/air and having a single-layered oxide film, (b) the oxide film of this defect, (c) SEM-EDS element maps (using Philips JEOL7000) corresponding to the area highlighted in (b).

    Fig. 5(a-b) shows a Type B entrainment defect having a multi-layered oxide film. The compact outer layers of the oxide films were enriched with fluorine and oxygen (Fig. 5c), while their relatively porous inner layers were only enriched with oxygen (i.e., poor in fluorine) and partly grew together, thus forming a sandwich-like structure. Therefore, it is speculated that the outer layer is the mixture of fluoride and oxide, while the inner layer is mainly oxide. Sulphur could only be recognized in the EDX spectrum and could not be clearly identified in the element map, which might be due to the small S-content in the cover gas (i.e., 0.5% volume content of SF6 in the cover gas). In this oxide film, aluminium was contained in the outer layer of this oxide film but could not be clearly detected in the inner layer. Moreover, the distribution of Al seems to be uneven. It can be found that, in the right side of the defect, aluminium exists in the film but its concentration can not be identified to be higher than the matrix. However, there is a small area with much higher aluminium concentration in the left side of the defect. Such an uneven distribution of aluminium was also observed in other defects (shown in the following), and it is the result of the formation of some oxide particles in or under the film.

    Fig. 5. (a) A Type B entrainment defect formed in SF6/air and having a multi-layered oxide film, (b) the oxide films of this defect have grown together, (c) SEM-EDS element maps (using Philips JEOL7000) corresponding to the area shown in (b).

    Figs. 4 and 5 show cross sectional observations of the entrainment defects formed in the AZ91 alloy sample cast under a cover gas of SF6/air. It is not sufficient to characterize the entrainment defects only by the figures observed from the two-dimensional section. To have a further understanding, the surface of the entrainment defects (i.e. the oxide film) was further studied by observing the fracture surface of the test bars.

    Fig. 6(a) shows fracture surfaces of an AZ91 alloy tensile test bar produced in SF6/air. Symmetrical dark regions can be seen on both sides of the fracture surfaces. Fig. 6(b) shows boundaries between the dark and bright regions. The bright region consisted of jagged and broken features, while the surface of the dark region was relatively smooth and flat. In addition, the EDS results (Fig. 6c-d and Table 2) show that fluorine, oxygen, sulphur, and nitrogen were only detected in the dark regions, indicating that the dark regions were surface protective films entrained into the melt. Therefore, it could be suggested that the dark regions were an entrainment defect with consideration of their symmetrical nature. Similar defects on fracture surfaces of Al-alloy castings have been previously reported [7]Nitrides were only found in the oxide films on the test-bar fracture surfaces but never detected in the cross-sectional samples shown in Figs. 4 and 5. An underlying reason is that the nitrides contained in these samples may have hydrolysed during the sample polishing process [54].

    Fig. 6. (a) A pair of the fracture surfaces of a AZ91 alloy tensile test bar produced under a cover gas of SF6/air. The dimension of the fracture surface is 5 mm × 6 mm, (b) a section of the boundary between the dark and bright regions shown in (a), (c-d) EDS spectrum of the (c) bright regions and (d) dark regions, (e) schematic of an entrainment defect contained in a test bar.

    Table 2. EDS results (wt.%) corresponding to the regions shown in Fig. 6 (cover gas: SF6/air).

    Empty CellCOMgFAlZnSN
    Dark region in Fig. 6(b)3.481.3279.130.4713.630.570.080.73
    Bright region in Fig. 6(b)3.5884.4811.250.68

    In conjunction with the cross-sectional observation of the defects shown in Figs. 4 and 5, the structure of an entrainment defect contained in a tensile test bar was sketched as shown in Fig. 6(e). The defect contained an entrained gas enclosed by its oxide film, creating a void section inside the test bar. When the tensile force applied on the defect during the fracture process, the crack was initiated at the void section and propagated along the entrainment defect, since cracks would be propagated along the weakest path [55]. Therefore, when the test bar was finally fractured, the oxide films of entrainment defect appeared on both fracture surfaces of the test bar, as shown in Fig. 6(a).

    3.2. Structure and composition of the entrainment defects formed in SF6/CO2

    Similar to the entrainment defect formed in SF6/air, the defects formed under a cover gas of 0.5%SF6/CO2 also had two types of oxide films (i.e., single-layered and multi-layered types). Fig. 7(a) shows an example of the entrainment defects containing a multi-layered oxide film. A magnified observation to the defect (Fig. 7b) shows that the inner layers of the oxide films had grown together, presenting a sandwich-like structure, which was similar to the defects formed in an atmosphere of SF6/air (Fig. 5b). An EDS spectrum (Fig. 7c) revealed that the joint area (inner layer) of this sandwich-like structure mainly contained magnesium oxides. Peaks of fluorine, sulphur, and aluminium were recognized in this EDS spectrum, but their amount was relatively small. In contrast, the outer layers of the oxide films were compact and composed of a mixture of fluorides and oxides (Fig. 7d-e).

    Fig. 7. (a) An example of entrainment defects formed in SF6/CO2 and having a multi-layered oxide film, (b) magnified observation of the defect, showing the inner layer of the oxide films has grown together, (c) EDS spectrum of the point denoted in (b), (d) outer layer of the oxide film, (e) SEM-EDS element maps (using Philips JEOL7000) corresponding to the area shown in (d).

    Fig. 8(a) shows an entrainment defect on the fracture surfaces of an AZ91 alloy tensile test bar, which was produced in an atmosphere of 0.5%SF6/CO2. The corresponding EDS results (Table 3) showed that oxide film contained fluorides and oxides. Sulphur and nitrogen were not detected. Besides, a magnified observation (Fig. 8b) indicated spots on the oxide film surface. The diameter of the spots ranged from hundreds of nanometres to a few micron meters.

    Fig. 8. (a) A pair of the fracture surfaces of a AZ91 alloy tensile test bar, produced in an atmosphere of SF6/CO2. The dimension of the fracture surface is 5 mm × 6 mm, (b) surface appearance of the oxide films on the fracture surfaces, showing spots on the film surface.

    To further reveal the structure and composition of the oxide film clearly, the cross-section of the oxide film on a test-bar fracture surface was onsite exposed using the FIB technique (Fig. 9). As shown in Fig. 9a, a continuous oxide film was found between the platinum coating layer and the Mg-Al alloy substrate. Fig. 9 (b-c) shows a magnified observation to oxide films, indicating a multi-layered structure (denoted by the red box in Fig. 9c). The bottom layer was enriched with fluorine and oxygen and should be the mixture of fluoride and oxide, which was similar to the “outer layer” shown in Figs. 5 and 7, while the only-oxygen-enriched top layer was similar to the “inner layer” shown in Figs. 5 and 7.

    Fig. 9. (a) A cross-sectional observation of the oxide film on the fracture surface of the AZ91 casting produced in SF6/CO2, exposed by FIB, (b) a magnified observation of area highlighted in (a), and (c) SEM-EDS elements map of the area shown in (b), obtained by CFEI Quanta 3D FEG FIB-SEM.

    Except the continuous film, some individual particles were also observed in or below the continuous film, as shown in Fig. 9. An Al-enriched particle was detected in the left side of the oxide film shown in Fig. 9b and might be speculated to be spinel Mg2AlO4 because it also contains abundant magnesium and oxygen elements. The existing of such Mg2AlO4 particles is responsible for the high concentration of aluminium in small areas of the observed film and the uneven distribution of aluminium, as shown in Fig. 5(c). Here it should be emphasized that, although the other part of the bottom layer of the continuous oxide film contains less aluminium than this Al-enriched particle, the Fig. 9c indicated that the amount of aluminium in this bottom layer was still non-negligible, especially when comparing with the outer layer of the film. Below the right side of the oxide film shown in Fig. 9b, a particle was detected and speculated to be MgO because it is rich in Mg and O. According to Wang’s result [56], lots of discrete MgO particles can be formed on the surface of the Mg melt by the oxidation of Mg melt and Mg vapor. The MgO particles observed in our present work may be formed due to the same reasons. While, due to the differences in experimental conditions, less Mg melt can be vapored or react with O2, thus only a few of MgO particles formed in our work. An enrichment of carbon was also found in the film, revealing that CO2 was able to react with the melt, thus forming carbon or carbides. This carbon concentration was consistent with the relatively high carbon content of the oxide film shown in Table 3 (i.e., the dark region). In the area next to the oxide film.

    Table 3. EDS results (wt.%) corresponding to the regions shown in Fig. 8 (cover gas: SF6/ CO2).

    Empty CellCOMgFAlZnSN
    Dark region in Fig. 8(a)7.253.6469.823.827.030.86
    Bright region in Fig. 8(a)2.100.4482.8313.261.36

    This cross-sectional observation of the oxide film on a test bar fracture surface (Fig. 9) further verified the schematic of the entrainment defect shown in Fig. 6(e). The entrainment defects formed in different atmospheres of SF6/CO2 and SF6/air had similar structures, but their compositions were different.

    3.3. Evolution of the oxide films in the oxidation cell

    The results in Section 3.1 and 3.2 have shown the structures and compositions of entrainment defects formed in AZ91 castings under cover gases of SF6/air and SF6/CO2. Different stages of the oxidation reaction may lead to the different structures and compositions of entrainment defects. Although Campbell has conjectured that an entrained gas may react with the surrounding melt, it is rarely reported that the reaction occurring between the Mg-alloy melt and entrapped cover gas. Previous researchers normally focus on the reaction between a Mg-alloy melt and the cover gas in an open environment [38,39,[46][47][48][49][50][51][52], which was different from the situation of a cover gas trapped into the melt. To further understand the formation of the entrainment defect in an AZ91 alloy, the evolution process of oxide films of the entrainment defect was further studied using an oxidation cell.

    Fig. 10 (a and d) shows a surface film held for 5 min in the oxidation cell, protected by 0.5%SF6/air. There was only one single layer consisting of fluoride and oxide (MgF2 and MgO). In this surface film. Sulphur was detected in the EDS spectrum, but its amount was too small to be recognized in the element map. The structure and composition of this oxide film was similar to the single-layered films of entrainment defects shown in Fig. 4.

    Fig. 10. Oxide films formed in the oxidation cell under a cover gas of 0.5%SF6/air and held at 700 °C for (a) 5 min; (b) 10 min; (c) 30 min, and (d-f) the SEM-EDS element maps (using Philips JEOL7000) corresponding to the oxide film shown in (a-c) respectively, (d) 5 min; (e) 10 min; (f) 30 min. The red points in (c and f) are the location references, denoting the boundary of the F-enriched layer in different element maps.

    After a holding time of 10 min, a thin (O, S)-enriched top layer (around 700 nm) appeared upon the preliminary F-enriched film, forming a multi-layered structure, as shown in Fig. 10(b and e). The thickness of the (O, S)-enriched top layer increased with increased holding time. As shown in Fig. 10(c and f), the oxide film held for 30 min also had a multi-layered structure, but the thickness of its (O, S)-enriched top layer (around 2.5 µm) was higher than the that of the 10-min oxide film. The multi-layered oxide films shown in Fig. 10(b-c) presented a similar appearance to the films of the sandwich-like defect shown in Fig. 5.

    The different structures of the oxide films shown in Fig. 10 indicated that fluorides in the cover gas would be preferentially consumed due to the reaction with the AZ91 alloy melt. After the depletion of fluorides, the residual cover gas reacted further with the liquid AZ91 alloy, forming the top (O, S)-enriched layer in the oxide film. Therefore, the different structures and compositions of entrainment defects shown in Figs. 4 and 5 may be due to an ongoing oxidation reaction between melt and entrapped cover gas.

    This multi-layered structure has not been reported in previous publications concerning the protective surface film formed on a Mg-alloy melt [38,[46][47][48][49][50][51]. This may be due to the fact that previous researchers carried out their experiments with an un-limited amount of cover gas, creating a situation where the fluorides in the cover gas were not able to become depleted. Therefore, the oxide film of an entrainment defect had behaviour traits similar to the oxide films shown in Fig. 10, but different from the oxide films formed on the Mg-alloy melt surface reported in [38,[46][47][48][49][50][51].

    Similar with the oxide films held in SF6/air, the oxide films formed in SF6/CO2 also had different structures with different holding times in the oxidation cell. Fig. 11(a) shows an oxide film, held on an AZ91 melt surface under a cover gas of 0.5%SF6/CO2 for 5 min. This film had a single-layered structure consisting of MgF2. The existence of MgO could not be confirmed in this film. After the holding time of 30 min, the film had a multi-layered structure; the inner layer was of a compact and uniform appearance and composed of MgF2, while the outer layer is the mixture of MgF2 and MgO. Sulphur was not detected in this film, which was different from the surface film formed in 0.5%SF6/air. Therefore, fluorides in the cover gas of 0.5%SF6/CO2 were also preferentially consumed at an early stage of the film growth process. Compared with the film formed in SF6/air, the MgO in film formed in SF6/CO2 appeared later and sulphide did not appear within 30 min. It may mean that the formation and evolution of film in SF6/air is faster than SF6/CO2. CO2 may have subsequently reacted with the melt to form MgO, while sulphur-containing compounds accumulated in the cover gas and reacted to form sulphide in very late stage (may after 30 min in oxidation cell).

    Fig. 11. Oxide films formed in the oxidation cell under a cover gas of 0.5%SF6/CO2, and their SEM-EDS element maps (using Philips JEOL7000). They were held at 700 °C for (a) 5 min; (b) 30 min. The red points in (b) are the location references, denoting the boundary between the top and bottom layers in the oxide film.

    4. Discussion

    4.1. Evolution of entrainment defects formed in SF6/air

    HSC software from Outokumpu HSC Chemistry for Windows (http://www.hsc-chemistry.net/) was used to carry out thermodynamic calculations needed to explore the reactions which might occur between the trapped gases and liquid AZ91 alloy. The solutions to the calculations suggest which products are most likely to form in the reaction process between a small amount of cover gas (i.e., the amount within a trapped bubble) and the AZ91-alloy melt.

    In the trials, the pressure was set to 1 atm, and the temperature set to 700 °C. The amount of the cover gas was assumed to be 7 × 10−7 kg, with a volume of approximately 0.57 cm3 (3.14 × 10−8 kmol) for 0.5%SF6/air, and 0.35 cm3 (3.12 × 10−8 kmol) for 0.5%SF6/CO2. The amount of the AZ91 alloy melt in contact with the trapped gas was assumed to be sufficient to complete all reactions. The decomposition products of SF6 were SF5, SF4, SF3, SF2, F2, S(g), S2(g) and F(g) [57][58][59][60].

    Fig. 12 shows the equilibrium diagram of the thermodynamic calculation of the reaction between the AZ91 alloy and 0.5%SF6/air. In the diagram, the reactants and products with less than 10−15 kmol have not been shown, as this was 5 orders of magnitude less than the amount of SF6 present (≈ 1.57 × 10−10 kmol) and therefore would not affect the observed process in a practical way.

    Fig. 12. An equilibrium diagram for the reaction between 7e-7 kg 0.5%SF6/air and a sufficient amount of AZ91 alloy. The X axis is the amount of AZ91 alloy melt having reacted with the entrained gas, and the vertical Y-axis is the amount of the reactants and products.

    This reaction process could be divided into 3 stages.

    Stage 1: The formation of fluorides. the AZ91 melt preferentially reacted with SF6 and its decomposition products, producing MgF2, AlF3, and ZnF2. However, the amount of ZnF2 may have been too small to be detected practically (1.25 × 10−12 kmol of ZnF2 compared with 3 × 10−10 kmol of MgF2), which may be the reason why Zn was not detected in any the oxide films shown in Sections 3.13.3. Meanwhile, sulphur accumulated in the residual gas as SO2.

    Stage 2: The formation of oxides. After the liquid AZ91 alloy had depleted all the available fluorides in the entrapped gas, the amount of AlF3 and ZnF2 quickly reduced due to a reaction with Mg. O2(g) and SO2 reacted with the AZ91 melt, forming MgO, Al2O3, MgAl2O4, ZnO, ZnSO4 and MgSO4. However, the amount of ZnO and ZnSO4 would have been too small to be found practically by EDS (e.g. 9.5 × 10−12 kmol of ZnO,1.38 × 10−14 kmol of ZnSO4, in contrast to 4.68 × 10−10 kmol of MgF2, when the amount of AZ91 on the X-axis is 2.5 × 10−9 kmol). In the experimental cases, the concentration of F in the cover gas is very low, whole the concentration f O is much higher. Therefore, the stage 1 and 2, i.e, the formation of fluoride and oxide may happen simultaneously at the beginning of the reaction, resulting in the formation of a singer-layered mixture of fluoride and oxide, as shown in Figs. 4 and 10(a). While an inner layer consisted of oxides but fluorides could form after the complete depletion of F element in the cover gas.

    Stages 1- 2 theoretically verified the formation process of the multi-layered structure shown in Fig. 10.

    The amount of MgAl2O4 and Al2O3 in the oxide film was of a sufficient amount to be detected, which was consistent with the oxide films shown in Fig. 4. However, the existence of aluminium could not be recognized in the oxide films grown in the oxidation cell, as shown in Fig. 10. This absence of Al may be due to the following reactions between the surface film and AZ91 alloy melt:(1)

    Al2O3 + 3Mg + = 3MgO + 2Al, △G(700 °C) = -119.82 kJ/mol(2)

    Mg + MgAl2O4 = MgO + Al, △G(700 °C) =-106.34 kJ/molwhich could not be simulated by the HSC software since the thermodynamic calculation was carried out under an assumption that the reactants were in full contact with each other. However, in a practical process, the AZ91 melt and the cover gas would not be able to be in contact with each other completely, due to the existence of the protective surface film.

    Stage 3: The formation of Sulphide and nitride. After a holding time of 30 min, the gas-phase fluorides and oxides in the oxidation cell had become depleted, allowing the melt reaction with the residual gas, forming an additional sulphur-enriched layer upon the initial F-enriched or (F, O)-enriched surface film, thus resulting in the observed multi-layered structure shown in Fig. 10 (b and c). Besides, nitrogen reacted with the AZ91 melt until all reactions were completed. The oxide film shown in Fig. 6 may correspond to this reaction stage due to its nitride content. However, the results shows that the nitrides were not detected in the polished samples shown in Figs. 4 and 5, but only found on the test bar fracture surfaces. The nitrides may have hydrolysed during the sample preparation process, as follows [54]:(3)

    Mg3N2 + 6H2O =3Mg(OH)2 + 2NH3↑(4)

    AlN+ 3H2O =Al(OH)3 + NH3

    In addition, Schmidt et al. [61] found that Mg3N2 and AlN could react to form ternary nitrides (Mg3AlnNn+2, n= 1, 2, 3…). HSC software did not contain the database of ternary nitrides, and it could not be added into the calculation. The oxide films in this stage may also contain ternary nitrides.

    4.2. Evolution of entrainment defects formed in SF6/CO2

    Fig. 13 shows the results of the thermodynamic calculation between AZ91 alloy and 0.5%SF6/CO2. This reaction processes can also be divided into three stages.

    Fig. 13. An equilibrium diagram for the reaction between 7e-7 kg 0.5%SF6/CO2 and a sufficient amount of AZ91 alloy. The X axis denotes the amount of Mg alloy melt having reacted with the entrained gas, and the vertical Y-axis denotes the amounts of the reactants and products.

    Stage 1: The formation of fluorides. SF6 and its decomposition products were consumed by the AZ91 melt, forming MgF2, AlF3, and ZnF2. As in the reaction of AZ91 in 0.5%SF6/air, the amount of ZnF2 was too small to be detected practically (1.51 × 10−13 kmol of ZnF2 compared with 2.67 × 10−10 kmol of MgF2). Sulphur accumulated in the residual trapped gas as S2(g) and a portion of the S2(g) reacted with CO2, to form SO2 and CO. The products in this reaction stage were consistent with the film shown in Fig. 11(a), which had a single layer structure that contained fluorides only.

    Stage 2: The formation of oxides. AlF3 and ZnF2 reacted with the Mg in the AZ91 melt, forming MgF2, Al and Zn. The SO2 began to be consumed, producing oxides in the surface film and S2(g) in the cover gas. Meanwhile, the CO2 directly reacted with the AZ91 melt, forming CO, MgO, ZnO, and Al2O3. The oxide films shown in Figs. 9 and 11(b) may correspond to this reaction stage due to their oxygen-enriched layer and multi-layered structure.

    The CO in the cover gas could further react with the AZ91 melt, producing C. This carbon may further react with Mg to form Mg carbides, when the temperature reduced (during solidification period) [62]. This may be the reason for the high carbon content in the oxide film shown in Figs. 89. Liang et al. [39] also reported carbon-detection in an AZ91 alloy surface film protected by SO2/CO2. The produced Al2O3 may be further combined with MgO, forming MgAl2O4 [63]. As discussed in Section 4.1, the alumina and spinel can react with Mg, causing an absence of aluminium in the surface films, as shown in Fig. 11.

    Stage 3: The formation of Sulphide. the AZ91 melt began to consume S2(g) in the residual entrapped gas, forming ZnS and MgS. These reactions did not occur until the last stage of the reaction process, which could be the reason why the S-content in the defect shown Fig. 7(c) was small.

    In summary, thermodynamic calculations indicate that the AZ91 melt will react with the cover gas to form fluorides firstly, then oxides and sulphides in the last. The oxide film in the different reaction stages would have different structures and compositions.

    4.3. Effect of the carrier gases on consumption of the entrained gas and the reproducibility of AZ91 castings

    The evolution processes of entrainment defects, formed in SF6/air and SF6/CO2, have been suggested in Sections 4.1 and 4.2. The theoretical calculations were verified with respect to the corresponding oxide films found in practical samples. The atmosphere within an entrainment defect could be efficiently consumed due to the reaction with liquid Mg-alloy, in a scenario dissimilar to the Al-alloy system (i.e., nitrogen in an entrained air bubble would not efficiently react with Al-alloy melt [64,65], however, nitrogen would be more readily consumed in liquid Mg alloys, commonly referred to as “nitrogen burning” [66]).

    The reaction between the entrained gas and the surrounding liquid Mg-alloy converted the entrained gas into solid compounds (e.g. MgO) within the oxide film, thus reducing the void volume of the entrainment defect and hence probably causing a collapse of the defect (e.g., if an entrained gas of air was depleted by the surrounding liquid Mg-alloy, under an assumption that the melt temperature is 700 °C and the depth of liquid Mg-alloy is 10 cm, the total volume of the final solid products would be 0.044% of the initial volume taken by the entrapped air).

    The relationship between the void volume reduction of entrainment defects and the corresponding casting properties has been widely studied in Al-alloy castings. Nyahumwa and Campbell [16] reported that the Hot Isostatic Pressing (HIP) process caused the entrainment defects in Al-alloy castings to collapse and their oxide surfaces forced into contact. The fatigue lives of their castings were improved after HIP. Nyahumwa and Campbell [16] also suggested a potential bonding of the double oxide films that were in contact with each other, but there was no direct evidence to support this. This binding phenomenon was further investigated by Aryafar et.al.[8], who re-melted two Al-alloy bars with oxide skins in a steel tube and then carried out a tensile strength test on the solidified sample. They found that the oxide skins of the Al-alloy bars strongly bonded with each other and became even stronger with an extension of the melt holding time, indicating a potential “healing” phenomenon due to the consumption of the entrained gas within the double oxide film structure. In addition, Raidszadeh and Griffiths [9,19] successfully reduced the negative effect of entrainment defects on the reproducibility of Al-alloy castings, by extending the melt holding time before solidification, which allowed the entrained gas to have a longer time to react with the surrounding melt.

    With consideration of the previous work mentioned, the consumption of the entrained gas in Mg-alloy castings may diminish the negative effect of entrainment defects in the following two ways.

    (1) Bonding phenomenon of the double oxide films. The sandwich-like structure shown in Fig. 5 and 7 indicated a potential bonding of the double oxide film structure. However, more evidence is required to quantify the increase in strength due to the bonding of the oxide films.

    (2) Void volume reduction of entrainment defects. The positive effect of void-volume reduction on the quality of castings has been widely demonstrated by the HIP process [67]. As the evolution processes discussed in Section 4.14.2, the oxide films of entrainment defects can grow together due to an ongoing reaction between the entrained gas and surrounding AZ91 alloy melt. The volume of the final solid products was significant small compared with the entrained gas (i.e., 0.044% as previously mentioned).

    Therefore, the consumption rate of the entrained gas (i.e., the growth rate of oxide films) may be a critical parameter for improving the quality of AZ91 alloy castings. The oxide film growth rate in the oxidization cell was accordingly further investigated.

    Fig. 14 shows a comparison of the surface film growth rates in different cover gases (i.e., 0.5%SF6/air and 0.5%SF6/CO2). 15 random points on each sample were selected for film thickness measurements. The 95% confidence interval (95%CI) was computed under an assumption that the variation of the film thickness followed a Gaussian distribution. It can be seen that all the surface films formed in 0.5%SF6/air grew faster than those formed in 0.5%SF6/CO2. The different growth rates suggested that the entrained-gas consumption rate of 0.5%SF6/air was higher than that of 0.5%SF6/CO2, which was more beneficial for the consumption of the entrained gas.

    Fig. 14. A comparison of the AZ91 alloy oxide film growth rates in 0.5%SF6/air and 0.5%SF6/CO2

    It should be noted that, in the oxidation cell, the contact area of liquid AZ91 alloy and cover gas (i.e. the size of the crucible) was relatively small with consideration of the large volume of melt and gas. Consequently, the holding time for the oxide film growth within the oxidation cell was comparatively long (i.e., 5–30 min). However, the entrainment defects contained in a real casting are comparatively very small (i.e., a few microns size as shown in Figs. 36, and [7]), and the entrained gas is fully enclosed by the surrounding melt, creating a relatively large contact area. Hence the reaction time for cover gas and the AZ91 alloy melt may be comparatively short. In addition, the solidification time of real Mg-alloy sand castings can be a few minutes (e.g. Guo [68] reported that a Mg-alloy sand casting with 60 mm diameter required 4 min to be solidified). Therefore, it can be expected that an entrained gas trapped during an Mg-alloy melt pouring process will be readily consumed by the surrounding melt, especially for sand castings and large-size castings, where solidification times are long.

    Therefore, the different cover gases (0.5%SF6/air and 0.5%SF6/CO2) associated with different consumption rates of the entrained gases may affect the reproducibility of the final castings. To verify this assumption, the AZ91 castings produced in 0.5%SF6/air and 0.5%SF6/CO2 were machined into test bars for mechanical evaluation. A Weibull analysis was carried out using both linear least square (LLS) method and non-linear least square (non-LLS) method [69].

    Fig. 15(a-b) shows a traditional 2-p linearized Weibull plot of the UTS and elongation of the AZ91 alloy castings, obtained by the LLS method. The estimator used is P= (i-0.5)/N, which was suggested to cause the lowest bias among all the popular estimators [69,70]. The casting produced in SF6/air has an UTS Weibull moduli of 16.9, and an elongation Weibull moduli of 5.0. In contrast, the UTS and elongation Weibull modulus of the casting produced in SF6/CO2 are 7.7 and 2.7 respectively, suggesting that the reproducibility of the casting protected by SF6/CO2 were much lower than that produced in SF6/air.

    Fig. 15. The Weibull modulus of AZ91 castings produced in different atmospheres, estimated by (a-b) the linear least square method, (c-d) the non-linear least square method, where SSR is the sum of residual squares.

    In addition, the author’s previous publication [69] demonstrated a shortcoming of the linearized Weibull plots, which may cause a higher bias and incorrect R2 interruption of the Weibull estimation. A Non-LLS Weibull estimation was therefore carried out, as shown in Fig. 15 (c-d). The UTS Weibull modulus of the SF6/air casting was 20.8, while the casting produced under SF6/CO2 had a lower UTS Weibull modulus of 11.4, showing a clear difference in their reproducibility. In addition, the SF6/air elongation (El%) dataset also had a Weibull modulus (shape = 5.8) higher than the elongation dataset of SF6/CO2 (shape = 3.1). Therefore, both the LLS and Non-LLS estimations suggested that the SF6/air casting has a higher reproducibility than the SF6/CO2 casting. It supports the method that the use of air instead of CO2 contributes to a quicker consumption of the entrained gas, which may reduce the void volume within the defects. Therefore, the use of 0.5%SF6/air instead of 0.5%SF6/CO2 (which increased the consumption rate of the entrained gas) improved the reproducibility of the AZ91 castings.

    However, it should be noted that not all the Mg-alloy foundries followed the casting process used in present work. The Mg-alloy melt in present work was degassed, thus reducing the effect of hydrogen on the consumption of the entrained gas (i.e., hydrogen could diffuse into the entrained gas, potentially suppressing the depletion of the entrained gas [7,71,72]). In contrast, in Mg-alloy foundries, the Mg-alloy melt is not normally degassed, since it was widely believed that there is not a ‘gas problem’ when casting magnesium and hence no significant change in tensile properties [73]. Although studies have shown the negative effect of hydrogen on the mechanical properties of Mg-alloy castings [41,42,73], a degassing process is still not very popular in Mg-alloy foundries.

    Moreover, in present work, the sand mould cavity was flushed with the SF6 cover gas prior to pouring [22]. However, not all the Mg-alloy foundries flushed the mould cavity in this way. For example, the Stone Foundry Ltd (UK) used sulphur powder instead of the cover-gas flushing. The entrained gas within their castings may be SO2/air, rather than the protective gas.

    Therefore, although the results in present work have shown that using air instead of CO2 improved the reproducibility of the final casting, it still requires further investigations to confirm the effect of carrier gases with respect to different industrial Mg-alloy casting processes.

    7. Conclusion

    Entrainment defects formed in an AZ91 alloy were observed. Their oxide films had two types of structure: single-layered and multi-layered. The multi-layered oxide film can grow together forming a sandwich-like structure in the final casting.2.

    Both the experimental results and the theoretical thermodynamic calculations demonstrated that fluorides in the trapped gas were depleted prior to the consumption of sulphur. A three-stage evolution process of the double oxide film defects has been suggested. The oxide films contained different combinations of compounds, depending on the evolution stage. The defects formed in SF6/air had a similar structure to those formed in SF6/CO2, but the compositions of their oxide films were different. The oxide-film formation and evolution process of the entrainment defects were different from that of the Mg-alloy surface films previous reported (i.e., MgO formed prior to MgF2).3.

    The growth rate of the oxide film was demonstrated to be greater under SF6/air than SF6/CO2, contributing to a quicker consumption of the damaging entrapped gas. The reproducibility of an AZ91 alloy casting improved when using SF6/air instead of SF6/CO2.

    Acknowledgements

    The authors acknowledge funding from the EPSRC LiME grant EP/H026177/1, and the help from Dr W.D. Griffiths and Mr. Adrian Carden (University of Birmingham). The casting work was carried out in University of Birmingham.

    Reference

    [1]

    M.K. McNutt, SALAZAR K.

    Magnesium, Compounds & Metal, U.S. Geological Survey and U.S. Department of the Interior

    Reston, Virginia (2013)

    Google Scholar[2]

    Magnesium

    Compounds & Metal, U.S. Geological Survey and U.S. Department of the Interior

    (1996)

    Google Scholar[3]

    I. Ostrovsky, Y. Henn

    ASTEC’07 International Conference-New Challenges in Aeronautics, Moscow (2007), pp. 1-5

    Aug 19-22

    View Record in ScopusGoogle Scholar[4]

    Y. Wan, B. Tang, Y. Gao, L. Tang, G. Sha, B. Zhang, N. Liang, C. Liu, S. Jiang, Z. Chen, X. Guo, Y. Zhao

    Acta Mater., 200 (2020), pp. 274-286

    ArticleDownload PDFView Record in Scopus[5]

    J.T.J. Burd, E.A. Moore, H. Ezzat, R. Kirchain, R. Roth

    Appl. Energy, 283 (2021), Article 116269

    ArticleDownload PDFView Record in Scopus[6]

    A.M. Lewis, J.C. Kelly, G.A. Keoleian

    Appl. Energy, 126 (2014), pp. 13-20

    ArticleDownload PDFView Record in Scopus[7]

    J. Campbell

    Castings

    Butterworth-Heinemann, Oxford (2004)

    Google Scholar[8]

    M. Aryafar, R. Raiszadeh, A. Shalbafzadeh

    J. Mater. Sci., 45 (2010), pp. 3041-3051 View PDF

    CrossRefView Record in Scopus[9]

    R. Raiszadeh, W.D. Griffiths

    Metall. Mater. Trans. B-Process Metall. Mater. Process. Sci., 42 (2011), pp. 133-143 View PDF

    CrossRefView Record in Scopus[10]

    R. Raiszadeh, W.D. Griffiths

    J. Alloy. Compd., 491 (2010), pp. 575-580

    ArticleDownload PDFView Record in Scopus[11]

    L. Peng, G. Zeng, T.C. Su, H. Yasuda, K. Nogita, C.M. Gourlay

    JOM, 71 (2019), pp. 2235-2244 View PDF

    CrossRefView Record in Scopus[12]

    S. Ganguly, A.K. Mondal, S. Sarkar, A. Basu, S. Kumar, C. Blawert

    Corros. Sci., 166 (2020)[13]

    G.E. Bozchaloei, N. Varahram, P. Davami, S.K. Kim

    Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 548 (2012), pp. 99-105

    View Record in Scopus[14]

    S. Fox, J. Campbell

    Scr. Mater., 43 (2000), pp. 881-886

    ArticleDownload PDFView Record in Scopus[15]

    M. Cox, R.A. Harding, J. Campbell

    Mater. Sci. Technol., 19 (2003), pp. 613-625

    View Record in Scopus[16]

    C. Nyahumwa, N.R. Green, J. Campbell

    Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 32 (2001), pp. 349-358

    View Record in Scopus[17]

    A. Ardekhani, R. Raiszadeh

    J. Mater. Eng. Perform., 21 (2012), pp. 1352-1362 View PDF

    CrossRefView Record in Scopus[18]

    X. Dai, X. Yang, J. Campbell, J. Wood

    Mater. Sci. Technol., 20 (2004), pp. 505-513

    View Record in Scopus[19]

    E.M. Elgallad, M.F. Ibrahim, H.W. Doty, F.H. Samuel

    Philos. Mag., 98 (2018), pp. 1337-1359 View PDF

    CrossRefView Record in Scopus[20]

    W.D. Griffiths, N.W. Lai

    Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 38A (2007), pp. 190-196 View PDF

    CrossRefView Record in Scopus[21]

    A.R. Mirak, M. Divandari, S.M.A. Boutorabi, J. Campbell

    Int. J. Cast Met. Res., 20 (2007), pp. 215-220 View PDF

    CrossRefView Record in Scopus[22]

    C. Cingi

    Laboratory of Foundry Engineering

    Helsinki University of Technology, Espoo, Finland (2006)

    Google Scholar[23]

    Y. Jia, J. Hou, H. Wang, Q. Le, Q. Lan, X. Chen, L. Bao

    J. Mater. Process. Technol., 278 (2020), Article 116542

    ArticleDownload PDFView Record in Scopus[24]

    S. Ouyang, G. Yang, H. Qin, S. Luo, L. Xiao, W. Jie

    Mater. Sci. Eng. A, 780 (2020), Article 139138

    ArticleDownload PDFView Record in Scopus[25]

    S.-m. Xiong, X.-F. Wang

    Trans. Nonferrous Met. Soc. China, 20 (2010), pp. 1228-1234

    ArticleDownload PDFView Record in Scopus[26]

    G.V. Research

    Grand View Research

    (2018)

    USA

    Google Scholar[27]

    T. Li, J. Davies

    Metall. Mater. Trans. A, 51 (2020), pp. 5389-5400 View PDF

    CrossRefView Record in Scopus[28]J.F. Fruehling, The University of Michigan, 1970.

    Google Scholar[29]

    S. Couling

    36th Annual World Conference on Magnesium, Norway (1979), pp. 54-57

    View Record in ScopusGoogle Scholar[30]

    S. Cashion, N. Ricketts, P. Hayes

    J. Light Met., 2 (2002), pp. 43-47

    ArticleDownload PDFView Record in Scopus[31]

    S. Cashion, N. Ricketts, P. Hayes

    J. Light Met., 2 (2002), pp. 37-42

    ArticleDownload PDFView Record in Scopus[32]

    K. Aarstad, G. Tranell, G. Pettersen, T.A. Engh

    Various Techniques to Study the Surface of Magnesium Protected by SF6

    TMS (2003)

    Google Scholar[33]

    S.-M. Xiong, X.-L. Liu

    Metall. Mater. Trans. A, 38 (2007), pp. 428-434 View PDF

    CrossRefView Record in Scopus[34]

    T.-S. Shih, J.-B. Liu, P.-S. Wei

    Mater. Chem. Phys., 104 (2007), pp. 497-504

    ArticleDownload PDFView Record in Scopus[35]

    G. Pettersen, E. Øvrelid, G. Tranell, J. Fenstad, H. Gjestland

    Mater. Sci. Eng. A, 332 (2002), pp. 285-294

    ArticleDownload PDFView Record in Scopus[36]

    H. Bo, L.B. Liu, Z.P. Jin

    J. Alloy. Compd., 490 (2010), pp. 318-325

    ArticleDownload PDFView Record in Scopus[37]

    A. Mirak, C. Davidson, J. Taylor

    Corros. Sci., 52 (2010), pp. 1992-2000

    ArticleDownload PDFView Record in Scopus[38]

    B.D. Lee, U.H. Beak, K.W. Lee, G.S. Han, J.W. Han

    Mater. Trans., 54 (2013), pp. 66-73 View PDF

    View Record in Scopus[39]

    W.Z. Liang, Q. Gao, F. Chen, H.H. Liu, Z.H. Zhao

    China Foundry, 9 (2012), pp. 226-230 View PDF

    CrossRef[40]

    U.I. Gol’dshleger, E.Y. Shafirovich

    Combust. Explos. Shock Waves, 35 (1999), pp. 637-644[41]

    A. Elsayed, S.L. Sin, E. Vandersluis, J. Hill, S. Ahmad, C. Ravindran, S. Amer Foundry

    Trans. Am. Foundry Soc., 120 (2012), pp. 423-429[42]

    E. Zhang, G.J. Wang, Z.C. Hu

    Mater. Sci. Technol., 26 (2010), pp. 1253-1258

    View Record in Scopus[43]

    N.R. Green, J. Campbell

    Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 173 (1993), pp. 261-266

    ArticleDownload PDFView Record in Scopus[44]

    C Reilly, MR Jolly, NR Green

    Proceedings of MCWASP XII – 12th Modelling of Casting, Welding and Advanced Solidifcation Processes, Vancouver, Canada (2009)

    Google Scholar[45]H.E. Friedrich, B.L. Mordike, Springer, Germany, 2006.

    Google Scholar[46]

    C. Zheng, B.R. Qin, X.B. Lou

    Proceedings of the 2010 International Conference on Mechanical, Industrial, and Manufacturing Technologies, ASME (2010), pp. 383-388

    Mimt 2010 View PDF

    CrossRefView Record in ScopusGoogle Scholar[47]

    S.M. Xiong, X.F. Wang

    Trans. Nonferrous Met. Soc. China, 20 (2010), pp. 1228-1234

    ArticleDownload PDFView Record in Scopus[48]

    S.M. Xiong, X.L. Liu

    Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 38A (2007), pp. 428-434 View PDF

    CrossRefView Record in Scopus[49]

    T.S. Shih, J.B. Liu, P.S. Wei

    Mater. Chem. Phys., 104 (2007), pp. 497-504

    ArticleDownload PDFView Record in Scopus[50]

    K. Aarstad, G. Tranell, G. Pettersen, T.A. Engh

    Magn. Technol. (2003), pp. 5-10[51]

    G. Pettersen, E. Ovrelid, G. Tranell, J. Fenstad, H. Gjestland

    Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 332 (2002), pp. 285-294

    ArticleDownload PDFView Record in Scopus[52]

    X.F. Wang, S.M. Xiong

    Corros. Sci., 66 (2013), pp. 300-307

    ArticleDownload PDFView Record in Scopus[53]

    S.H. Nie, S.M. Xiong, B.C. Liu

    Mater. Sci. Eng. A-Struct. Mater. Prop. Microstruct. Process., 422 (2006), pp. 346-351

    ArticleDownload PDFView Record in Scopus[54]

    C. Bauer, A. Mogessie, U. Galovsky

    Zeitschrift Fur Metallkunde, 97 (2006), pp. 164-168 View PDF

    CrossRef[55]

    Q.G. Wang, D. Apelian, D.A. Lados

    J. Light Met., 1 (2001), pp. 73-84

    ArticleDownload PDFView Record in Scopus[56]

    S. Wang, Y. Wang, Q. Ramasse, Z. Fan

    Metall. Mater. Trans. A, 51 (2020), pp. 2957-2974[57]

    S. Hayashi, W. Minami, T. Oguchi, H.J. Kim

    Kag. Kog. Ronbunshu, 35 (2009), pp. 411-415 View PDF

    CrossRefView Record in Scopus[58]

    K. Aarstad

    Norwegian University of Science and Technology

    (2004)

    Google Scholar[59]

    R.L. Wilkins

    J. Chem. Phys., 51 (1969), p. 853

    -&

    View Record in Scopus[60]

    O. Kubaschewski, K. Hesselemam

    Thermo-Chemical Properties of Inorganic Substances

    Springer-Verlag, Belin (1991)

    Google Scholar[61]

    R. Schmidt, M. Strobele, K. Eichele, H.J. Meyer

    Eur. J. Inorg. Chem. (2017), pp. 2727-2735 View PDF

    CrossRefView Record in Scopus[62]

    B. Hu, Y. Du, H. Xu, W. Sun, W.W. Zhang, D. Zhao

    J. Min. Metall. Sect. B-Metall., 46 (2010), pp. 97-103

    View Record in Scopus[63]

    O. Salas, H. Ni, V. Jayaram, K.C. Vlach, C.G. Levi, R. Mehrabian

    J. Mater. Res., 6 (1991), pp. 1964-1981

    View Record in Scopus[64]

    S.S.S. Kumari, U.T.S. Pillai, B.C. Pai

    J. Alloy. Compd., 509 (2011), pp. 2503-2509

    ArticleDownload PDFView Record in Scopus[65]

    H. Scholz, P. Greil

    J. Mater. Sci., 26 (1991), pp. 669-677

    View Record in Scopus[66]

    P. Biedenkopf, A. Karger, M. Laukotter, W. Schneider

    Magn. Technol., 2005 (2005), pp. 39-42

    View Record in Scopus[67]

    H.V. Atkinson, S. Davies

    Metall. Mater. Trans. A, 31 (2000), pp. 2981-3000 View PDF

    CrossRefView Record in Scopus[68]

    E.J. Guo, L. Wang, Y.C. Feng, L.P. Wang, Y.H. Chen

    J. Therm. Anal. Calorim., 135 (2019), pp. 2001-2008 View PDF

    CrossRefView Record in Scopus[69]

    T. Li, W.D. Griffiths, J. Chen

    Metall. Mater. Trans. A-Phys. Metall. Mater. Sci., 48A (2017), pp. 5516-5528 View PDF

    CrossRefView Record in Scopus[70]

    M. Tiryakioglu, D. Hudak

    J. Mater. Sci., 42 (2007), pp. 10173-10179 View PDF

    CrossRefView Record in Scopus[71]

    Y. Yue, W.D. Griffiths, J.L. Fife, N.R. Green

    Proceedings of the 1st International Conference on 3d Materials Science (2012), pp. 131-136 View PDF

    CrossRefView Record in ScopusGoogle Scholar[72]

    R. Raiszadeh, W.D. Griffiths

    Metall. Mater. Trans. B-Process Metall. Mater. Process. Sci., 37 (2006), pp. 865-871

    View Record in Scopus[73]

    Z.C. Hu, E.L. Zhang, S.Y. Zeng

    Mater. Sci. Technol., 24 (2008), pp. 1304-1308 View PDF

    CrossRefView Record in Scopus

    Fig. 1. Schematic of lap welding for 6061/5182 aluminum alloys.

    알루미늄 합금 겹침 용접 중 용접 형성, 용융 흐름 및 입자 구조에 대한 사인파 발진 레이저 빔의 영향

    린 첸 가오 양 미시 옹 장 춘밍 왕
    Lin Chen , Gaoyang Mi , Xiong Zhang , Chunming Wang *
    중국 우한시 화중과학기술대학 재료공학부, 430074

    Effects of sinusoidal oscillating laser beam on weld formation, melt flow and grain structure during aluminum alloys lap welding

    Abstract

    A numerical model of 1.5 mm 6061/5182 aluminum alloys thin sheets lap joints under laser sinusoidal oscillation (sine) welding and laser welding (SLW) weld was developed to simulate temperature distribution and melt flow. Unlike the common energy distribution of SLW, the sinusoidal oscillation of laser beam greatly homogenized the energy distribution and reduced the energy peak. The energy peaks were located at both sides of the sine weld, resulting in the tooth-shaped sectional formation. This paper illustrated the effect of the temperature gradient (G) and solidification rate (R) on the solidification microstructure by simulation. Results indicated that the center of the sine weld had a wider area with low G/R, promoting the formation of a wider equiaxed grain zone, and the columnar grains were slenderer because of greater GR. The porosity-free and non-penetration welds were obtained by the laser sinusoidal oscillation. The reasons were that the molten pool volume was enlarged, the volume proportion of keyhole was reduced and the turbulence in the molten pool was gentled, which was observed by the high-speed imaging and simulation results of melt flow. The tensile test of both welds showed a tensile fracture form along the fusion line, and the tensile strength of sine weld was significantly better than that of the SLW weld. This was because that the wider equiaxed grain area reduced the tendency of cracks and the finer grain size close to the fracture location. Defect-free and excellent welds are of great significance to the new energy vehicles industry.

    온도 분포 및 용융 흐름을 시뮬레이션하기 위해 레이저 사인파 진동 (사인) 용접 및 레이저 용접 (SLW) 용접에서 1.5mm 6061/5182 알루미늄 합금 박판 랩 조인트 의 수치 모델이 개발되었습니다. SLW의 일반적인 에너지 분포와 달리 레이저 빔의 사인파 진동은 에너지 분포를 크게 균질화하고 에너지 피크를 줄였습니다. 에너지 피크는 사인 용접의 양쪽에 위치하여 톱니 모양의 단면이 형성되었습니다. 이 논문은 온도 구배(G)와 응고 속도 의 영향을 설명했습니다.(R) 시뮬레이션에 의한 응고 미세 구조. 결과는 사인 용접의 중심이 낮은 G/R로 더 넓은 영역을 가짐으로써 더 넓은 등축 결정립 영역의 형성을 촉진하고 더 큰 GR로 인해 주상 결정립 이 더 가늘다는 것을 나타냅니다. 다공성 및 비관통 용접은 레이저 사인파 진동에 의해 얻어졌습니다. 그 이유는 용융 풀의 부피가 확대되고 열쇠 구멍의 부피 비율이 감소하며 용융 풀의 난류가 완만해졌기 때문이며, 이는 용융 흐름의 고속 이미징 및 시뮬레이션 결과에서 관찰되었습니다. 두 용접부 의 인장시험 은 융착선을 따라 인장파괴형태를인장강도사인 용접의 경우 SLW 용접보다 훨씬 우수했습니다. 이는 등축 결정립 영역이 넓을수록 균열 경향이 감소하고 파단 위치에 근접한 입자 크기가 미세 하기 때문입니다. 결함이 없고 우수한 용접은 신에너지 자동차 산업에 매우 중요합니다.

    Fig. 1. Schematic of lap welding for 6061/5182 aluminum alloys.
    Fig. 1. Schematic of lap welding for 6061/5182 aluminum alloys.
    Fig. 2. Finite element mesh.
    Fig. 2. Finite element mesh.
    Fig. 3. Weld morphologies of cross-section and upper surface for the two welds: (a) sine pattern weld; (b) SLW weld.
    Fig. 3. Weld morphologies of cross-section and upper surface for the two welds: (a) sine pattern weld; (b) SLW weld.
    Fig. 4. Calculation of laser energy distribution: (a)-(c) sine pattern weld; (d)-(f) SLW weld.
    Fig. 4. Calculation of laser energy distribution: (a)-(c) sine pattern weld; (d)-(f) SLW weld.
    Fig. 5. The partially melted region of zone A.
    Fig. 5. The partially melted region of zone A.
    Fig. 6. The simulated profiles of melted region for the two welds: (a) SLW weld; (b) sine pattern weld.
    Fig. 6. The simulated profiles of melted region for the two welds: (a) SLW weld; (b) sine pattern weld.
    Fig. 7. The temperature field simulation results of cross section for sine pattern weld.
    Fig. 7. The temperature field simulation results of cross section for sine pattern weld.
    Fig. 8. Dynamic behavior of the molten pool at the same time interval of 0.004 s within one oscillating period: (a) SLW weld; (b) sine pattern weld.
    Fig. 8. Dynamic behavior of the molten pool at the same time interval of 0.004 s within one oscillating period: (a) SLW weld; (b) sine pattern weld.
    Fig. 9. The temperature field and flow field of the molten pool for the SLW weld: (a)~(f) t = 80 ms~100 ms.
    Fig. 9. The temperature field and flow field of the molten pool for the SLW weld: (a)~(f) t = 80 ms~100 ms.
    Fig. 10. The temperature field and flow field of the molten pool for the sine pattern weld: (a)~(f) t = 151 ms~171 ms.
    Fig. 10. The temperature field and flow field of the molten pool for the sine pattern weld: (a)~(f) t = 151 ms~171 ms.
    Fig. 11. The evolution of the molten pool volume and keyhole depth within one period.
    Fig. 11. The evolution of the molten pool volume and keyhole depth within one period.
    Fig. 12. The X-ray inspection results for the two welds: (a) SLW weld, (b) sine pattern weld.
    Fig. 12. The X-ray inspection results for the two welds: (a) SLW weld, (b) sine pattern weld.
    Fig. 13. Comparison of the solidification parameters for sine and SLW patterns: (a) the temperature field simulated results of the molten pool upper surfaces; (b) temperature gradient G and solidification rate R along the molten pool boundary isotherm from weld centerline to the fusion boundary; (c) G/R; (d) GR.
    Fig. 13. Comparison of the solidification parameters for sine and SLW patterns: (a) the temperature field simulated results of the molten pool upper surfaces; (b) temperature gradient G and solidification rate R along the molten pool boundary isotherm from weld centerline to the fusion boundary; (c) G/R; (d) GR.
    Fig. 14. The EBSD results of equiaxed grain zone in the weld center of: (a) sine pattern weld; (b) SLW weld; (c) grain size.
    Fig. 14. The EBSD results of equiaxed grain zone in the weld center of: (a) sine pattern weld; (b) SLW weld; (c) grain size.
    Fig. 15. (a) EBSD results of horizontal sections of SLW weld and sine pattern weld; (b) The columnar crystal widths of SLW weld and sine pattern weld.
    Fig. 15. (a) EBSD results of horizontal sections of SLW weld and sine pattern weld; (b) The columnar crystal widths of SLW weld and sine pattern weld.
    Fig. 16. (a) The tensile test results of the two welds; (b) Fracture location of SLW weld; (b) Fracture location of sine pattern weld.
    Fig. 16. (a) The tensile test results of the two welds; (b) Fracture location of SLW weld; (b) Fracture location of sine pattern weld.

    Keywords

    Laser welding, Sinusoidal oscillating, Energy distribution, Numerical simulation, Molten pool flow, Grain structure

    References

    Chen, X., 2014. Study on laser-MAG Hybrid Weaving Welding Charateristics. Master
    thesis. Harbin Institute of Technology, China.
    Chen, G., Wang, B., Mao, S., Zhong, P., He, J., 2019. Research on the “∞”-shaped laser
    scanning welding process for aluminum alloy. Opt. Laser Technol. 115, 32–41.
    Cho, W.-I., Na, S.-J., Cho, M.-H., Lee, J.-S., 2010. Numerical study of alloying element
    distribution in CO2 laser–GMA hybrid welding. Comput. Mater. Sci. 49, 792–800.
    Cho, W.-I., Na, S.-J., Thomy, C., Vollertsen, F., 2012. Numerical simulation of molten
    pool dynamics in high power disk laser welding. J. Mater. Process. Technol. 212,
    262–275.
    Das, A., Butterworth, I., Masters, I., Williams, D., 2018. Microstructure and mechanical
    properties of gap-bridged remote laser welded (RLW) automotive grade AA 5182
    joints. Mater. Charact. 145, 697–712.
    Fetzer, F., Sommer, M., Weber, R., Weberpals, J.-P., Graf, T., 2018. Reduction of pores by
    means of laser beam oscillation during remote welding of AlMgSi. Opt. Lasers Eng.
    108, 68–77.
    Geng, S., Jiang, P., Shao, X., Guo, L., Gao, X., 2020. Heat transfer and fluid flow and their
    effects on the solidification microstructure in full-penetration laser welding of
    aluminum sheet. J. Mater. Sci. Technol. 46, 50–63.
    Hagenlocher, C., Sommer, M., Fetzer, F., Weber, R., Graf, T., 2018a. Optimization of the
    solidification conditions by means of beam oscillation during laser beam welding of
    aluminum. Mater. Des. 160, 1178–1185.
    Hagenlocher, C., Weller, D., Weber, R., Graf, T., 2018b. Reduction of the hot cracking
    susceptibility of laser beam welds in AlMgSi alloys by increasing the number of grain
    boundaries. Sci. Technol. Weld. Join. 24, 313–319.
    Hagenlocher, C., Fetzer, F., Weller, D., Weber, R., Graf, T., 2019. Explicit analytical
    expressions for the influence of welding parameters on the grain structure of laser
    beam welds in aluminium alloys. Mater. Des. 174, 107791.
    Han, X., Tang, X., Wang, T., Shao, C., Lu, F., Cui, H., 2018. Role of ambient pressure in
    keyhole dynamics based on beam transmission path method for laser welding on Al
    alloy. Int. J. Adv. Manuf. Technol. 99, 1639–1651.
    Hao, K., Li, G., Gao, M., Zeng, X., 2015. Weld formation mechanism of fiber laser
    oscillating welding of austenitic stainless steel. J. Mater. Process. Technol. 225,
    77–83.
    Hirt, C.W., Nichols, B.D., 1981. Volume of fluid (VOF) method for the dynamics of free
    boundaries. J. Comput. Phys. 39, 201–225.
    Jiang, Z., Chen, X., Li, H., Lei, Z., Chen, Y., Wu, S., Wang, Y., 2020. Grain refinement and
    laser energy distribution during laser oscillating welding of Invar alloy. Mater. Des.
    186, 108195.
    Kaplan, A., 1994. A model of deep penetration laser welding based on calculation of the
    keyhole profile. J. Phys. D Appl. Phys. 27, 1805–1814.
    Kou, S., 2002. Welding Metallurgy, 2nd ed. Wiley-Interscience, New Jersey, USA.
    Kuryntsev, S.V., Gilmutdinov, A.K., 2015. The effect of laser beam wobbling mode in
    welding process for structural steels. Int. J. Adv. Manuf. Technol. 81, 1683–1691.
    Li, P., Nie, F., Dong, H., Li, S., Yang, G., Zhang, H., 2018. Pulse MIG welding of 6061-T6/
    A356-T6 aluminum alloy dissimilar T-joint. J. Mater. Eng. Perform. 27, 4760–4769.
    Liu, T., Mu, Z., Hu, R., Pang, S., 2019. Sinusoidal oscillating laser welding of 7075
    aluminum alloy: hydrodynamics, porosity formation and optimization. Int. J. Heat
    Mass Transf. 140, 346–358.
    Seto, N., Katayama, S., Matsunawa, A., 2000. High-speed simultaneous observation of
    plasma and keyhole behavior during high power CO2 laser welding: effect of
    shielding gas on porosity formation. J. Laser Appl. 12, 245–250.
    Tang, Z., Vollertsen, F., 2014. Influence of grain refinement on hot cracking in laser
    welding of aluminum. Weld. World 58, 355–366.
    Wang, L., Gao, M., Zhang, C., Zeng, X., 2016. Effect of beam oscillating pattern on weld
    characterization of laser welding of AA6061-T6 aluminum alloy. Mater. Des. 108,
    707–717.
    Wang, L., Gao, M., Zeng, X., 2018. Experiment and prediction of weld morphology for
    laser oscillating welding of AA6061 aluminium alloy. Sci. Technol. Weld. Join. 24,
    334–341.
    Yamazaki, Y., Abe, Y., Hioki, Y., Nakatani, M., Kitagawa, A., Nakata, K., 2016.
    Fundamental study of narrow-gap welding with oscillation laser beam. Weld. Int. 30,
    699–707.
    Yuan, Z., Tu, Y., Yuan, T., Zhang, Y., Huang, Y., 2021. Size effects on mechanical
    properties of pure industrial aluminum sheet for micro/meso scale plastic
    deformation: experiment and modeling. J. Alloys. Compd. 859, 157752.
    Zou, J., 2016. Characteristics of laser scanning welding process for 5A06 aluminum alloy
    thick plate with narrow gap. Materials Processing Engineering. Harbin Welding
    Institute, China. Master thesis.

    Figure 3. Comparison of water surface profiles over porous media with 12 mm particle diameter in laboratory measurements (symbols) and numerical results (lines).

    다공층에 대한 돌발 댐 붕괴의 3차원 유동 수치해석 시뮬레이션

    A. Safarzadeh1*, P. Mohsenzadeh2, S. Abbasi3
    1 Professor of Civil Eng., Water Engineering and Mineral Waters Research Center, Univ. of Mohaghegh Ardabili,Ardabil, Iran
    2 M.Sc., Graduated of Civil-Hydraulic Structures Eng., Faculty of Eng., Univ. of Mohaghegh Ardabili, Ardabil, Iran
    3 M.Sc., Graduated of Civil -Hydraulic Structures Eng., Faculty of Eng., Univ. of Mohaghegh Ardabili, Ardabil, Iran Safarzadeh@uma.ac.ir

    Highlights

    유체 이동에 의해 생성된 RBF는 Ls-Dyna에서 Fluent, ICFD ALE 및 SPH 방법으로 시뮬레이션되었습니다.
    RBF의 과예측은 유체가 메인 도메인에서 고속으로 분리될 때 발생합니다.
    이 과잉 예측은 요소 크기, 시간 단계 크기 및 유체 모델에 따라 다릅니다.
    유체 성능을 검증하려면 최대 RBF보다 임펄스가 권장됩니다.

    Abstract

    Dam break is a very important problem due to its effects on economy, security, human casualties and environmental consequences. In this study, 3D flow due to dam break over the porous substrate is numerically simulated and the effect of porosity, permeability and thickness of the porous bed and the water depth in the porous substrate are investigated. Classic models of dam break over a rigid bed and water infiltration through porous media were studied and results of the numerical simulations are compared with existing laboratory data. Validation of the results is performed by comparing the water surface profiles and wave front position with dam break on rigid and porous bed. Results showed that, due to the effect of dynamic wave in the initial stage of dam break, a local peak occurs in the flood hydrograph. The presence of porous bed reduces the acceleration of the flood wave relative to the flow over the solid bed and it decreases with the increase of the permeability of the bed. By increasing the permeability of the bed, the slope of the ascending limb of the flood hydrograph and the peak discharge drops. Furthermore, if the depth and permeability of the bed is such that the intrusive flow reaches the rigid substrate under the porous bed, saturation of the porous bed, results in a sharp increase in the slope of the flood hydrograph. The maximum values of the peak discharge at the end of the channel with porous bed occurred in saturated porous bed conditions.

    댐 붕괴는 경제, 보안, 인명 피해 및 환경적 영향으로 인해 매우 중요한 문제입니다. 본 연구에서는 다공성 기재에 대한 댐 파괴로 인한 3차원 유동을 수치적으로 시뮬레이션하고 다공성 기재의 다공성, 투과도 및 다공성 층의 두께 및 수심의 영향을 조사합니다. 단단한 바닥에 대한 댐 파괴 및 다공성 매체를 통한 물 침투의 고전 모델을 연구하고 수치 시뮬레이션 결과를 기존 실험실 데이터와 비교합니다. 결과 검증은 강체 및 다공성 베드에서 댐 파단과 수면 프로파일 및 파면 위치를 비교하여 수행됩니다. 그 결과 댐파괴 초기의 동적파동의 영향으로 홍수수문곡선에서 국부첨두가 발생하는 것으로 나타났다. 다공성 베드의 존재는 고체 베드 위의 유동에 대한 홍수파의 가속을 감소시키고 베드의 투과성이 증가함에 따라 감소합니다. 베드의 투수성을 증가시켜 홍수 수문곡선의 오름차순 경사와 첨두방류량이 감소한다. 더욱이, 만약 층의 깊이와 투과성이 관입 유동이 다공성 층 아래의 단단한 기질에 도달하는 정도라면, 다공성 층의 포화는 홍수 수문곡선의 기울기의 급격한 증가를 초래합니다. 다공층이 있는 채널의 끝단에서 최대 방전 피크값은 포화 다공층 조건에서 발생하였다.

    Keywords

    Keywords: Dams Break, 3D modeling, Porous Bed, Permeability, Flood wave

    Reference

    [1] D.L. Fread, In: Maidment, D.R. (Ed.), Flow Routing in Handbook of Hydrology, McGraw-Hill Inc., New York, USA, pp. 10(1) (1993) 1-36.
    [2] M. Morris, CADAM: Concerted Action on Dambreak Modeling – Final Report, Rep. SR 571. HR Wallingford, 2000.
    [3] H. Chanson, The Hydraulics of Open Channel Flows: an Introduction, ButterworthHeinemann, Oxford, 2004.
    [4] A. Ritter, Die Fortpflanzung der Wasserwellen (The Propagation of Water Waves), Zeitschrift Verein Deutscher Ingenieure, 36 (33) (1892) 947–954 [in German].
    [5] B. Ghimire, Hydraulic Analysis of Free-Surface Flows into Highly Permeable Porous Media and its Applications, Phd. Thesis, Kyoto University, 2009.
    [6] R. Dressler, Hydraulic Resistance Effect Upon the Dam-Break Function, Journal of Research of the National Bureau of Standards, 49 (3) 1952.
    [7] G. Lauber, and W.H. Hager, Experiments to Dambreak Wave: horizontal channel, Journal of Hydraulic Research. 36 (3) (1998) 291–307.
    [8] L.W. Tan, and V.H. Chu, Lagrangian Block Hydrodynamics of Macro Resistance in a River-Flow Model,
    [9] L. Tan, V.H. Lauber and Hager’s Dam-Break Wave Data for Numerical Model Validation, Journal of Hydraulic Research, 47 (4) (2009) 524-528.
    [10] S. Mambretti, E.D. Larcan, and D. Wrachien, 1D Modelling of Dam-Break Surges with Floating Debris, J. of Biosystems engineering, 100 (2) (2008) 297-308.
    [11] M. Pilotti, M. Tomirotti, G. Valerio, and B. Bacchi, Simplified Method for the Characterization of the Hydrograph Following a Sudden Partial Dam Break, Journal of Hydraulic Engineering, 136 (10) (2010) 693-704.
    [12] T.J. Chang, H.M. Kao, K.H. Chang, and Mi.H. Hsu, Numerical Simulation of ShallowWater Dam Break Flows in Open Channels Using Smoothed Particle Hydrodynamics, J. Hydraul. Eng., 408 (78–90) 2011.
    [13] T. Tawatchai, and W. Rattanapitikon, 2-D Modelling of Dambreak Wave Propagation on Initially Dry Bed, Thammasat Int. J. Sc. 4 (3) 1999.
    [14] Y.F. Le, Experimental Study of landslide Dam-Break Flood over Erodible Bed in open Channels. Journal of Hydrodynamics, Ser. B, 21 (5) 2006.
    [15] O. Castro-Orgaz, & H. Chanson, Ritter’s Dry-Bed Dam-Break Flows: Positive and Negative Wave Dynamics, J. of Environmental Fluid Mechanics, 17 (4) (2017) 665-694.
    [16] A. Jozdani, A.R. Kabiri-Samani, Application of Image Processing Method to Analysis of Flood Behavior Due to Dam Break, 9th Iranian Hydraulic Conference. Univ. of Tarbiat Moddares, 2011.(in persian)
    [17] A. Safarzadeh, Three Dimensional Hydrodynamics of Sudden Dam Break in Curved Channels, Journal of Modares Civil Engineering, 17(3) (2017) 77-86. (in persian)
    [18] P. C. Carman, Fluid Flow Through Granular Beds, Transactions, Institution of Chem. Eng. Res. Des. 75 (Dec): S32–S48, London, 15, (1937) 150-166.
    [19] P. Forchheimer, Wasserbewegung Durch Boden. Z. Ver. Deutsch. Ing. 45 (1901) 1782– 1788.
    [20] S. Ergun, Fluid Flow through Packed Columns. Chemical Engineering Progress, 48(2) (1952) 89-93.
    [21] A. Parsaei, S. Dehdar-Behbahani, Numerical Modeling of Cavitation on Spillway’s Flip Bucket, Frontiers of Structural and Civil Engineering, 10 (4) (2016) 438-444.
    [22] S. Dehdar-Behbahani, A. Parsaei, Numerical Modeling of Flow Pattern in Dam Spillway’s Guide Wall. Case study: Balaroud dam, Iran, Alexandria Engineering Journal, 55(1) (2016) 467-473.
    [23] A. Parsaei, AH. Haghiabi, A. Moradnejad, CFD Modeling of Flow Pattern in Spillway’s ACCEPTED MANUSCRIPT 19 Approach Channel, Sustainable Water Resources Management, 1(3) (2015) 245-251.
    [24] SH. Najafian, H. Yonesi, A. Parsaei, PH. Torabi, Physical and Numerical Modeling of Flow in Heterogeneous Roughness Non-Prismatic Compound Open Channel, Irrigation and Drainage Structures Engineering Research, 17(66) (2016) 87-104.
    [25] SH. Najafian, H. Yonesi, A. Parsaei, PH. Torabi, Physical and Numerical Modeling of Flow Properties in Prismatic Compound Open Channel with Heterogeneous Roughness, Irrigation and Drainage Structures Engineering Research, 18(68) (2017) 1-16.
    [26] A. Safarzadeh, S.H. Mohajeri, Hydrodynamics of Rectangular Broad-Crested Porous Weirs, Journal of Irrig. & Drain. Eng., 144(10) (2018) 1-12.
    [27] M. Fathi-moghaddam, M.T. Sadrabadi, M, Rahamnshahi, Numerical Simulation of the Hydraulic Performance of Triangular and Trapezoidal Gabion Weirs in Free Flow Condition, Journal of Flow Measurement & Instrumentation, 62 (2018) 93-104.
    [28] A. Parsaei, A. Moradnejad, Numerical Modeling of Flow Pattern in Spillway Approach Channel, Jordan Journal of Civil Engineering, 12(1) (2018) 1-9.

    Fig. 1 Multi-physics phenomena in the laser-material interaction zone

    COMPARISON BETWEEN GREEN AND
    INFRARED LASER IN LASER POWDER BED
    FUSION OF PURE COPPER THROUGH HIGH
    FIDELITY NUMERICAL MODELLING AT MESOSCALE

    316-L 스테인리스강의 레이저 분말 베드 융합 중 콜드 스패터 형성의 충실도 높은 수치 모델링

    W.E. ALPHONSO1*, M. BAYAT1 and J.H. HATTEL1
    *Corresponding author
    1Technical University of Denmark (DTU), 2800, Kgs, Lyngby, Denmark

    ABSTRACT

    L-PBF(Laser Powder Bed Fusion)는 금속 적층 제조(MAM) 기술로, 기존 제조 공정에 비해 부품 설계 자유도, 조립품 통합, 부품 맞춤화 및 낮은 툴링 비용과 같은 여러 이점을 산업에 제공합니다.

    전기 코일 및 열 관리 장치는 일반적으로 높은 전기 및 열 전도성 특성으로 인해 순수 구리로 제조됩니다. 따라서 순동의 L-PBF가 가능하다면 기하학적으로 최적화된 방열판과 자유형 전자코일을 제작할 수 있습니다.

    그러나 L-PBF로 조밀한 순동 부품을 생산하는 것은 적외선에 대한 낮은 광 흡수율과 높은 열전도율로 인해 어렵습니다. 기존의 L-PBF 시스템에서 조밀한 구리 부품을 생산하려면 적외선 레이저의 출력을 500W 이상으로 높이거나 구리의 광흡수율이 높은 녹색 레이저를 사용해야 합니다.

    적외선 레이저 출력을 높이면 후면 반사로 인해 레이저 시스템의 광학 구성 요소가 손상되고 렌즈의 열 광학 현상으로 인해 공정이 불안정해질 수 있습니다. 이 작업에서 FVM(Finite Volume Method)에 기반한 다중 물리학 중간 규모 수치 모델은 Flow-3D에서 개발되어 용융 풀 역학과 궁극적으로 부품 품질을 제어하는 ​​물리적 현상 상호 작용을 조사합니다.

    녹색 레이저 열원과 적외선 레이저 열원은 기판 위의 순수 구리 분말 베드에 단일 트랙 증착을 생성하기 위해 개별적으로 사용됩니다.

    용융 풀 역학에 대한 레이저 열원의 유사하지 않은 광학 흡수 특성의 영향이 탐구됩니다. 수치 모델을 검증하기 위해 단일 트랙이 구리 분말 베드에 증착되고 시뮬레이션된 용융 풀 모양과 크기가 비교되는 실험이 수행되었습니다.

    녹색 레이저는 광흡수율이 높아 전도 및 키홀 모드 용융이 가능하고 적외선 레이저는 흡수율이 낮아 키홀 모드 용융만 가능하다. 레이저 파장에 대한 용융 모드의 변화는 궁극적으로 기계적, 전기적 및 열적 특성에 영향을 미치는 열 구배 및 냉각 속도에 대한 결과를 가져옵니다.

    Laser Powder Bed Fusion (L-PBF) is a Metal Additive Manufacturing (MAM) technology which offers several advantages to industries such as part design freedom, consolidation of assemblies, part customization and low tooling cost over conventional manufacturing processes. Electric coils and thermal management devices are generally manufactured from pure copper due to its high electrical and thermal conductivity properties. Therefore, if L-PBF of pure copper is feasible, geometrically optimized heat sinks and free-form electromagnetic coils can be manufactured. However, producing dense pure copper parts by L-PBF is difficult due to low optical absorptivity to infrared radiation and high thermal conductivity. To produce dense copper parts in a conventional L-PBF system either the power of the infrared laser must be increased above 500W, or a green laser should be used for which copper has a high optical absorptivity. Increasing the infrared laser power can damage the optical components of the laser systems due to back reflections and create instabilities in the process due to thermal-optical phenomenon of the lenses. In this work, a multi-physics meso-scale numerical model based on Finite Volume Method (FVM) is developed in Flow-3D to investigate the physical phenomena interaction which governs the melt pool dynamics and ultimately the part quality. A green laser heat source and an infrared laser heat source are used individually to create single track deposition on pure copper powder bed above a substrate. The effect of the dissimilar optical absorptivity property of laser heat sources on the melt pool dynamics is explored. To validate the numerical model, experiments were conducted wherein single tracks are deposited on a copper powder bed and the simulated melt pool shape and size are compared. As the green laser has a high optical absorptivity, a conduction and keyhole mode melting is possible while for the infrared laser only keyhole mode melting is possible due to low absorptivity. The variation in melting modes with respect to the laser wavelength has an outcome on thermal gradient and cooling rates which ultimately affect the mechanical, electrical, and thermal properties.

    Keywords

    Pure Copper, Laser Powder Bed Fusion, Finite Volume Method, multi-physics

    Fig. 1 Multi-physics phenomena in the laser-material interaction zone
    Fig. 1 Multi-physics phenomena in the laser-material interaction zone
    Fig. 2 Framework for single laser track simulation model including powder bed and substrate (a) computational domain with boundaries (b) discretization of the domain with uniform quad mesh.
    Fig. 2 Framework for single laser track simulation model including powder bed and substrate (a) computational domain with boundaries (b) discretization of the domain with uniform quad mesh.
    Fig. 3 2D melt pool contours from the numerical model compared to experiments [16] for (a) VED = 65 J/mm3 at 7 mm from the beginning of the single track (b) VED = 103 J/mm3 at 3 mm from the beginning of the single track (c) VED = 103 J/mm3 at 7 mm from the beginning of the single track. In the 2D contour, the non-melted region is indicated in blue, and the melted region is indicated by red and green when the VED is 65 J/mm3 and 103 J/mm3 respectively.
    Fig. 3 2D melt pool contours from the numerical model compared to experiments [16] for (a) VED = 65 J/mm3 at 7 mm from the beginning of the single track (b) VED = 103 J/mm3 at 3 mm from the beginning of the single track (c) VED = 103 J/mm3 at 7 mm from the beginning of the single track. In the 2D contour, the non-melted region is indicated in blue, and the melted region is indicated by red and green when the VED is 65 J/mm3 and 103 J/mm3 respectively.
    Fig. 4 3D temperature contour plots of during single track L-PBF process at time1.8 µs when (a) VED = 65 J/mm3 (b) VED = 103 J/mm3 along with 2D melt pool contours at 5 mm from the laser initial position. In the 2D contour, the non-melted region is indicated in blue, and the melted region is indicated by red and green when the VED is 65 J/mm3 and 103 J/mm3 respectively.
    Fig. 4 3D temperature contour plots of during single track L-PBF process at time1.8 µs when (a) VED = 65 J/mm3 (b) VED = 103 J/mm3 along with 2D melt pool contours at 5 mm from the laser initial position. In the 2D contour, the non-melted region is indicated in blue, and the melted region is indicated by red and green when the VED is 65 J/mm3 and 103 J/mm3 respectively.

    References

    [1] L. Jyothish Kumar, P. M. Pandey, and D. I. Wimpenny, 3D printing and additive
    manufacturing technologies. Springer Singapore, 2018. doi: 10.1007/978-981-13-0305-0.
    [2] T. DebRoy et al., “Additive manufacturing of metallic components – Process, structure
    and properties,” Progress in Materials Science, vol. 92, pp. 112–224, 2018, doi:
    10.1016/j.pmatsci.2017.10.001.
    [3] C. S. Lefky, B. Zucker, D. Wright, A. R. Nassar, T. W. Simpson, and O. J. Hildreth,
    “Dissolvable Supports in Powder Bed Fusion-Printed Stainless Steel,” 3D Printing and
    Additive Manufacturing, vol. 4, no. 1, pp. 3–11, 2017, doi: 10.1089/3dp.2016.0043.
    [4] J. L. Bartlett and X. Li, “An overview of residual stresses in metal powder bed fusion,”
    Additive Manufacturing, vol. 27, no. January, pp. 131–149, 2019, doi:
    10.1016/j.addma.2019.02.020.
    [5] I. H. Ahn, “Determination of a process window with consideration of effective layer
    thickness in SLM process,” International Journal of Advanced Manufacturing
    Technology, vol. 105, no. 10, pp. 4181–4191, 2019, doi: 10.1007/s00170-019-04402-w.

    [6] R. McCann et al., “In-situ sensing, process monitoring and machine control in Laser
    Powder Bed Fusion: A review,” Additive Manufacturing, vol. 45, no. May, 2021, doi:
    10.1016/j.addma.2021.102058.
    [7] M. Bayat et al., “Keyhole-induced porosities in Laser-based Powder Bed Fusion (L-PBF)
    of Ti6Al4V: High-fidelity modelling and experimental validation,” Additive
    Manufacturing, vol. 30, no. August, p. 100835, 2019, doi: 10.1016/j.addma.2019.100835.
    [8] M. Bayat, S. Mohanty, and J. H. Hattel, “Multiphysics modelling of lack-of-fusion voids
    formation and evolution in IN718 made by multi-track/multi-layer L-PBF,” International
    Journal of Heat and Mass Transfer, vol. 139, pp. 95–114, 2019, doi:
    10.1016/j.ijheatmasstransfer.2019.05.003.
    [9] S. D. Jadhav, L. R. Goossens, Y. Kinds, B. van Hooreweder, and K. Vanmeensel, “Laserbased powder bed fusion additive manufacturing of pure copper,” Additive Manufacturing,
    vol. 42, no. March, 2021, doi: 10.1016/j.addma.2021.101990.
    [10] S. D. Jadhav, S. Dadbakhsh, L. Goossens, J. P. Kruth, J. van Humbeeck, and K.
    Vanmeensel, “Influence of selective laser melting process parameters on texture evolution
    in pure copper,” Journal of Materials Processing Technology, vol. 270, no. January, pp.
    47–58, 2019, doi: 10.1016/j.jmatprotec.2019.02.022.
    [11] H. Siva Prasad, F. Brueckner, J. Volpp, and A. F. H. Kaplan, “Laser metal deposition of
    copper on diverse metals using green laser sources,” International Journal of Advanced
    Manufacturing Technology, vol. 107, no. 3–4, pp. 1559–1568, 2020, doi: 10.1007/s00170-
    020-05117-z.
    [12] L. R. Goossens, Y. Kinds, J. P. Kruth, and B. van Hooreweder, “On the influence of
    thermal lensing during selective laser melting,” Solid Freeform Fabrication 2018:
    Proceedings of the 29th Annual International Solid Freeform Fabrication Symposium – An
    Additive Manufacturing Conference, SFF 2018, no. December, pp. 2267–2274, 2020.
    [13] M. Bayat, V. K. Nadimpalli, D. B. Pedersen, and J. H. Hattel, “A fundamental investigation
    of thermo-capillarity in laser powder bed fusion of metals and alloys,” International
    Journal of Heat and Mass Transfer, vol. 166, p. 120766, 2021, doi:
    10.1016/j.ijheatmasstransfer.2020.120766.
    [14] H. Chen, Q. Wei, Y. Zhang, F. Chen, Y. Shi, and W. Yan, “Powder-spreading mechanisms
    in powder-bed-based additive manufacturing: Experiments and computational modeling,”
    Acta Materialia, vol. 179, pp. 158–171, 2019, doi: 10.1016/j.actamat.2019.08.030.
    [15] S. K. Nayak, S. K. Mishra, C. P. Paul, A. N. Jinoop, and K. S. Bindra, “Effect of energy
    density on laser powder bed fusion built single tracks and thin wall structures with 100 µm
    preplaced powder layer thickness,” Optics and Laser Technology, vol. 125, May 2020, doi:
    10.1016/j.optlastec.2019.106016.
    [16] G. Nordet et al., “Absorptivity measurements during laser powder bed fusion of pure
    copper with a 1 kW cw green laser,” Optics & Laser Technology, vol. 147, no. April 2021,
    p. 107612, 2022, doi: 10.1016/j.optlastec.2021.107612.
    [17] M. Hummel, C. Schöler, A. Häusler, A. Gillner, and R. Poprawe, “New approaches on
    laser micro welding of copper by using a laser beam source with a wavelength of 450 nm,”
    Journal of Advanced Joining Processes, vol. 1, no. February, p. 100012, 2020, doi:
    10.1016/j.jajp.2020.100012.
    [18] M. Hummel, M. Külkens, C. Schöler, W. Schulz, and A. Gillner, “In situ X-ray
    tomography investigations on laser welding of copper with 515 and 1030 nm laser beam
    sources,” Journal of Manufacturing Processes, vol. 67, no. April, pp. 170–176, 2021, doi:
    10.1016/j.jmapro.2021.04.063.
    [19] L. Gargalis et al., “Determining processing behaviour of pure Cu in laser powder bed
    fusion using direct micro-calorimetry,” Journal of Materials Processing Technology, vol.
    294, no. March, p. 117130, 2021, doi: 10.1016/j.jmatprotec.2021.117130.
    [20] A. Mondal, D. Agrawal, and A. Upadhyaya, “Microwave heating of pure copper powder
    with varying particle size and porosity,” Journal of Microwave Power and
    Electromagnetic Energy, vol. 43, no. 1, pp. 4315–43110, 2009, doi:
    10.1080/08327823.2008.11688599.

    Fig. 1. Schematic figure showing the PREP with additional gas flowing on the end face of electrode.

    플라즈마 회전 전극 공정 중 분말 형성에 대한 공정 매개변수 및 냉각 가스의 영향

    Effects of process parameters and cooling gas on powder formation during the plasma rotating electrode process

    Yujie Cuia Yufan Zhaoa1 Haruko Numatab Kenta Yamanakaa Huakang Biana Kenta Aoyagia AkihikoChibaa
    aInstitute for Materials Research, Tohoku University, Sendai 980-8577, JapanbDepartment of Materials Processing, Graduate School of Engineering, Tohoku University, Sendai 980-8577, Japan

    Highlights

    •The limitation of increasing the rotational speed in decreasing powder size was clarified.

    •Cooling and disturbance effects varied with the gas flowing rate.

    •Inclined angle of the residual electrode end face affected powder formation.

    •Additional cooling gas flowing could be applied to control powder size.

    Abstract

    The plasma rotating electrode process (PREP) is rapidly becoming an important powder fabrication method in additive manufacturing. However, the low production rate of fine PREP powder limits the development of PREP. Herein, we investigated different factors affecting powder formation during PREP by combining experimental methods and numerical simulations. The limitation of increasing the rotation electrode speed in decreasing powder size is attributed to the increased probability of adjacent droplets recombining and the decreased tendency of granulation. The effects of additional Ar/He gas flowing on the rotational electrode on powder formation is determined through the cooling effect, the disturbance effect, and the inclined effect of the residual electrode end face simultaneously. A smaller-sized powder was obtained in the He atmosphere owing to the larger inclined angle of the residual electrode end face compared to the Ar atmosphere. Our research highlights the route for the fabrication of smaller-sized powders using PREP.

    플라즈마 회전 전극 공정(PREP)은 적층 제조 에서 중요한 분말 제조 방법으로 빠르게 자리잡고 있습니다. 그러나 미세한 PREP 분말의 낮은 생산율은 PREP의 개발을 제한합니다. 여기에서 우리는 실험 방법과 수치 시뮬레이션을 결합하여 PREP 동안 분말 형성에 영향을 미치는 다양한 요인을 조사했습니다. 분말 크기 감소에서 회전 전극 속도 증가의 한계는 인접한 액적 재결합 확률 증가 및 과립화 경향 감소에 기인합니다.. 회전 전극에 흐르는 추가 Ar/He 가스가 분말 형성에 미치는 영향은 냉각 효과, 외란 효과 및 잔류 전극 단면의 경사 효과를 통해 동시에 결정됩니다. He 분위기에서는 Ar 분위기에 비해 잔류 전극 단면의 경사각이 크기 때문에 더 작은 크기의 분말이 얻어졌다. 우리의 연구는 PREP를 사용하여 더 작은 크기의 분말을 제조하는 경로를 강조합니다.

    Keywords

    Plasma rotating electrode process

    Ti-6Al-4 V alloy, Rotating speed, Numerical simulation, Gas flowing, Powder size

    Introduction

    With the development of additive manufacturing, there has been a significant increase in high-quality powder production demand [1,2]. The initial powder characteristics are closely related to the uniform powder spreading [3,4], packing density [5], and layer thickness observed during additive manufacturing [6], thus determining the mechanical properties of the additive manufactured parts [7,8]. Gas atomization (GA) [9–11], centrifugal atomization (CA) [12–15], and the plasma rotating electrode process (PREP) are three important powder fabrication methods.

    Currently, GA is the dominant powder fabrication method used in additive manufacturing [16] for the fabrication of a wide range of alloys [11]. GA produces powders by impinging a liquid metal stream to droplets through a high-speed gas flow of nitrogen, argon, or helium. With relatively low energy consumption and a high fraction of fine powders, GA has become the most popular powder manufacturing technology for AM.

    The entrapped gas pores are generally formed in the powder after solidification during GA, in which the molten metal is impacted by a high-speed atomization gas jet. In addition, satellites are formed in GA powder when fine particles adhere to partially molten particles.

    The gas pores of GA powder result in porosity generation in the additive manufactured parts, which in turn deteriorates its mechanical properties because pores can become crack initiation sites [17]. In CA, a molten metal stream is poured directly onto an atomizer disc spinning at a high rotational speed. A thin film is formed on the surface of the disc, which breaks into small droplets due to the centrifugal force. Metal powder is obtained when these droplets solidify.

    Compared with GA powder, CA powder exhibits higher sphericity, lower impurity content, fewer satellites, and narrower particle size distribution [12]. However, very high speed is required to obtain fine powder by CA. In PREP, the molten metal, melted using the plasma arc, is ejected from the rotating rod through centrifugal force. Compared with GA powder, PREP-produced powders also have higher sphericity and fewer pores and satellites [18].

    For instance, PREP-fabricated Ti6Al-4 V alloy powder with a powder size below 150 μm exhibits lower porosity than gas-atomized powder [19], which decreases the porosity of additive manufactured parts. Furthermore, the process window during electron beam melting was broadened using PREP powder compared to GA powder in Inconel 718 alloy [20] owing to the higher sphericity of the PREP powder.

    In summary, PREP powder exhibits many advantages and is highly recommended for powder-based additive manufacturing and direct energy deposition-type additive manufacturing. However, the low production rate of fine PREP powder limits the widespread application of PREP powder in additive manufacturing.

    Although increasing the rotating speed is an effective method to decrease the powder size [21,22], the reduction in powder size becomes smaller with the increased rotating speed [23]. The occurrence of limiting effects has not been fully clarified yet.

    Moreover, the powder size can be decreased by increasing the rotating electrode diameter [24]. However, these methods are quite demanding for the PREP equipment. For instance, it is costly to revise the PREP equipment to meet the demand of further increasing the rotating speed or electrode diameter.

    Accordingly, more feasible methods should be developed to further decrease the PREP powder size. Another factor that influences powder formation is the melting rate [25]. It has been reported that increasing the melting rate decreases the powder size of Inconel 718 alloy [26].

    In contrast, the powder size of SUS316 alloy was decreased by decreasing the plasma current within certain ranges. This was ascribed to the formation of larger-sized droplets from fluid strips with increased thickness and spatial density at higher plasma currents [27]. The powder size of NiTi alloy also decreases at lower melting rates [28]. Consequently, altering the melting rate, varied with the plasma current, is expected to regulate the PREP powder size.

    Furthermore, gas flowing has a significant influence on powder formation [27,29–31]. On one hand, the disturbance effect of gas flowing promotes fluid granulation, which in turn contributes to the formation of smaller-sized powder [27]. On the other hand, the cooling effect of gas flowing facilitates the formation of large-sized powder due to increased viscosity and surface tension. However, there is a lack of systematic research on the effect of different gas flowing on powder formation during PREP.

    Herein, the authors systematically studied the effects of rotating speed, electrode diameter, plasma current, and gas flowing on the formation of Ti-6Al-4 V alloy powder during PREP as additive manufactured Ti-6Al-4 V alloy exhibits great application potential [32]. Numerical simulations were conducted to explain why increasing the rotating speed is not effective in decreasing powder size when the rotation speed reaches a certain level. In addition, the different factors incited by the Ar/He gas flowing on powder formation were clarified.

    Fig. 1. Schematic figure showing the PREP with additional gas flowing on the end face of electrode.
    Fig. 1. Schematic figure showing the PREP with additional gas flowing on the end face of electrode.

    References

    [1] W. Ding, G. Chen, M. Qin, Y. He, X. Qu, Low-cost Ti powders for additive manufacturing treated by fluidized bed, Powder Technol. 350 (2019) 117–122, https://doi.org/
    10.1016/j.powtec.2019.03.042.
    [2] W.S.W. Harun, M.S.I.N. Kamariah, N. Muhamad, S.A.C. Ghani, F. Ahmad, Z. Mohamed,
    A review of powder additive manufacturing processes for metallic biomaterials,
    Powder Technol. 327 (2018) 128–151, https://doi.org/10.1016/j.powtec.2017.12.
    058.
    [3] M. Ahmed, M. Pasha, W. Nan, M. Ghadiri, A simple method for assessing powder
    spreadability for additive manufacturing, Powder Technol. 367 (2020) 671–679,
    https://doi.org/10.1016/j.powtec.2020.04.033.
    [4] W. Nan, M. Pasha, M. Ghadiri, Numerical simulation of particle flow and segregation
    during roller spreading process in additive manufacturing, Powder Technol. 364
    (2020) 811–821, https://doi.org/10.1016/j.powtec.2019.12.023.
    [5] A. Averardi, C. Cola, S.E. Zeltmann, N. Gupta, Effect of particle size distribution on the
    packing of powder beds : a critical discussion relevant to additive manufacturing,
    Mater. Today Commun. 24 (2020) 100964, https://doi.org/10.1016/j.mtcomm.
    2020.100964.
    [6] K. Riener, N. Albrecht, S. Ziegelmeier, R. Ramakrishnan, L. Haferkamp, A.B. Spierings,
    G.J. Leichtfried, Influence of particle size distribution and morphology on the properties of the powder feedstock as well as of AlSi10Mg parts produced by laser powder bed fusion (LPBF), Addit. Manuf. 34 (2020) 101286, https://doi.org/10.1016/j.
    addma.2020.101286.
    [7] W.S.W. Harun, N.S. Manam, M.S.I.N. Kamariah, S. Sharif, A.H. Zulkifly, I. Ahmad, H.
    Miura, A review of powdered additive manufacturing techniques for Ti-6Al-4V biomedical applications, Powder Technol. 331 (2018) 74–97, https://doi.org/10.1016/j.
    powtec.2018.03.010.
    [8] A.T. Sutton, C.S. Kriewall, M.C. Leu, J.W. Newkirk, Powder characterisation techniques and effects of powder characteristics on part properties in powder-bed fusion processes, Virtual Phys. Prototyp. 12 (2017) (2017) 3–29, https://doi.org/10.
    1080/17452759.2016.1250605.
    [9] G. Chen, Q. Zhou, S.Y. Zhao, J.O. Yin, P. Tan, Z.F. Li, Y. Ge, J. Wang, H.P. Tang, A pore
    morphological study of gas-atomized Ti-6Al-4V powders by scanning electron microscopy and synchrotron X-ray computed tomography, Powder Technol. 330
    (2018) 425–430, https://doi.org/10.1016/j.powtec.2018.02.053.
    [10] Y. Feng, T. Qiu, Preparation, characterization and microwave absorbing properties of
    FeNi alloy prepared by gas atomization method, J. Alloys Compd. 513 (2012)
    455–459, https://doi.org/10.1016/j.jallcom.2011.10.079.

    [11] I.E. Anderson, R.L. Terpstra, Progress toward gas atomization processing with increased uniformity and control, Mater. Sci. Eng. A 326 (2002) 101–109, https://
    doi.org/10.1016/S0921-5093(01)01427-7.
    [12] P. Phairote, T. Plookphol, S. Wisutmethangoon, Design and development of a centrifugal atomizer for producing zinc metal powder, Int. J. Appl. Phys. Math. 2 (2012)
    77–82, https://doi.org/10.7763/IJAPM.2012.V2.58.
    [13] L. Tian, I. Anderson, T. Riedemann, A. Russell, Production of fine calcium powders by
    centrifugal atomization with rotating quench bath, Powder Technol. 308 (2017)
    84–93, https://doi.org/10.1016/j.powtec.2016.12.011.
    [14] M. Eslamian, J. Rak, N. Ashgriz, Preparation of aluminum/silicon carbide metal matrix composites using centrifugal atomization, Powder Technol. 184 (2008) 11–20,
    https://doi.org/10.1016/j.powtec.2007.07.045.
    [15] T. Plookphol, S. Wisutmethangoon, S. Gonsrang, Influence of process parameters on
    SAC305 lead-free solder powder produced by centrifugal atomization, Powder
    Technol. 214 (2011) 506–512, https://doi.org/10.1016/j.powtec.2011.09.015.
    [16] M.Z. Gao, B. Ludwig, T.A. Palmer, Impact of atomization gas on characteristics of austenitic stainless steel powder feedstocks for additive manufacturing, Powder
    Technol. 383 (2021) 30–42, https://doi.org/10.1016/j.powtec.2020.12.005.
    [17] X. Shui, K. Yamanaka, M. Mori, Y. Nagata, A. Chiba, Effects of post-processing on cyclic fatigue response of a titanium alloy additively manufactured by electron beam
    melting, Mater. Sci. Eng. A 680 (2017) 239–248, https://doi.org/10.1016/j.msea.
    2016.10.059.
    [18] C. Wang, X.H. Zhao, Y.C. Ma, Q.X. Wang, Y.J. Lai, S.J. Liang, Study of the spherical
    HoCu powders prepared by supreme-speed plasma rotating electrode process,
    Powder Metallurgy Technology 38 (3) (2020) 227–233, https://doi.org/10.19591/
    j.cnki.cn11-1974/tf.2020.03.011 (in Chinese).
    [19] G. Chen, S.Y. Zhao, P. Tan, J. Wang, C.S. Xiang, H.P. Tang, A comparative study of Ti6Al-4V powders for additive manufacturing by gas atomization, plasma rotating
    electrode process and plasma atomization, Powder Technol. 333 (2018) 38–46,
    https://doi.org/10.1016/j.powtec.2018.04.013.
    [20] Y. Zhao, K. Aoyagi, Y. Daino, K. Yamanaka, A. Chiba, Significance of powder feedstock
    characteristics in defect suppression of additively manufactured Inconel 718, Addit.
    Manuf. 34 (2020) 101277, https://doi.org/10.1016/j.addma.2020.101277.
    [21] Y. Nie, J. Tang, B. Yang, Q. Lei, S. Yu, Y. Li, Comparison in characteristic and atomization behavior of metallic powders produced by plasma rotating electrode process,
    Adv. Powder Technol. 31 (2020) 2152–2160, https://doi.org/10.1016/j.apt.2020.03.
    006.
    [22] Y. Cui, Y. Zhao, H. Numata, H. Bian, K. Wako, K. Yamanaka, K. Aoyagi, C. Zhang, A.
    Chiba, Effects of plasma rotating electrode process parameters on the particle size
    distribution and microstructure of Ti-6Al-4 V alloy powder, Powder Technol 376
    (2020) 363–372, https://doi.org/10.1016/j.powtec.2020.08.027.
    [23] J. Tang, Y. Nie, Q. Lei, Y. Li, Characteristics and atomization behavior of Ti-6Al-4V
    powder produced by plasma rotating electrode process Adv, Powder Technol. 10
    (2019) 2330–2337, https://doi.org/10.1016/j.apt.2019.07.015.
    [24] M. Zdujić, D. Uskoković, Production of atomized metal and alloy powders by the rotating electrode process, Sov. Powder Metall. Met. Ceram. 29 (1990) 673–683,
    https://doi.org/10.1007/BF00795571.
    [25] L. Zhang, Y. Zhao, Particle size distribution of tin powder produced by centrifugal
    atomisation using rotating cups, Powder Technol. 318 (2017) 62–67, https://doi.
    org/10.1016/j.powtec.2017.05.038.
    [26] Y. Liu, S. Liang, Z. Han, J. Song, Q. Wang, A novel model of calculating particle sizes in
    plasma rotating electrode process for superalloys, Powder Technol. 336 (2018)
    406–414, https://doi.org/10.1016/j.powtec.2018.06.002.
    [27] Y. Zhao, Y. Cui, H. Numata, H. Bian, K. Wako, K. Yamanaka, Centrifugal granulation
    behavior in metallic powder fabrication by plasma rotating electrode process, Sci.
    Rep. (2020) 1–15, https://doi.org/10.1038/s41598-020-75503-w.
    [28] T. Hsu, C. Wei, L. Wu, Y. Li, A. Chiba, M. Tsai, Nitinol powders generate from plasma
    rotation electrode process provide clean powder for biomedical devices used with
    suitable size, spheroid surface and pure composition, Sci. Rep. 8 (2018) 1–8,
    https://doi.org/10.1038/s41598-018-32101-1.
    [29] M. Wei, S. Chen, M. Sun, J. Liang, C. Liu, M. Wang, Atomization simulation and preparation of 24CrNiMoY alloy steel powder using VIGA technology at high gas pressure, Powder Technol. 367 (2020) 724–739, https://doi.org/10.1016/j.powtec.
    2020.04.030.
    [30] Y. Tan, X. Zhu, X.Y. He, B. Ding, H. Wang, Q. Liao, H. Li, Granulation characteristics of
    molten blast furnace slag by hybrid centrifugal-air blast technique, Powder Technol.
    323 (2018) 176–185, https://doi.org/10.1016/j.powtec.2017.09.040.
    [31] P. Xu, D.H. Liu, J. Hu, G.Y. Lin, Synthesis of Ni-Ti composite powder by radio frequency plasma spheroidization process, Nonferrous Metals Science and Engineering
    39 (1) (2020) 67–71 , (in Chinese) 10.13264/j.cnki.ysjskx.2020.01.011.
    [32] H. Mehboob, F. Tarlochan, A. Mehboob, S.H. Chang, S. Ramesh, W.S.W. Harun, K.
    Kadirgama, A novel design, analysis and 3D printing of Ti-6Al-4V alloy bioinspired porous femoral stem, J. Mater. Sci. Mater. Med. 31 (2020) 78, https://doi.
    org/10.1007/s10856-020-06420-7.
    [33] FLOW-3D® Version 11.2 [Computer software]. , Flow Science, Inc., Santa Fe, NM,
    2017https://www.flow3d.com.
    [34] M. Boivineau, C. Cagran, D. Doytier, V. Eyraud, M.H. Nadal, B. Wilthan, G. Pottlacher,
    Thermophysical properties of solid and liquid Ti-6Al-4V (TA6V) alloy, Int. J.
    Thermophys. 27 (2006) 507–529, https://doi.org/10.1007/PL00021868.
    [35] J. Liu, Q. Qin, Q. Yu, The effect of size distribution of slag particles obtained in dry
    granulation on blast furnace slag cement strength, Powder Technol. 362 (2020)
    32–36, https://doi.org/10.1016/j.powtec.2019.11.115.
    [36] M. Tanaka, S. Tashiro, A study of thermal pinch effect of welding arcs, J. Japan Weld.
    Soc. 25 (2007) 336–342, https://doi.org/10.2207/qjjws.25.336 (in Japanese).
    [37] T. Kamiya, A. Kayano, Disintegration of viscous fluid in the ligament state purged
    from a rotating disk, J. Chem. Eng. JAPAN. 4 (1971) 364–369, https://doi.org/10.
    1252/jcej.4.364.
    [38] T. Kamiya, An analysis of the ligament-type disintegration of thin liquid film at the
    edge of a rotating disk, J. Chem. Eng. Japan. 5 (1972) 391–396, https://doi.org/10.
    1252/jcej.5.391.
    [39] J. Burns, C. Ramshaw, R. Jachuck, Measurement of liquid film thickness and the determination of spin-up radius on a rotating disc using an electrical resistance technique, Chem. Eng. Sci. 58 (2003) 2245–2253, https://doi.org/10.1016/S0009-2509
    (03)00091-5.
    [40] J. Rauscher, R. Kelly, J. Cole, An asymptotic solution for the laminar flow of a thin film
    on a rotating disk, J. Appl. Mech. Trans. ASME 40 (1973) 43–47, https://doi.org/10.
    1115/1.3422970

    Gating System Design Based on Numerical Simulation and Production Experiment Verification of Aluminum Alloy Bracket Fabricated by Semi-solid Rheo-Die Casting Process

    Gating System Design Based on Numerical Simulation and Production Experiment Verification of Aluminum Alloy Bracket Fabricated by Semi-solid Rheo-Die Casting Process

    반고체 레오 다이 캐스팅 공정으로 제작된 알루미늄 합금 브래킷의 수치 시뮬레이션 및 생산 실험 검증을 기반으로 한 게이팅 시스템 설계

    International Journal of Metalcasting volume 16, pages878–893 (2022)Cite this article

    Abstract

    In this study a gating system including sprue, runner and overflows for semi-solid rheocasting of aluminum alloy was designed by means of numerical simulations with a commercial software. The effects of pouring temperature, mold temperature and injection speed on the filling process performance of semi-solid die casting were studied. Based on orthogonal test analysis, the optimal die casting process parameters were selected, which were metal pouring temperature 590 °C, mold temperature 260 °C and injection velocity 0.5 m/s. Semi-solid slurry preparation process of Swirled Enthalpy Equilibration Device (SEED) was used for die casting production experiment. Aluminum alloy semi-solid bracket components were successfully produced with the key die casting process parameters selected, which was consistent with the simulation result. The design of semi-solid gating system was further verified by observing and analyzing the microstructure of different zones of the casting. The characteristic parameters, particle size and shape factor of microstructure of the produced semi-solid casting showed that the semi-solid aluminum alloy components are of good quality.

    이 연구에서 알루미늄 합금의 반고체 레오캐스팅을 위한 스프루, 러너 및 오버플로를 포함하는 게이팅 시스템은 상용 소프트웨어를 사용한 수치 시뮬레이션을 통해 설계되었습니다. 주입 온도, 금형 온도 및 사출 속도가 반고체 다이캐스팅의 충전 공정 성능에 미치는 영향을 연구했습니다. 직교 테스트 분석을 기반으로 금속 주입 온도 590°C, 금형 온도 260°C 및 사출 속도 0.5m/s인 최적의 다이 캐스팅 공정 매개변수가 선택되었습니다. Swirled Enthalpy Equilibration Device(SEED)의 반고체 슬러리 제조 공정을 다이캐스팅 생산 실험에 사용하였다. 알루미늄 합금 반고체 브래킷 구성 요소는 시뮬레이션 결과와 일치하는 주요 다이 캐스팅 공정 매개변수를 선택하여 성공적으로 생산되었습니다. 반고체 게이팅 시스템의 설계는 주조의 다른 영역의 미세 구조를 관찰하고 분석하여 추가로 검증되었습니다. 생산된 반고체 주조물의 특성 매개변수, 입자 크기 및 미세 구조의 형상 계수는 반고체 알루미늄 합금 부품의 품질이 양호함을 보여주었습니다.

    Gating System Design Based on Numerical Simulation and Production Experiment Verification of Aluminum Alloy Bracket Fabricated by Semi-solid Rheo-Die Casting Process
    Gating System Design Based on Numerical Simulation and Production Experiment Verification of Aluminum Alloy Bracket Fabricated by Semi-solid Rheo-Die Casting Process

    References

    1. G. Li, H. Lu, X. Hu et al., Current progress in rheoforming of wrought aluminum alloys: a review. Met. Open Access Metall. J. 10(2), 238 (2020)CAS Google Scholar 
    2. G. Eisaabadi, A. Nouri, Effect of Sr on the microstructure of electromagnetically stirred semi-solid hypoeutectic Al–Si alloys. Int. J. Metalcast. 12, 292–297 (2018). https://doi.org/10.1007/s40962-017-0161-8CAS Article Google Scholar 
    3. C. Xghab, D. Qza, E. Spma et al., Blistering in semi-solid die casting of aluminium alloys and its avoidance. Acta Mater. 124, 446–455 (2017)Article Google Scholar 
    4. M. Modigell, J. Koke, Rheological modelling on semi-solid metal alloys and simulation of thixocasting processes. J. Mater. Process. Technol. 111(1–3), 53–58 (2001)CAS Article Google Scholar 
    5. A. Pola, M. Tocci, P. Kapranos, Microstructure and properties of semi-solid aluminum alloys: a literature review. Met. Open Access Metall. J. 8(3), 181 (2018)Google Scholar 
    6. M.C. Flemings, Behavior of metal alloys in the semisolid state. Metall. Trans. B 22, 269–293 (1991). https://doi.org/10.1007/BF02651227Article Google Scholar 
    7. Q. Zhu, Semi-solid moulding: competition to cast and machine from forging in making automotive complex components. Trans. Nonferrous Met. Soc. China 20, 1042–1047 (2010)Article Google Scholar 
    8. K. Prapasajchavet, Y. Harada, S. Kumai, Microstructure analysis of Al–5.5 at.%Mg alloy semi-solid slurry by Weck’s reagent. Int. J. Metalcast. 11(1), 123 (2017). https://doi.org/10.1007/s40962-016-0084-9Article Google Scholar 
    9. P. Das, S.K. Samanta, S. Tiwari, P. Dutta, Die filling behaviour of semi solid A356 Al alloy slurry during rheo pressure die casting. Trans. Indian Inst. Met. 68(6), 1215–1220 (2015). https://doi.org/10.1007/s12666-015-0706-6CAS Article Google Scholar 
    10. B. Zhou, S. Lu, K. Xu et al., Microstructure and simulation of semisolid aluminum alloy castings in the process of stirring integrated transfer-heat (SIT) with water cooling. Int. J. Metalcast. 14(2), 396–408 (2019). https://doi.org/10.1007/s40962-019-00357-6CAS Article Google Scholar 
    11. S. Ji, Z. Fan, Solidification behavior of Sn–15 wt Pct Pb alloy under a high shear rate and high intensity of turbulence during semisolid processing. Metall. Mater. Trans. A. 33(11), 3511–3520 (2002). https://doi.org/10.1007/s11661-002-0338-4Article Google Scholar 
    12. P. Kapranos, P.J. Ward, H.V. Atkinson, D.H. Kirkwood, Near net shaping by semi-solid metal processing. Mater. Des. 21, 387–394 (2000). https://doi.org/10.1016/S0261-3069(99)00077-1Article Google Scholar 
    13. H.V. Atkinson, Alloys for semi-solid processing. Solid State Phenom. 192–193, 16–27 (2013)Google Scholar 
    14. L. Rogal, Critical assessment: opportunities in developing semi-solid processing: aluminium, magnesium, and high-temperature alloys. Mater. Sci. Technol. Mst A Publ. Inst. Met. 33, 759–764 (2017)CAS Article Google Scholar 
    15. H. Guo, Rheo-diecasting process for semi-solid aluminum alloys. J. Wuhan Univ. Technol. Mater. Sci. Ed. 22(004), 590–595 (2007)CAS Article Google Scholar 
    16. T. Chucheep, J. Wannasin, R. Canyook, T. Rattanochaikul, S. Janudom, S. Wisutmethangoon, M.C. Flemings, Characterization of flow behavior of semi-solid slurries with low solid fractions. Metall. Mater. Trans. A 44(10), 4754–4763 (2013)CAS Article Google Scholar 
    17. M. Li, Y.D. Li, W.L. Yang et al., Effects of forming processes on microstructures and mechanical properties of A356 aluminum alloy prepared by self-inoculation method. Mater. Res. 22(3) (2019)
    18. P. Côté, M.E. Larouche, X.G. Chen et al., New developments with the SEED technology. Solid State Phenom. 192(3), 373–378 (2012)Article Google Scholar 
    19. I. Dumanić, S. Jozić, D. Bajić et al., Optimization of semi-solid high-pressure die casting process by computer simulation, Taguchi method and grey relational analysis. Inter Metalcast. 15, 108–118 (2021). https://doi.org/10.1007/s40962-020-00422-5Article Google Scholar 
    20. Y. Bai et al., Numerical simulation on the rheo-diecasting of the semi-solid A356 aluminum alloy. Int. J. Miner. Metall. Mater. 16, 422 (2009). https://doi.org/10.1016/S1674-4799(09)60074-1CAS Article Google Scholar 
    21. B.C. Bhunia, Studies on die filling of A356 Al alloy and development of a steering knuckle component using rheo pressure die casting system. J. Mater. Process. Technol. 271, 293–311 (2019). https://doi.org/10.1016/j.jmatprotec.2019.04.014CAS Article Google Scholar 
    22. A. Guo, J. Zhao, C. Xu et al., Effects of pouring temperature and electromagnetic stirring on porosity and mechanical properties of A357 aluminum alloy rheo-diecasting. J. Mater. Eng. Perform. (2018). https://doi.org/10.1007/s11665-018-3310-1Article Google Scholar 
    23. C.G. Kang, S.M. Lee, B.M. Kim, A study of die design of semi-solid die casting according to gate shape and solid fraction. J. Mater. Process. Technol. 204(1–3), 8–21 (2008)CAS Article Google Scholar 
    24. Z. Liu, W. Mao, T. Wan et al., Study on semi-solid A380 aluminum alloy slurry prepared by water-cooling serpentine channel and its rheo-diecasting. Met. Mater. Int. (2020). https://doi.org/10.1007/s12540-020-00672-2Article Google Scholar 
    25. Z.Y. Liu, W.M. Mao, W.P. Wang et al., Investigation of rheo-diecasting mold filling of semi-solid A380 aluminum alloy slurry. Int. J. Miner. Metall. Mater. 24(006), 691–700 (2017)CAS Article Google Scholar 
    26. M. Arif, M.Z. Omar, N. Muhamad et al., Microstructural evolution of solid-solution-treated Zn–22Al in the semisolid state. J. Mater. Sci. Technol. 29(008), 765–774 (2013)CAS Article Google Scholar 

    Keywords

    • semi-solid rheo-die casting
    • gating system
    • process parameters
    • numerical simulation
    • microstructure
    Fig. 2: Scheme of the LED photo-crosslinking and 3D-printing section of the microfluidic/3D-printing device. The droplet train is transferred from the chip microchannel into a microtubing in a straight section with nearly identical inner channel and inner microtubing diameter. Further downstream, the microtubing passes an LED-section for fast photo cross-linking to generate the microgels. This section is contained in an aluminum encasing to avoid premature crosslinking of polymer precursor in upstream channel sections by stray light. Subsequently, the microtubing is integrated into a 3D-printhead, where the microgels are jammed into a filament that is directly 3D-printed into the scaffold.

    On-Chip Fabrication and In-Flow 3D-Printing of Cell-Laden Microgel Constructs: From Chip to Scaffold Materials in One Integral Process

    세포가 함유된 마이크로겔의 온칩 제작 및 인-플로우 3D 프린팅
    구성:하나의 통합 프로세스에서 칩에서 스캐폴드 재료까지

    Vollmer, Gültekin Tamgüney, Aldo Boccacini
    Submitted date: 10/05/2021 • Posted date: 11/05/2021
    Licence: CC BY-NC-ND 4.0

    바이오프린팅은 세포가 실린 스캐폴드의 제조를 위한 유력한 기술로 발전했습니다. 바이오잉크는 바이오프린팅의 가장 중요한 구성요소입니다. 최근 마이크로겔은 세포 보호 및 세포 미세 환경 제어를 가능하게 하는 매우 유망한 바이오 잉크로 도입되었습니다. 그러나 이들의 미세유체 제작은 본질적으로 한계가 있는 것으로 보입니다.

    여기에서 우리는 안정적인 스캐폴드에 직접 유입되는 바이오프린팅과 함께 세포가 실린 마이크로겔의 미세유체 생산을 위한 미세유체 및 3D 인쇄의 직접 결합을 소개합니다. 방법론은 세포를 단분산 미세 방울로 연속 온칩 캡슐화하여 후속 유입 교차 연결을 통해 세포가 함유된 마이크로겔을 생성할 수 있으며, 이는 미세관을 종료한 후 자동으로 얇은 연속 마이크로겔 필라멘트로 끼이게 됩니다.

    3D 프린트 헤드로의 통합으로 독립형 3차원 스캐폴드에 필라멘트를 직접 유입 인쇄할 수 있습니다. 이 방법은 다양한 교차 연결 방법 및 세포주에 대해 설명됩니다. 이러한 발전으로 미세유체학은 더 이상 바이오 제조의 병목을 초래하는 현상이 아닙니다.

    Bioprinting has evolved into a thriving technology for the fabrication of cell-laden scaffolds. Bioinks are the most critical component for bioprinting. Recently, microgels have been introduced as a very promising bioink enabling cell protection and the control of the cellular microenvironment. However, their microfluidic fabrication inherently seemed to be a limitation. Here we introduce a direct coupling of microfluidics and 3D-printing for the microfluidic production of cell-laden microgels with direct in-flow bioprinting into stable scaffolds. The methodology enables the continuous on-chip encapsulation of cells into monodisperse microdroplets with subsequent in-flow cross-linking to produce cell-laden microgels, which after exiting a microtubing are automatically jammed into thin continuous microgel filaments. The integration into a 3D printhead allows direct in-flow printing of the filaments into free-standing three-dimensional scaffolds. The method is demonstrated for different cross-linking methods and cell lines. With this advancement, microfluidics is no longer a bottleneck for biofabrication.

    Fig. 1: Three-dimensional schematic view of the multilayer double 3D-focusing microfluidic channel system, (b) control of droplet diameter via the Capiilary number Ca, and accessible hydrodynamic regimes for droplet production: squeezing (c), dripping (d) and jetting (e). The scale bars are 200 µm.
    Fig. 1: Three-dimensional schematic view of the multilayer double 3D-focusing microfluidic channel system, (b) control of droplet diameter via the Capiilary number Ca, and accessible hydrodynamic regimes for droplet production: squeezing (c), dripping (d) and jetting (e). The scale bars are 200 µm.
    Fig. 2: Scheme of the LED photo-crosslinking and 3D-printing section of the microfluidic/3D-printing device. The droplet train is transferred from the chip microchannel into a microtubing in a straight section with nearly identical inner channel and inner microtubing diameter. Further downstream, the microtubing passes an LED-section for fast photo cross-linking to generate the microgels. This section is contained in an aluminum encasing to avoid premature crosslinking of polymer precursor in upstream channel sections by stray light. Subsequently, the microtubing is integrated into a 3D-printhead, where the microgels are jammed into a filament that is directly 3D-printed into the scaffold.
    Fig. 2: Scheme of the LED photo-crosslinking and 3D-printing section of the microfluidic/3D-printing device. The droplet train is transferred from the chip microchannel into a microtubing in a straight section with nearly identical inner channel and inner microtubing diameter. Further downstream, the microtubing passes an LED-section for fast photo cross-linking to generate the microgels. This section is contained in an aluminum encasing to avoid premature crosslinking of polymer precursor in upstream channel sections by stray light. Subsequently, the microtubing is integrated into a 3D-printhead, where the microgels are jammed into a filament that is directly 3D-printed into the scaffold.
    Fig. 3: a) Photograph of a standard meander-shaped layer fabricated by microgel filament deposition printing. The lines have a thickness of 300 µm. b) photograph of a cross-bar pattern obtained by on-top deposition of several microgel filaments. The average linewidth is 1 mm. c) photograph of a donut-shaped microgel construct. The microgels have been fluorescently labelled by FITC-dextran to demonstrate the intrinsic microporosity corresponding to the black non-fluorescent regions, d) light microscopy image of a construct edge showing that fused adhesive microgels form a continuous, three-dimensional selfsupporting scaffold with intrinsic micropores.
    Fig. 3: a) Photograph of a standard meander-shaped layer fabricated by microgel filament deposition printing. The lines have a thickness of 300 µm. b) photograph of a cross-bar pattern obtained by on-top deposition of several microgel filaments. The average linewidth is 1 mm. c) photograph of a donut-shaped microgel construct. The microgels have been fluorescently labelled by FITC-dextran to demonstrate the intrinsic microporosity corresponding to the black non-fluorescent regions, d) light microscopy image of a construct edge showing that fused adhesive microgels form a continuous, three-dimensional selfsupporting scaffold with intrinsic micropores.
    Fig. 4: a) Scheme of the perfusion chamber consisting of an upstream and downstream chamber, perfusion ports, and removable scaffolds to stabilize the microgel construct during 3D-printing, b) photograph of a microgel construct in the perfusion chamber directly after printing and removal of the scaffolds, c) confocal microscopy image of the permeation front of a fluorescent dye, where the high dye concentration in the micropores can be clearly seen, d) confocal microscopy image of YFP-labelled HEK-cells within a microgel construct.
    Fig. 4: a) Scheme of the perfusion chamber consisting of an upstream and downstream chamber, perfusion ports, and removable scaffolds to stabilize the microgel construct during 3D-printing, b) photograph of a microgel construct in the perfusion chamber directly after printing and removal of the scaffolds, c) confocal microscopy image of the permeation front of a fluorescent dye, where the high dye concentration in the micropores can be clearly seen, d) confocal microscopy image of YFP-labelled HEK-cells within a microgel construct.
    Fig. 5: a) Layer-by-layer printing of microgel construct with integrated perfusion channel. After printing of the first layer, a hollow perfusion channel is inserted. Subsequently, the second and third layers are printed. b) The construct is directly printed into a perfusion chamber. The perfusion chamber provides whole construct permeation via flows cin and cout, as well as independent flow through the perfusion channel via flows vin and vout. c) Photograph of a perfusion chamber containing the construct directly after printing. The flow of the fluorescein solution through the integrated PVA hollow channel is clearly visible.
    Fig. 5: a) Layer-by-layer printing of microgel construct with integrated perfusion channel. After printing of the first layer, a hollow perfusion channel is inserted. Subsequently, the second and third layers are printed. b) The construct is directly printed into a perfusion chamber. The perfusion chamber provides whole construct permeation via flows cin and cout, as well as independent flow through the perfusion channel via flows vin and vout. c) Photograph of a perfusion chamber containing the construct directly after printing. The flow of the fluorescein solution through the integrated PVA hollow channel is clearly visible.
    Fig. 6: a) Photograph of an alginate capsule fiber formed after exiting the microtube. b) Confocal fluorescence microscopy image of part of a 3D-printed alginate capsule construct. The fluorescence arises from encapsulated fluorescently labelled polystyrene microbeads to demonstrate the integrity and stability of the alginate capsules.
    Fig. 6: a) Photograph of an alginate capsule fiber formed after exiting the microtube. b) Confocal fluorescence microscopy image of part of a 3D-printed alginate capsule construct. The fluorescence arises from encapsulated fluorescently labelled polystyrene microbeads to demonstrate the integrity and stability of the alginate capsules.

    Keywords

    biomaterials, microgels, microfluidics, 3D printing, bioprinting

    References

    1. A. Atala, Chem. Rev. 2020, 120, 10545-10546.
    2. J. Groll, J. A. Burdick, D. W. Cho, B. Derby, M. Gelinsky, S. C. Heilshorn, T. Jüngst, J. Malda, V. A
      Mironov, K. Nakayama, A. Ovisanikov, W. Sun, S. Takeuchi, J. J. Yoo, T. B. F. Woodfield,
      Biofabrication 2019, 11, 013001.
    3. W. Sun, B. Starly, A. C. Daly, J. A. Burdick, J. Groll, G. Skeldon, W. Shu, Y. Sakai, M. Shinohara,
      M. Nishikawa, J. Jang, D.-W. Cho, M. Nie, S. Takeuchi, S. Ostrovidov, A. Khademhosseini, R. D. Kamm,
      V. Mironov, L. Moroni, I. T. Ozbolat, Biofabrication 2020, 12, 022002.
    4. R. Levato, T. Juengst, R. G. Scheuring, T. Blunk, J. Groll, J. Malda, Adv. Mater. 2020, 32, 1906423.
    5. C. B. Highley, K. H. Song, A. C. Daly, J. A. Burdick, Adv. Sci. 2019, 6, 1801076.
    6. D. Velasco, E. Tumarkin, E. Kumacheva, Small 2012, 8, 1633-1642.
    7. W. Jiang, M. Li, Z. Chen, K. W. Leong, Lab Chip 2016, 16, 4482-4506.
    8. A. C. Daly, L. Riley, T. Segura, J. A. Burdick, Nat. Rev. 2020, 5, 20-43.
    9. A. S. Mao, B. Özkale, N. J. Shah, K. H. Vining, T. Descombes, L. Zhang, C. M. Tringides, S.-W.
      Wong, J.-W. Shin, D. T. Scadden, D. A. Weitz, D. J. Mooney, Proc. Natl. Acad. Sci. 2019, 116, 15392-15397.
    10. S. R. Pajoumshariati, M. Azizi, D. Wesner, P. G. Miller, M. L. Shuler, A. Abbaspourrad, ACS Appl.
      Mater. Interfaces 2018, 10, 9235-9246.
    11. A. S. Mao, J.-W. Shin, S. Utech, H. Wang, O. Uzun, W. Li, M. Cooper, Y. Hu, L. Zhang, D. A.
      Weitz, D. J. Mooney, Nat. Mater. 2017, 16, 236-243.
    12. P. S. Lienemann, T. Rossow, A. S. Mao, Q. Vallmajo-Martin, M. Ehrbar, D. J. Mooney, Lab Chip, 2017, 17, 727.
    13. F. Chen, J. Xue, J. Zhang, M. Bai, X. Yu, X.; C. Fan, Y. Zhao, J. Am. Chem. Soc. 2020, 142, 2889-2896.
    14. Q. Feng, Q. Li, H. Wen, J. Chen, M. Liang, H. Huang, D. Lan, H. Dong, X. Cao, Adv. Funct. Mater.,
      2019, 29, 1096690.
    15. L. P. B. Guerzoni, T. Yoshinari, D. B. Gehlen, D. Rommel, T. Haraszti, M. Akashi, L. De Laporte,
      Biomacromolecules 2019, 20, 3746-3754
    16. T. Rossow, J. A. Heyman, A. J. Ehrlicher, A. Langhoff, D. A. Weitz, R. Haag, S. Seiffert, J. Am.
      Chem. Soc. 2012, 134, 4983-4989.
    17. E. Kapourani, F. Neumann, K. Achazi, J. Dernedde, R. Haag, Macromol. Bioscience 2018, 18,1800116
    18. H. Wang, H. Liu, H. Liu, W. Su, W. Chen, J. Qin, Adv. Mater. Technol. 2019, 4, 1800632.
    19. C. Fan, S.-H. Zhan, Z.-X. Dong, W. Yang, W.-S. Deng, X. Liu, P. Suna, D.-A. Wang, Mater. Sci. Eng. C 2019, 108, 110399.
    20. A. M. Compaan, K. Song, W. Chai, Y. Huang, ACS Appl. Mater. Interfaces 2020, 12, 7855-7868.
    21. S. L. Anna, H. C. Mayer, Phys. Fluids 2006, 18, 121512.
    22. T. Ward, M. Faivre, M. Abkarian, H. A. Stone, Electrophoresis 2005, 26, 3716-3724.
    23. F. Lapierre, N. Wu, Y. Zhu, Proc. SPIE 2011, 8204, 82040H-1.
    24. C. A. Stan, S. K. Y. Tang, G. M. Whitesides, Anal. Chem. 2009, 81, 2399-2402.
    25. J. Tan, J. H. Xu, S. W. Li, G. S. Luo, Chem. Eng. J. 2008, 136, 306-311.
    26. R.-C. Luo, C.-H. Chen, Soft 2012, 1, 1-23.
    27. C. H. Choi, J. H. Jung, T. S. Hwang, C. S. Lee, Macromol. Res. 2009, 17, 163-167.
    28. A. J. D. Krüger, O. Bakirman, P. B. Guerzoni, A. Jans, D. B. Gehlen, D. Rommel, T. Haraszti, A. J.
      C. Kuehne, L. De Laporte, Adv. Mater. 2019, 31, 1903668.
    29. D. B. Kolesky, K. A. Homan, M. A. Skylar-Scott, J. A. Lewis, Proc. Natl. Acad. Sci. 2016, 113, 3179-3184
    30. A. K. Miri, I. Mirzaee, S. Hassan, S. M. Oskui, D. Nieto, A. Khademhosseini, Y. S. Zhang, Lab Chip 2019, 19, 2019.
    31. F. A. Plamper, W. Richtering Acc. Chem. Res. 2017, 50, 131-140.
    32. S. Sun, M. Li, A. Liu, Int. J. Adhesion Adhesives 2013, 41, 98-106.
    Figure 3.10: Snapshots of Temperature Profile for Single Track in Keyhole Regime (P = 250W and V = 0.5m/s) at the Preheating Temperature of 100 °C

    Multiscale Process Modeling of Residual Deformation and Defect Formation for Laser Powder Bed Fusion Additive Manufacturing

    Qian Chen, PhD
    University of Pittsburgh, 2021

    레이저 분말 베드 퓨전(L-PBF) 적층 제조(AM)는 우수한 기계적 특성으로 그물 모양에 가까운 복잡한 부품을 생산할 수 있습니다. 그러나 빌드 실패 및 다공성과 같은 결함으로 이어지는 원치 않는 잔류 응력 및 왜곡이 L-PBF의 광범위한 적용을 방해하고 있습니다.

    L-PBF의 잠재력을 최대한 실현하기 위해 잔류 변형, 용융 풀 및 다공성 형성을 예측하는 다중 규모 모델링 방법론이 개발되었습니다. L-PBF의 잔류 변형 및 응력을 부품 규모에서 예측하기 위해 고유 변형 ​​방법을 기반으로 하는 다중 규모 프로세스 모델링 프레임워크가 제안됩니다.

    고유한 변형 벡터는 마이크로 스케일에서 충실도가 높은 상세한 다층 프로세스 시뮬레이션에서 추출됩니다. 균일하지만 이방성인 변형은 잔류 왜곡 및 응력을 예측하기 위해 준 정적 평형 유한 요소 분석(FEA)에서 레이어별로 L-PBF 부품에 적용됩니다.

    부품 규모에서의 잔류 변형 및 응력 예측 외에도 분말 규모의 다중물리 모델링을 수행하여 공정 매개변수, 예열 온도 및 스패터링 입자에 의해 유도된 용융 풀 변동 및 결함 형성을 연구합니다. 이러한 요인과 관련된 용융 풀 역학 및 다공성 형성 메커니즘은 시뮬레이션 및 실험을 통해 밝혀졌습니다.

    제안된 부품 규모 잔류 응력 및 왜곡 모델을 기반으로 경로 계획 방법은 큰 잔류 변형 및 건물 파손을 방지하기 위해 주어진 형상에 대한 레이저 스캐닝 경로를 조정하기 위해 개발되었습니다.

    연속 및 아일랜드 스캐닝 전략을 위한 기울기 기반 경로 계획이 공식화되고 공식화된 컴플라이언스 및 스트레스 최소화 문제에 대한 전체 감도 분석이 수행됩니다. 이 제안된 경로 계획 방법의 타당성과 효율성은 AconityONE L-PBF 시스템을 사용하여 실험적으로 입증되었습니다.

    또한 기계 학습을 활용한 데이터 기반 프레임워크를 개발하여 L-PBF에 대한 부품 규모의 열 이력을 예측합니다. 본 연구에서는 실시간 열 이력 예측을 위해 CNN(Convolutional Neural Network)과 RNN(Recurrent Neural Network)을 포함하는 순차적 기계 학습 모델을 제안합니다.

    유한 요소 해석과 비교하여 100배의 예측 속도 향상이 달성되어 실제 제작 프로세스보다 빠른 예측이 가능하고 실시간 온도 프로파일을 사용할 수 있습니다.

    Laser powder bed fusion (L-PBF) additive manufacturing (AM) is capable of producing complex parts near net shape with good mechanical properties. However, undesired residual stress and distortion that lead to build failure and defects such as porosity are preventing broader applications of L-PBF. To realize the full potential of L-PBF, a multiscale modeling methodology is developed to predict residual deformation, melt pool, and porosity formation. To predict the residual deformation and stress in L-PBF at part-scale, a multiscale process modeling framework based on inherent strain method is proposed.

    Inherent strain vectors are extracted from detailed multi-layer process simulation with high fidelity at micro-scale. Uniform but anisotropic strains are then applied to L-PBF part in a layer-by-layer fashion in a quasi-static equilibrium finite element analysis (FEA) to predict residual distortion and stress. Besides residual distortion and stress prediction at part scale, multiphysics modeling at powder scale is performed to study the melt pool variation and defect formation induced by process parameters, preheating temperature and spattering particles. Melt pool dynamics and porosity formation mechanisms associated with these factors are revealed through simulation and experiments.

    Based on the proposed part-scale residual stress and distortion model, path planning method is developed to tailor the laser scanning path for a given geometry to prevent large residual deformation and building failures. Gradient based path planning for continuous and island scanning strategy is formulated and full sensitivity analysis for the formulated compliance- and stress-minimization problem is performed.

    The feasibility and effectiveness of this proposed path planning method is demonstrated experimentally using the AconityONE L-PBF system. In addition, a data-driven framework utilizing machine learning is developed to predict the thermal history at part-scale for L-PBF.

    In this work, a sequential machine learning model including convolutional neural network (CNN) and recurrent neural network (RNN), long shortterm memory unit, is proposed for real-time thermal history prediction. A 100x prediction speed improvement is achieved compared to the finite element analysis which makes the prediction faster than real fabrication process and real-time temperature profile available.

    Figure 1.1: Schematic Overview of Metal Laser Powder Bed Fusion Process [2]
    Figure 1.1: Schematic Overview of Metal Laser Powder Bed Fusion Process [2]
    Figure 1.2: Commercial Powder Bed Fusion Systems
    Figure 1.2: Commercial Powder Bed Fusion Systems
    Figure 1.3: Commercial Metal Components Fabricated by Powder Bed Fusion Additive Manufacturing: (a) GE Fuel Nozzle; (b) Stryker Hip Biomedical Implant.
    Figure 1.3: Commercial Metal Components Fabricated by Powder Bed Fusion Additive Manufacturing: (a) GE Fuel Nozzle; (b) Stryker Hip Biomedical Implant.
    Figure 2.1: Proposed Multiscale Process Simulation Framework
    Figure 2.1: Proposed Multiscale Process Simulation Framework
    Figure 2.2: (a) Experimental Setup for In-situ Thermocouple Measurement in the EOS M290 Build Chamber; (b) Themocouple Locations on the Bottom Side of the Substrate.
    Figure 2.2: (a) Experimental Setup for In-situ Thermocouple Measurement in the EOS M290 Build Chamber; (b) Themocouple Locations on the Bottom Side of the Substrate.
    Figure 2.3: (a) Finite Element Model for Single Layer Thermal Analysis; (b) Deposition Layer
    Figure 2.3: (a) Finite Element Model for Single Layer Thermal Analysis; (b) Deposition Layer
    Figure 2.4: Core-skin layer: (a) Surface Morphology; (b) Scanning Strategy; (c) Transient Temperature Distribution and Temperature History at (d) Point 1; (e) Point 2 and (f) Point 3
    Figure 2.4: Core-skin layer: (a) Surface Morphology; (b) Scanning Strategy; (c) Transient Temperature Distribution and Temperature History at (d) Point 1; (e) Point 2 and (f) Point 3
    Figure 2.5: (a) Scanning Orientation of Each Layer; (b) Finite Element Model for Micro-scale Representative Volume
    Figure 2.5: (a) Scanning Orientation of Each Layer; (b) Finite Element Model for Micro-scale Representative Volume
    Figure 2.6: Bottom Layer (a) Thermal History; (b) Plastic Strain and (c) Elastic Strain Evolution History
    Figure 2.6: Bottom Layer (a) Thermal History; (b) Plastic Strain and (c) Elastic Strain Evolution History
    Figure 2.7: Bottom Layer Inherent Strain under Default Process Parameters along Horizontal Scanning Path
    Figure 2.7: Bottom Layer Inherent Strain under Default Process Parameters along Horizontal Scanning Path
    Figure 2.8: Snapshots of the Element Activation Process
    Figure 2.8: Snapshots of the Element Activation Process
    Figure 2.9: Double Cantilever Beam Structure Built by the EOS M290 DMLM Process (a) Before and (b) After Cutting off; (c) Faro Laser ScanArm V3 for Distortion Measurement
    Figure 2.9: Double Cantilever Beam Structure Built by the EOS M290 DMLM Process (a) Before and (b) After Cutting off; (c) Faro Laser ScanArm V3 for Distortion Measurement
    Figure 2.10: Square Canonical Structure Built by the EOS M290 DMLM Process
    Figure 2.10: Square Canonical Structure Built by the EOS M290 DMLM Process
    Figure 2.11: Finite Element Mesh for the Square Canonical and Snapshots of Element Activation Process
    Figure 2.11: Finite Element Mesh for the Square Canonical and Snapshots of Element Activation Process
    Figure 2.12: Simulated Distortion Field for the Double Cantilever Beam before Cutting off the Supports: (a) Inherent Strain Method; (b) Simufact Additive 3.1
    Figure 2.12: Simulated Distortion Field for the Double Cantilever Beam before Cutting off the Supports: (a) Inherent Strain Method; (b) Simufact Additive 3.1
    Figure 3.10: Snapshots of Temperature Profile for Single Track in Keyhole Regime (P = 250W and V = 0.5m/s) at the Preheating Temperature of 100 °C
    Figure 3.10: Snapshots of Temperature Profile for Single Track in Keyhole Regime (P = 250W and V = 0.5m/s) at the Preheating Temperature of 100 °C
    s) at the Preheating Temperature of 500 °C
    s) at the Preheating Temperature of 500 °C
    Figure 3.15: Melt Pool Cross Section Comparison Between Simulation and Experiment for Single Track
    Figure 3.15: Melt Pool Cross Section Comparison Between Simulation and Experiment for Single Track

    Bibliography

    [1] I. Astm, ASTM52900-15 Standard Terminology for Additive Manufacturing—General
    Principles—Terminology, ASTM International, West Conshohocken, PA 3(4) (2015) 5.
    [2] W.E. King, A.T. Anderson, R.M. Ferencz, N.E. Hodge, C. Kamath, S.A. Khairallah, A.M.
    Rubenchik, Laser powder bed fusion additive manufacturing of metals; physics, computational,
    and materials challenges, Applied Physics Reviews 2(4) (2015) 041304.
    [3] W. Yan, Y. Lu, K. Jones, Z. Yang, J. Fox, P. Witherell, G. Wagner, W.K. Liu, Data-driven
    characterization of thermal models for powder-bed-fusion additive manufacturing, Additive
    Manufacturing (2020) 101503.
    [4] K. Dai, L. Shaw, Thermal and stress modeling of multi-material laser processing, Acta
    Materialia 49(20) (2001) 4171-4181.
    [5] K. Dai, L. Shaw, Distortion minimization of laser-processed components through control of
    laser scanning patterns, Rapid Prototyping Journal 8(5) (2002) 270-276.
    [6] S.S. Bo Cheng, Kevin Chou, Stress and deformation evaluations of scanning strategy effect in
    selective laser melting, Additive Manufacturing (2017).
    [7] C. Fu, Y. Guo, Three-dimensional temperature gradient mechanism in selective laser melting
    of Ti-6Al-4V, Journal of Manufacturing Science and Engineering 136(6) (2014) 061004.
    [8] P. Prabhakar, W.J. Sames, R. Dehoff, S.S. Babu, Computational modeling of residual stress
    formation during the electron beam melting process for Inconel 718, Additive Manufacturing 7
    (2015) 83-91.
    [9] A. Hussein, L. Hao, C. Yan, R. Everson, Finite element simulation of the temperature and
    stress fields in single layers built without-support in selective laser melting, Materials & Design
    (1980-2015) 52 (2013) 638-647.
    [10] P.Z. Qingcheng Yang, Lin Cheng, Zheng Min, Minking Chyu, Albert C. To, articleFinite
    element modeling and validation of thermomechanicalbehavior of Ti-6Al-4V in directed energy
    deposition additivemanufacturing, Additive Manufacturing (2016).
    [11] E.R. Denlinger, J. Irwin, P. Michaleris, Thermomechanical Modeling of Additive
    Manufacturing Large Parts, Journal of Manufacturing Science and Engineering 136(6) (2014)
    061007.
    [12] E.R. Denlinger, M. Gouge, J. Irwin, P. Michaleris, Thermomechanical model development
    and in situ experimental validation of the Laser Powder-Bed Fusion process, Additive
    Manufacturing 16 (2017) 73-80.
    [13] V.J. Erik R Denlinger, G.V. Srinivasan, Tahany EI-Wardany, Pan Michaleris, Thermal
    modeling of Inconel 718 processed with powder bed fusionand experimental validation using in
    situ measurements, Additive Manufacturing 11 (2016) 7-15.
    [14] N. Patil, D. Pal, H.K. Rafi, K. Zeng, A. Moreland, A. Hicks, D. Beeler, B. Stucker, A
    Generalized Feed Forward Dynamic Adaptive Mesh Refinement and Derefinement Finite Element
    Framework for Metal Laser Sintering—Part I: Formulation and Algorithm Development, Journal
    of Manufacturing Science and Engineering 137(4) (2015) 041001.
    [15] D. Pal, N. Patil, K.H. Kutty, K. Zeng, A. Moreland, A. Hicks, D. Beeler, B. Stucker, A
    Generalized Feed-Forward Dynamic Adaptive Mesh Refinement and Derefinement FiniteElement Framework for Metal Laser Sintering—Part II: Nonlinear Thermal Simulations and
    Validations, Journal of Manufacturing Science and Engineering 138(6) (2016) 061003.
    [16] N. Keller, V. Ploshikhin, New method for fast predictions of residual stress and distortion of
    AM parts, Solid Freeform Fabrication Symposium, Austin, Texas, 2014, pp. 1229-1237.
    [17] S.A. Khairallah, A.T. Anderson, 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 Materialia 108 (2016) 36-45.
    [18] 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 Materialia 114
    (2016) 33-42.
    [19] A.A. Martin, N.P. Calta, S.A. Khairallah, J. Wang, P.J. Depond, A.Y. Fong, V. Thampy, G.M.
    Guss, A.M. Kiss, K.H. Stone, Dynamics of pore formation during laser powder bed fusion additive
    manufacturing, Nature communications 10(1) (2019) 1987.
    [20] R. Shi, S.A. Khairallah, T.T. Roehling, T.W. Heo, J.T. McKeown, M.J. Matthews,
    Microstructural control in metal laser powder bed fusion additive manufacturing using laser beam
    shaping strategy, Acta Materialia (2019).
    [21] S.A. Khairallah, A.A. Martin, J.R. Lee, G. Guss, N.P. Calta, J.A. Hammons, M.H. Nielsen,
    K. Chaput, E. Schwalbach, M.N. Shah, Controlling interdependent meso-nanosecond dynamics
    and defect generation in metal 3D printing, Science 368(6491) (2020) 660-665.
    [22] W. Yan, W. Ge, Y. Qian, S. Lin, B. Zhou, W.K. Liu, F. Lin, G.J. Wagner, Multi-physics
    modeling of single/multiple-track defect mechanisms in electron beam selective melting, Acta
    Materialia 134 (2017) 324-333.
    [23] S. Shrestha, Y. Kevin Chou, A Numerical Study on the Keyhole Formation During Laser
    Powder Bed Fusion Process, Journal of Manufacturing Science and Engineering 141(10) (2019).
    [24] S. Shrestha, B. Cheng, K. Chou, An Investigation into Melt Pool Effective Thermal
    Conductivity for Thermal Modeling of Powder-Bed Electron Beam Additive Manufacturing.
    [25] D. Rosenthal, Mathematical theory of heat distribution during welding and cutting, Welding
    journal 20 (1941) 220-234.
    [26] P. Promoppatum, S.-C. Yao, P.C. Pistorius, A.D. Rollett, A comprehensive comparison of the
    analytical and numerical prediction of the thermal history and solidification microstructure of
    Inconel 718 products made by laser powder-bed fusion, Engineering 3(5) (2017) 685-694.
    [27] M. Tang, P.C. Pistorius, J.L. Beuth, Prediction of lack-of-fusion porosity for powder bed
    fusion, Additive Manufacturing 14 (2017) 39-48.
    [28] T. Moran, P. Li, D. Warner, N. Phan, Utility of superposition-based finite element approach
    for part-scale thermal simulation in additive manufacturing, Additive Manufacturing 21 (2018)
    215-219.
    [29] Y. Yang, M. Knol, F. van Keulen, C. Ayas, A semi-analytical thermal modelling approach
    for selective laser melting, Additive Manufacturing 21 (2018) 284-297.
    [30] B. Cheng, S. Shrestha, K. Chou, Stress and deformation evaluations of scanning strategy
    effect in selective laser melting, Additive Manufacturing 12 (2016) 240-251.
    [31] L.H. Ahmed Hussein, Chunze Yan, Richard Everson, Finite element simulation of the
    temperature and stress fields in single layers built without-support in selective laser melting,
    Materials and Design 52 (2013) 638-647.
    [32] H. Peng, D.B. Go, R. Billo, S. Gong, M.R. Shankar, B.A. Gatrell, J. Budzinski, P. Ostiguy,
    R. Attardo, C. Tomonto, Part-scale model for fast prediction of thermal distortion in DMLS
    additive manufacturing; Part 2: a quasi-static thermo-mechanical model, Austin, Texas (2016).
    [33] M.F. Zaeh, G. Branner, Investigations on residual stresses and deformations in selective laser
    melting, Production Engineering 4(1) (2010) 35-45.
    [34] C. Li, C. Fu, Y. Guo, F. Fang, A multiscale modeling approach for fast prediction of part
    distortion in selective laser melting, Journal of Materials Processing Technology 229 (2016) 703-
    712.
    [35] C. Li, Z. Liu, X. Fang, Y. Guo, On the Simulation Scalability of Predicting Residual Stress
    and Distortion in Selective Laser Melting, Journal of Manufacturing Science and Engineering
    140(4) (2018) 041013.
    [36] S. Afazov, W.A. Denmark, B.L. Toralles, A. Holloway, A. Yaghi, Distortion Prediction and
    Compensation in Selective Laser Melting, Additive Manufacturing 17 (2017) 15-22.
    [37] Y. Lee, W. Zhang, Modeling of heat transfer, fluid flow and solidification microstructure of
    nickel-base superalloy fabricated by laser powder bed fusion, Additive Manufacturing 12 (2016)
    178-188.
    [38] L. Scime, J. Beuth, A multi-scale convolutional neural network for autonomous anomaly
    detection and classification in a laser powder bed fusion additive manufacturing process, Additive
    Manufacturing 24 (2018) 273-286.
    [39] L. Scime, J. Beuth, Using machine learning to identify in-situ melt pool signatures indicative
    of flaw formation in a laser powder bed fusion additive manufacturing process, Additive
    Manufacturing 25 (2019) 151-165.
    [40] X. Xie, J. Bennett, S. Saha, Y. Lu, J. Cao, W.K. Liu, Z. Gan, Mechanistic data-driven
    prediction of as-built mechanical properties in metal additive manufacturing, npj Computational
    Materials 7(1) (2021) 1-12.
    [41] C. Wang, X. Tan, S. Tor, C. Lim, Machine learning in additive manufacturing: State-of-theart and perspectives, Additive Manufacturing (2020) 101538.
    [42] J. Li, R. Jin, Z.Y. Hang, Integration of physically-based and data-driven approaches for
    thermal field prediction in additive manufacturing, Materials & Design 139 (2018) 473-485.
    [43] M. Mozaffar, A. Paul, R. Al-Bahrani, S. Wolff, A. Choudhary, A. Agrawal, K. Ehmann, J.
    Cao, Data-driven prediction of the high-dimensional thermal history in directed energy deposition
    processes via recurrent neural networks, Manufacturing letters 18 (2018) 35-39.
    [44] A. Paul, M. Mozaffar, Z. Yang, W.-k. Liao, A. Choudhary, J. Cao, A. Agrawal, A real-time
    iterative machine learning approach for temperature profile prediction in additive manufacturing
    processes, 2019 IEEE International Conference on Data Science and Advanced Analytics (DSAA),
    IEEE, 2019, pp. 541-550.
    [45] S. Clijsters, T. Craeghs, J.-P. Kruth, A priori process parameter adjustment for SLM process
    optimization, Innovative developments on virtual and physical prototyping, Taylor & Francis
    Group., 2012, pp. 553-560.
    [46] R. Mertens, S. Clijsters, K. Kempen, J.-P. Kruth, Optimization of scan strategies in selective
    laser melting of aluminum parts with downfacing areas, Journal of Manufacturing Science and
    Engineering 136(6) (2014) 061012.
    [47] J.-P. Kruth, J. Deckers, E. Yasa, R. Wauthlé, Assessing and comparing influencing factors of
    residual stresses in selective laser melting using a novel analysis method, Proceedings of the
    institution of mechanical engineers, Part B: Journal of Engineering Manufacture 226(6) (2012)
    980-991.
    [48] Y. Lu, S. Wu, Y. Gan, T. Huang, C. Yang, L. Junjie, J. Lin, Study on the microstructure,
    mechanical property and residual stress of SLM Inconel-718 alloy manufactured by differing
    island scanning strategy, Optics & Laser Technology 75 (2015) 197-206.
    [49] E. Foroozmehr, R. Kovacevic, Effect of path planning on the laser powder deposition process:
    thermal and structural evaluation, The International Journal of Advanced Manufacturing
    Technology 51(5-8) (2010) 659-669.
    [50] L.H. Ahmed Hussein, Chunze Yan, Richard Everson, Finite element simulation of the
    temperature and stress fields in single layers built without-support in selective laser melting,
    Materials and Design (2013).
    [51] J.-P. Kruth, M. Badrossamay, E. Yasa, J. Deckers, L. Thijs, J. Van Humbeeck, Part and
    material properties in selective laser melting of metals, Proceedings of the 16th international
    symposium on electromachining, 2010, pp. 1-12.
    [52] L. Thijs, K. Kempen, J.-P. Kruth, J. Van Humbeeck, Fine-structured aluminium products with
    controllable texture by selective laser melting of pre-alloyed AlSi10Mg powder, Acta Materialia
    61(5) (2013) 1809-1819.
    [53] D. Ding, Z.S. Pan, D. Cuiuri, H. Li, A tool-path generation strategy for wire and arc additive
    manufacturing, The international journal of advanced manufacturing technology 73(1-4) (2014)
    173-183.
    [54] B.E. Carroll, T.A. Palmer, A.M. Beese, Anisotropic tensile behavior of Ti–6Al–4V
    components fabricated with directed energy deposition additive manufacturing, Acta Materialia
    87 (2015) 309-320.
    [55] D. Ding, Z. Pan, D. Cuiuri, H. Li, A practical path planning methodology for wire and arc
    additive manufacturing of thin-walled structures, Robotics and Computer-Integrated
    Manufacturing 34 (2015) 8-19.
    [56] D. Ding, Z. Pan, D. Cuiuri, H. Li, S. van Duin, N. Larkin, Bead modelling and implementation
    of adaptive MAT path in wire and arc additive manufacturing, Robotics and Computer-Integrated
    Manufacturing 39 (2016) 32-42.
    [57] R. Ponche, O. Kerbrat, P. Mognol, J.-Y. Hascoet, A novel methodology of design for Additive
    Manufacturing applied to Additive Laser Manufacturing process, Robotics and ComputerIntegrated Manufacturing 30(4) (2014) 389-398.
    [58] D.E. Smith, R. Hoglund, Continuous fiber angle topology optimization for polymer fused
    fillament fabrication, Annu. Int. Solid Free. Fabr. Symp. Austin, TX, 2016.
    [59] J. Liu, J. Liu, H. Yu, H. Yu, Concurrent deposition path planning and structural topology
    optimization for additive manufacturing, Rapid Prototyping Journal 23(5) (2017) 930-942.
    [60] Q. Xia, T. Shi, Optimization of composite structures with continuous spatial variation of fiber
    angle through Shepard interpolation, Composite Structures 182 (2017) 273-282.
    [61] C. Kiyono, E. Silva, J. Reddy, A novel fiber optimization method based on normal distribution
    function with continuously varying fiber path, Composite Structures 160 (2017) 503-515.
    [62] C.J. Brampton, K.C. Wu, H.A. Kim, New optimization method for steered fiber composites
    using the level set method, Structural and Multidisciplinary Optimization 52(3) (2015) 493-505.
    [63] J. Liu, A.C. To, Deposition path planning-integrated structural topology optimization for 3D
    additive manufacturing subject to self-support constraint, Computer-Aided Design 91 (2017) 27-
    45.
    [64] H. Shen, J. Fu, Z. Chen, Y. Fan, Generation of offset surface for tool path in NC machining
    through level set methods, The International Journal of Advanced Manufacturing Technology
    46(9-12) (2010) 1043-1047.
    [65] C. Zhuang, Z. Xiong, H. Ding, High speed machining tool path generation for pockets using
    level sets, International Journal of Production Research 48(19) (2010) 5749-5766.
    [66] K.C. Mills, Recommended values of thermophysical properties for selected commercial
    alloys, Woodhead Publishing2002.
    [67] S.S. Sih, J.W. Barlow, The prediction of the emissivity and thermal conductivity of powder
    beds, Particulate Science and Technology 22(4) (2004) 427-440.
    [68] L. Dong, A. Makradi, S. Ahzi, Y. Remond, Three-dimensional transient finite element
    analysis of the selective laser sintering process, Journal of materials processing technology 209(2)
    (2009) 700-706.
    [69] J.J. Beaman, J.W. Barlow, D.L. Bourell, R.H. Crawford, H.L. Marcus, K.P. McAlea, Solid
    freeform fabrication: a new direction in manufacturing, Kluwer Academic Publishers, Norwell,
    MA 2061 (1997) 25-49.
    [70] G. Bugeda Miguel Cervera, G. Lombera, Numerical prediction of temperature and density
    distributions in selective laser sintering processes, Rapid Prototyping Journal 5(1) (1999) 21-26.
    [71] T. Mukherjee, W. Zhang, T. DebRoy, An improved prediction of residual stresses and
    distortion in additive manufacturing, Computational Materials Science 126 (2017) 360-372.
    [72] A.J. Dunbar, E.R. Denlinger, M.F. Gouge, P. Michaleris, Experimental validation of finite
    element modeling for laser powderbed fusion deformation, Additive Manufacturing 12 (2016)
    108-120.
    [73] J. Goldak, A. Chakravarti, M. Bibby, A new finite element model for welding heat sources,
    Metallurgical and Materials Transactions B 15(2) (1984) 299-305.
    [74] J. Liu, Q. Chen, Y. Zhao, W. Xiong, A. To, Quantitative Texture Prediction of Epitaxial
    Columnar Grains in Alloy 718 Processed by Additive Manufacturing, Proceedings of the 9th
    International Symposium on Superalloy 718 & Derivatives: Energy, Aerospace, and Industrial
    Applications, Springer, 2018, pp. 749-755.
    [75] J. Irwin, P. Michaleris, A line heat input model for additive manufacturing, Journal of
    Manufacturing Science and Engineering 138(11) (2016) 111004.
    [76] M. Gouge, J. Heigel, P. Michaleris, T. Palmer, Modeling forced convection in the thermal
    simulation of laser cladding processes, International Journal of Advanced Manufacturing
    Technology 79 (2015).
    [77] J. Heigel, P. Michaleris, E. Reutzel, Thermo-mechanical model development and validation
    of directed energy deposition additive manufacturing of Ti–6Al–4V, Additive manufacturing 5
    (2015) 9-19.
    [78] E.R. Denlinger, J.C. Heigel, P. Michaleris, Residual stress and distortion modeling of electron
    beam direct manufacturing Ti-6Al-4V, Proceedings of the Institution of Mechanical Engineers,
    Part B: Journal of Engineering Manufacture 229(10) (2015) 1803-1813.
    [79] X. Liang, Q. Chen, L. Cheng, Q. Yang, A. To, A modified inherent strain method for fast
    prediction of residual deformation in additive manufacturing of metal parts, 2017 Solid Freeform
    Fabrication Symposium Proceedings, Austin, Texas, 2017.
    [80] X. Liang, L. Cheng, Q. Chen, Q. Yang, A. To, A Modified Method for Estimating Inherent
    Strains from Detailed Process Simulation for Fast Residual Distortion Prediction of Single-Walled
    Structures Fabricated by Directed Energy Deposition, Additive Manufacturing 23 (2018) 471-486.
    [81] L. Sochalski-Kolbus, E.A. Payzant, P.A. Cornwell, T.R. Watkins, S.S. Babu, R.R. Dehoff,
    M. Lorenz, O. Ovchinnikova, C. Duty, Comparison of residual stresses in Inconel 718 simple parts
    made by electron beam melting and direct laser metal sintering, Metallurgical and Materials
    Transactions A 46(3) (2015) 1419-1432.
    [82] P. Mercelis, J.-P. Kruth, Residual stresses in selective laser sintering and selective laser
    melting, Rapid Prototyping Journal 12(5) (2006) 254-265.
    [83] N. Hodge, R. Ferencz, J. Solberg, Implementation of a thermomechanical model for the
    simulation of selective laser melting, Computational Mechanics 54(1) (2014) 33-51.
    [84] A.S. Wu, D.W. Brown, M. Kumar, G.F. Gallegos, W.E. King, An experimental investigation
    into additive manufacturing-induced residual stresses in 316L stainless steel, Metallurgical and
    Materials Transactions A 45(13) (2014) 6260-6270.
    [85] C. Li, J. liu, Y. Guo, Efficient predictive model of part distortion and residual stress in
    selective laser melting, Solid Freeform Fabrication 2016, 2017.
    [86] Y. Zhao, Y. Koizumi, K. Aoyagi, D. Wei, K. Yamanaka, A. Chiba, Molten pool behavior and
    effect of fluid flow on solidification conditions in selective electron beam melting (SEBM) of a
    biomedical Co-Cr-Mo alloy, Additive Manufacturing 26 (2019) 202-214.
    [87] J.-H. Cho, S.-J. Na, Implementation of real-time multiple reflection and Fresnel absorption of
    laser beam in keyhole, Journal of Physics D: Applied Physics 39(24) (2006) 5372.
    [88] Q. Guo, C. Zhao, M. Qu, L. Xiong, L.I. Escano, S.M.H. Hojjatzadeh, N.D. Parab, K. Fezzaa,
    W. Everhart, T. Sun, In-situ characterization and quantification of melt pool variation under
    constant input energy density in laser powder-bed fusion additive manufacturing process, Additive
    Manufacturing (2019).
    [89] E. Assuncao, S. Williams, D. Yapp, Interaction time and beam diameter effects on the
    conduction mode limit, Optics and Lasers in Engineering 50(6) (2012) 823-828.
    [90] R. Cunningham, C. Zhao, N. Parab, C. Kantzos, J. Pauza, K. Fezzaa, T. Sun, A.D. Rollett,
    Keyhole threshold and morphology in laser melting revealed by ultrahigh-speed x-ray imaging,
    Science 363(6429) (2019) 849-852.
    [91] W. Tan, N.S. Bailey, Y.C. Shin, Investigation of keyhole plume and molten pool based on a
    three-dimensional dynamic model with sharp interface formulation, Journal of Physics D: Applied
    Physics 46(5) (2013) 055501.
    [92] W. Tan, Y.C. Shin, Analysis of multi-phase interaction and its effects on keyhole dynamics
    with a multi-physics numerical model, Journal of Physics D: Applied Physics 47(34) (2014)
    345501.
    [93] R. Fabbro, K. Chouf, Keyhole modeling during laser welding, Journal of applied Physics
    87(9) (2000) 4075-4083.
    [94] Q. Guo, C. Zhao, M. Qu, L. Xiong, S.M.H. Hojjatzadeh, L.I. Escano, N.D. Parab, K. Fezzaa,
    T. Sun, L. Chen, In-situ full-field mapping of melt flow dynamics in laser metal additive
    manufacturing, Additive Manufacturing 31 (2020) 100939.
    [95] Y. Ueda, K. Fukuda, K. Nakacho, S. Endo, A new measuring method of residual stresses with
    the aid of finite element method and reliability of estimated values, Journal of the Society of Naval
    Architects of Japan 1975(138) (1975) 499-507.
    [96] M.R. Hill, D.V. Nelson, The inherent strain method for residual stress determination and its
    application to a long welded joint, ASME-PUBLICATIONS-PVP 318 (1995) 343-352.
    [97] H. Murakawa, Y. Luo, Y. Ueda, Prediction of welding deformation and residual stress by
    elastic FEM based on inherent strain, Journal of the society of Naval Architects of Japan 1996(180)
    (1996) 739-751.
    [98] M. Yuan, Y. Ueda, Prediction of residual stresses in welded T-and I-joints using inherent
    strains, Journal of Engineering Materials and Technology, Transactions of the ASME 118(2)
    (1996) 229-234.
    [99] L. Zhang, P. Michaleris, P. Marugabandhu, Evaluation of applied plastic strain methods for
    welding distortion prediction, Journal of Manufacturing Science and Engineering 129(6) (2007)
    1000-1010.
    [100] M. Bugatti, Q. Semeraro, Limitations of the Inherent Strain Method in Simulating Powder
    Bed Fusion Processes, Additive Manufacturing 23 (2018) 329-346.
    [101] L. Cheng, X. Liang, J. Bai, Q. Chen, J. Lemon, A. To, On Utilizing Topology Optimization
    to Design Support Structure to Prevent Residual Stress Induced Build Failure in Laser Powder Bed
    Metal Additive Manufacturing, Additive Manufacturing (2019).
    [102] Q. Chen, X. Liang, D. Hayduke, J. Liu, L. Cheng, J. Oskin, R. Whitmore, A.C. To, An
    inherent strain based multiscale modeling framework for simulating part-scale residual
    deformation for direct metal laser sintering, Additive Manufacturing 28 (2019) 406-418.
    [103] S. Osher, J.A. Sethian, Fronts propagating with curvature-dependent speed: algorithms based
    on Hamilton-Jacobi formulations, Journal of computational physics 79(1) (1988) 12-49.
    [104] M.Y. Wang, X. Wang, D. Guo, A level set method for structural topology optimization,
    Computer methods in applied mechanics and engineering 192(1) (2003) 227-246.
    [105] G. Allaire, F. Jouve, A.-M. Toader, Structural optimization using sensitivity analysis and a
    level-set method, Journal of computational physics 194(1) (2004) 363-393.
    [106] Y. Wang, Z. Luo, Z. Kang, N. Zhang, A multi-material level set-based topology and shape
    optimization method, Computer Methods in Applied Mechanics and Engineering 283 (2015)
    1570-1586.
    [107] P. Dunning, C. Brampton, H. Kim, Simultaneous optimisation of structural topology and
    material grading using level set method, Materials Science and Technology 31(8) (2015) 884-894.
    [108] P. Liu, Y. Luo, Z. Kang, Multi-material topology optimization considering interface
    behavior via XFEM and level set method, Computer methods in applied mechanics and
    engineering 308 (2016) 113-133.
    [109] J. Liu, Q. Chen, Y. Zheng, R. Ahmad, J. Tang, Y. Ma, Level set-based heterogeneous object
    modeling and optimization, Computer-Aided Design (2019).
    [110] J. Liu, Q. Chen, X. Liang, A.C. To, Manufacturing cost constrained topology optimization
    for additive manufacturing, Frontiers of Mechanical Engineering 14(2) (2019) 213-221.
    [111] Z. Kang, Y. Wang, Integrated topology optimization with embedded movable holes based
    on combined description by material density and level sets, Computer methods in applied
    mechanics and engineering 255 (2013) 1-13.
    [112] P.D. Dunning, H. Alicia Kim, A new hole insertion method for level set based structural
    topology optimization, International Journal for Numerical Methods in Engineering 93(1) (2013)
    118-134.
    [113] J.A. Sethian, A fast marching level set method for monotonically advancing fronts,
    Proceedings of the National Academy of Sciences 93(4) (1996) 1591-1595.
    [114] J.A. Sethian, Level set methods and fast marching methods: evolving interfaces in
    computational geometry, fluid mechanics, computer vision, and materials science, Cambridge
    university press1999.
    [115] C. Le, J. Norato, T. Bruns, C. Ha, D. Tortorelli, Stress-based topology optimization for
    continua, Structural and Multidisciplinary Optimization 41(4) (2010) 605-620.
    [116] A. Takezawa, G.H. Yoon, S.H. Jeong, M. Kobashi, M. Kitamura, Structural topology
    optimization with strength and heat conduction constraints, Computer Methods in Applied
    Mechanics and Engineering 276 (2014) 341-361.
    [117] S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural computation 9(8) (1997)
    1735-1780.
    [118] A. Krizhevsky, I. Sutskever, G.E. Hinton, Imagenet classification with deep convolutional
    neural networks, Advances in neural information processing systems 25 (2012) 1097-1105.
    [119] K. Simonyan, A. Zisserman, Very deep convolutional networks for large-scale image
    recognition, arXiv preprint arXiv:1409.1556 (2014).
    [120] K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, Proceedings
    of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770-778.
    [121] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A.
    Khosla, M. Bernstein, Imagenet large scale visual recognition challenge, International journal of
    computer vision 115(3) (2015) 211-252.
    [122] S. Ren, K. He, R. Girshick, J. Sun, Faster r-cnn: Towards real-time object detection with
    region proposal networks, Advances in neural information processing systems 28 (2015) 91-99.
    [123] E.J. Schwalbach, S.P. Donegan, M.G. Chapman, K.J. Chaput, M.A. Groeber, A discrete
    source model of powder bed fusion additive manufacturing thermal history, Additive
    Manufacturing 25 (2019) 485-498.
    [124] D.G. Duffy, Green’s functions with applications, Chapman and Hall/CRC2015.
    [125] J. Martínez-Frutos, D. Herrero-Pérez, Efficient matrix-free GPU implementation of fixed
    grid finite element analysis, Finite Elements in Analysis and Design 104 (2015) 61-71.
    [126] F. Dugast, P. Apostolou, A. Fernandez, W. Dong, Q. Chen, S. Strayer, R. Wicker, A.C. To,
    Part-scale thermal process modeling for laser powder bed fusion with matrix-free method and GPU
    computing, Additive Manufacturing 37 (2021) 101732.
    [127] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A.N. Gomez, Ł. Kaiser, I.
    Polosukhin, Attention is all you need, Advances in neural information processing systems, 2017,
    pp. 5998-6008.
    [128] J. Devlin, M.-W. Chang, K. Lee, K. Toutanova, Bert: Pre-training of deep bidirectional
    transformers for language understanding, arXiv preprint arXiv:1810.04805 (2018).

    Figure 3 Simulation PTC pipes enhanced with copper foam and nanoparticles in FLOW-3D software.

    다공성 미디어 및 나노유체에 의해 강화된 수집기로 태양광 CCHP 시스템의 최적화

    Optimization of Solar CCHP Systems with Collector Enhanced by Porous Media and Nanofluid


    Navid Tonekaboni,1Mahdi Feizbahr,2 Nima Tonekaboni,1Guang-Jun Jiang,3,4 and Hong-Xia Chen3,4

    Abstract

    태양열 집열기의 낮은 효율은 CCHP(Solar Combined Cooling, Heating, and Power) 사이클의 문제점 중 하나로 언급될 수 있습니다. 태양계를 개선하기 위해 나노유체와 다공성 매체가 태양열 집열기에 사용됩니다.

    다공성 매질과 나노입자를 사용하는 장점 중 하나는 동일한 조건에서 더 많은 에너지를 흡수할 수 있다는 것입니다. 이 연구에서는 평균 일사량이 1b인 따뜻하고 건조한 지역의 600 m2 건물의 전기, 냉방 및 난방을 생성하기 위해 다공성 매질과 나노유체를 사용하여 태양열 냉난방 복합 발전(SCCHP) 시스템을 최적화했습니다.

    본 논문에서는 침전물이 형성되지 않는 lb = 820 w/m2(이란) 정도까지 다공성 물질에서 나노유체의 최적량을 계산하였다. 이 연구에서 태양열 집열기는 구리 다공성 매체(95% 다공성)와 CuO 및 Al2O3 나노 유체로 향상되었습니다.

    나노유체의 0.1%-0.6%가 작동 유체로 물에 추가되었습니다. 나노유체의 0.5%가 태양열 집열기 및 SCCHP 시스템에서 가장 높은 에너지 및 엑서지 효율 향상으로 이어지는 것으로 밝혀졌습니다.

    본 연구에서 포물선형 집열기(PTC)의 최대 에너지 및 엑서지 효율은 각각 74.19% 및 32.6%입니다. 그림 1은 태양 CCHP의 주기를 정확하게 설명하기 위한 그래픽 초록으로 언급될 수 있습니다.

    The low efficiency of solar collectors can be mentioned as one of the problems in solar combined cooling, heating, and power (CCHP) cycles. For improving solar systems, nanofluid and porous media are used in solar collectors. One of the advantages of using porous media and nanoparticles is to absorb more energy under the same conditions. In this research, a solar combined cooling, heating, and power (SCCHP) system has been optimized by porous media and nanofluid for generating electricity, cooling, and heating of a 600 m2 building in a warm and dry region with average solar radiation of Ib = 820 w/m2 in Iran. In this paper, the optimal amount of nanofluid in porous materials has been calculated to the extent that no sediment is formed. In this study, solar collectors were enhanced with copper porous media (95% porosity) and CuO and Al2O3 nanofluids. 0.1%–0.6% of the nanofluids were added to water as working fluids; it is found that 0.5% of the nanofluids lead to the highest energy and exergy efficiency enhancement in solar collectors and SCCHP systems. Maximum energy and exergy efficiency of parabolic thermal collector (PTC) riches in this study are 74.19% and 32.6%, respectively. Figure 1 can be mentioned as a graphical abstract for accurately describing the cycle of solar CCHP.

    1. Introduction

    Due to the increase in energy consumption, the use of clean energy is one of the important goals of human societies. In the last four decades, the use of cogeneration cycles has increased significantly due to high efficiency. Among clean energy, the use of solar energy has become more popular due to its greater availability [1]. Low efficiency of energy production, transmission, and distribution system makes a new system to generate simultaneously electricity, heating, and cooling as an essential solution to be widely used. The low efficiency of the electricity generation, transmission, and distribution system makes the CCHP system a basic solution to eliminate waste of energy. CCHP system consists of a prime mover (PM), a power generator, a heat recovery system (produce extra heating/cooling/power), and thermal energy storage (TES) [2]. Solar combined cooling, heating, and power (SCCHP) has been started three decades ago. SCCHP is a system that receives its propulsive force from solar energy; in this cycle, solar collectors play the role of propulsive for generating power in this system [3].

    Increasing the rate of energy consumption in the whole world because of the low efficiency of energy production, transmission, and distribution system causes a new cogeneration system to generate electricity, heating, and cooling energy as an essential solution to be widely used. Building energy utilization fundamentally includes power required for lighting, home electrical appliances, warming and cooling of building inside, and boiling water. Domestic usage contributes to an average of 35% of the world’s total energy consumption [4].

    Due to the availability of solar energy in all areas, solar collectors can be used to obtain the propulsive power required for the CCHP cycle. Solar energy is the main source of energy in renewable applications. For selecting a suitable area to use solar collectors, annual sunshine hours, the number of sunny days, minus temperature and frosty days, and the windy status of the region are essentially considered [5]. Iran, with an average of more than 300 sunny days, is one of the suitable countries to use solar energy. Due to the fact that most of the solar radiation is in the southern regions of Iran, also the concentration of cities is low in these areas, and transmission lines are far apart, one of the best options is to use CCHP cycles based on solar collectors [6]. One of the major problems of solar collectors is their low efficiency [7]. Low efficiency increases the area of collectors, which increases the initial cost of solar systems and of course increases the initial payback period. To increase the efficiency of solar collectors and improve their performance, porous materials and nanofluids are used to increase their workability.

    There are two ways to increase the efficiency of solar collectors and mechanical and fluid improvement. In the first method, using porous materials or helical filaments inside the collector pipes causes turbulence of the flow and increases heat transfer. In the second method, using nanofluids or salt and other materials increases the heat transfer of water. The use of porous materials has grown up immensely over the past twenty years. Porous materials, especially copper porous foam, are widely used in solar collectors. Due to the high contact surface area, porous media are appropriate candidates for solar collectors [8]. A number of researchers investigated Solar System performance in accordance with energy and exergy analyses. Zhai et al. [9] reviewed the performance of a small solar-powered system in which the energy efficiency was 44.7% and the electrical efficiency was 16.9%.

    Abbasi et al. [10] proposed an innovative multiobjective optimization to optimize the design of a cogeneration system. Results showed the CCHP system based on an internal diesel combustion engine was the applicable alternative at all regions with different climates. The diesel engine can supply the electrical requirement of 31.0% and heating demand of 3.8% for building.

    Jiang et al. [11] combined the experiment and simulation together to analyze the performance of a cogeneration system. Moreover, some research focused on CCHP systems using solar energy. It integrated sustainable and renewable technologies in the CCHP, like PV, Stirling engine, and parabolic trough collector (PTC) [21215].

    Wang et al. [16] optimized a cogeneration solar cooling system with a Rankine cycle and ejector to reach the maximum total system efficiency of 55.9%. Jing et al. analyzed a big-scale building with the SCCHP system and auxiliary heaters to produced electrical, cooling, and heating power. The maximum energy efficiency reported in their work is 46.6% [17]. Various optimization methods have been used to improve the cogeneration system, minimum system size, and performance, such as genetic algorithm [1819].

    Hirasawa et al. [20] investigated the effect of using porous media to reduce thermal waste in solar systems. They used the high-porosity metal foam on top of the flat plate solar collector and observed that thermal waste decreased by 7% due to natural heat transfer. Many researchers study the efficiency improvement of the solar collector by changing the collector’s shapes or working fluids. However, the most effective method is the use of nanofluids in the solar collector as working fluid [21]. In the experimental study done by Jouybari et al. [22], the efficiency enhancement up to 8.1% was achieved by adding nanofluid in a flat plate collector. In this research, by adding porous materials to the solar collector, collector efficiency increased up to 92% in a low flow regime. Subramani et al. [23] analyzed the thermal performance of the parabolic solar collector with Al2O3 nanofluid. They conducted their experiments with Reynolds number range 2401 to 7202 and mass flow rate 0.0083 to 0.05 kg/s. The maximum efficiency improvement in this experiment was 56% at 0.05 kg/s mass flow rate.

    Shojaeizadeh et al. [24] investigated the analysis of the second law of thermodynamic on the flat plate solar collector using Al2O3/water nanofluid. Their research showed that energy efficiency rose up to 1.9% and the exergy efficiency increased by a maximum of 0.72% compared to pure water. Tiwari et al. [25] researched on the thermal performance of solar flat plate collectors for working fluid water with different nanofluids. The result showed that using 1.5% (optimum) particle volume fraction of Al2O3 nanofluid as an absorbing medium causes the thermal efficiency to enhance up to 31.64%.

    The effect of porous media and nanofluids on solar collectors has already been investigated in the literature but the SCCHP system with a collector embedded by both porous media and nanofluid for enhancing the ratio of nanoparticle in nanofluid for preventing sedimentation was not discussed. In this research, the amount of energy and exergy of the solar CCHP cycles with parabolic solar collectors in both base and improved modes with a porous material (copper foam with 95% porosity) and nanofluid with different ratios of nanoparticles was calculated. In the first step, it is planned to design a CCHP system based on the required load, and, in the next step, it will analyze the energy and exergy of the system in a basic and optimize mode. In the optimize mode, enhanced solar collectors with porous material and nanofluid in different ratios (0.1%–0.7%) were used to optimize the ratio of nanofluids to prevent sedimentation.

    2. Cycle Description

    CCHP is one of the methods to enhance energy efficiency and reduce energy loss and costs. The SCCHP system used a solar collector as a prime mover of the cogeneration system and assisted the boiler to generate vapor for the turbine. Hot water flows from the expander to the absorption chiller in summer or to the radiator or fan coil in winter. Finally, before the hot water wants to flow back to the storage tank, it flows inside a heat exchanger for generating domestic hot water [26].

    For designing of solar cogeneration system and its analysis, it is necessary to calculate the electrical, heating (heating load is the load required for the production of warm water and space heating), and cooling load required for the case study considered in a residential building with an area of 600 m2 in the warm region of Iran (Zahedan). In Table 1, the average of the required loads is shown for the different months of a year (average of electrical, heating, and cooling load calculated with CARRIER software).Table 1 The average amount of electric charges, heating load, and cooling load used in the different months of the year in the city of Zahedan for a residential building with 600 m2.

    According to Table 1, the maximum magnitude of heating, cooling, and electrical loads is used to calculate the cogeneration system. The maximum electric load is 96 kW, the maximum amount of heating load is 62 kW, and the maximum cooling load is 118 kW. Since the calculated loads are average, all loads increased up to 10% for the confidence coefficient. With the obtained values, the solar collector area and other cogeneration system components are calculated. The cogeneration cycle is capable of producing 105 kW electric power, 140 kW cooling capacity, and 100 kW heating power.

    2.1. System Analysis Equations

    An analysis is done by considering the following assumptions:(1)The system operates under steady-state conditions(2)The system is designed for the warm region of Iran (Zahedan) with average solar radiation Ib = 820 w/m2(3)The pressure drops in heat exchangers, separators, storage tanks, and pipes are ignored(4)The pressure drop is negligible in all processes and no expectable chemical reactions occurred in the processes(5)Potential, kinetic, and chemical exergy are not considered due to their insignificance(6)Pumps have been discontinued due to insignificance throughout the process(7)All components are assumed adiabatic

    Schematic shape of the cogeneration cycle is shown in Figure 1 and all data are given in Table 2.

    Figure 1 Schematic shape of the cogeneration cycle.Table 2 Temperature and humidity of different points of system.

    Based on the first law of thermodynamic, energy analysis is based on the following steps.

    First of all, the estimated solar radiation energy on collector has been calculated:where α is the heat transfer enhancement coefficient based on porous materials added to the collector’s pipes. The coefficient α is increased by the porosity percentage, the type of porous material (in this case, copper with a porosity percentage of 95), and the flow of fluid to the collector equation.

    Collector efficiency is going to be calculated by the following equation [9]:

    Total energy received by the collector is given by [9]

    Also, the auxiliary boiler heat load is [2]

    Energy consumed from vapor to expander is calculated by [2]

    The power output form by the screw expander [9]:

    The efficiency of the expander is 80% in this case [11].

    In this step, cooling and heating loads were calculated and then, the required heating load to reach sanitary hot water will be calculated as follows:

    First step: calculating the cooling load with the following equation [9]:

    Second step: calculating heating loads [9]:

    Then, calculating the required loud for sanitary hot water will be [9]

    According to the above-mentioned equations, efficiency is [9]

    In the third step, calculated exergy analysis as follows.

    First, the received exergy collector from the sun is calculated [9]:

    In the previous equation, f is the constant of air dilution.

    The received exergy from the collector is [9]

    In the case of using natural gas in an auxiliary heater, the gas exergy is calculated from the following equation [12]:

    Delivering exergy from vapor to expander is calculated with the following equation [9]:

    In the fourth step, the exergy in cooling and heating is calculated by the following equation:

    Cooling exergy in summer is calculated [9]:

    Heating exergy in winter is calculated [9]:

    In the last step based on thermodynamic second law, exergy efficiency has been calculated from the following equation and the above-mentioned calculated loads [9]:

    3. Porous Media

    The porous medium that filled the test section is copper foam with a porosity of 95%. The foams are determined in Figure 2 and also detailed thermophysical parameters and dimensions are shown in Table 3.

    Figure 2 Copper foam with a porosity of 95%.Table 3 Thermophysical parameters and dimensions of copper foam.

    In solar collectors, copper porous materials are suitable for use at low temperatures and have an easier and faster manufacturing process than ceramic porous materials. Due to the high coefficient conductivity of copper, the use of copper metallic foam to increase heat transfer is certainly more efficient in solar collectors.

    Porous media and nanofluid in solar collector’s pipes were simulated in FLOW-3D software using the finite-difference method [27]. Nanoparticles Al2O3 and CUO are mostly used in solar collector enhancement. In this research, different concentrations of nanofluid are added to the parabolic solar collectors with porous materials (copper foam with porosity of 95%) to achieve maximum heat transfer in the porous materials before sedimentation. After analyzing PTC pipes with the nanofluid flow in FLOW-3D software, for energy and exergy efficiency analysis, Carrier software results were used as EES software input. Simulation PTC with porous media inside collector pipe and nanofluids sedimentation is shown in Figure 3.

    Figure 3 Simulation PTC pipes enhanced with copper foam and nanoparticles in FLOW-3D software.

    3.1. Nano Fluid

    In this research, copper and silver nanofluids (Al2O3, CuO) have been added with percentages of 0.1%–0.7% as the working fluids. The nanoparticle properties are given in Table 4. Also, system constant parameters are presented in Table 4, which are available as default input in the EES software.Table 4 Properties of the nanoparticles [9].

    System constant parameters for input in the software are shown in Table 5.Table 5 System constant parameters.

    The thermal properties of the nanofluid can be obtained from equations (18)–(21). The basic fluid properties are indicated by the index (bf) and the properties of the nanoparticle silver with the index (np).

    The density of the mixture is shown in the following equation [28]:where ρ is density and ϕ is the nanoparticles volume fraction.

    The specific heat capacity is calculated from the following equation [29]:

    The thermal conductivity of the nanofluid is calculated from the following equation [29]:

    The parameter β is the ratio of the nanolayer thickness to the original particle radius and, usually, this parameter is taken equal to 0.1 for the calculated thermal conductivity of the nanofluids.

    The mixture viscosity is calculated as follows [30]:

    In all equations, instead of water properties, working fluids with nanofluid are used. All of the above equations and parameters are entered in the EES software for calculating the energy and exergy of solar collectors and the SCCHP cycle. All calculation repeats for both nanofluids with different concentrations of nanofluid in the solar collector’s pipe.

    4. Results and Discussion

    In the present study, relations were written according to Wang et al. [16] and the system analysis was performed to ensure the correctness of the code. The energy and exergy charts are plotted based on the main values of the paper and are shown in Figures 4 and 5. The error rate in this simulation is 1.07%.

    Figure 4 Verification charts of energy analysis results.

    Figure 5 Verification charts of exergy analysis results.

    We may also investigate the application of machine learning paradigms [3141] and various hybrid, advanced optimization approaches that are enhanced in terms of exploration and intensification [4255], and intelligent model studies [5661] as well, for example, methods such as particle swarm optimizer (PSO) [6062], differential search (DS) [63], ant colony optimizer (ACO) [616465], Harris hawks optimizer (HHO) [66], grey wolf optimizer (GWO) [5367], differential evolution (DE) [6869], and other fusion and boosted systems [4146485054557071].

    At the first step, the collector is modified with porous copper foam material. 14 cases have been considered for the analysis of the SCCHP system (Table 6). It should be noted that the adding of porous media causes an additional pressure drop inside the collector [922263072]. All fourteen cases use copper foam with a porosity of 95 percent. To simulate the effect of porous materials and nanofluids, the first solar PTC pipes have been simulated in the FLOW-3D software and then porous media (copper foam with porosity of 95%) and fluid flow with nanoparticles (AL2O3 and CUO) are generated in the software. After analyzing PTC pipes in FLOW-3D software, for analyzing energy and exergy efficiency, software outputs were used as EES software input for optimization ratio of sedimentation and calculating energy and exergy analyses.Table 6 Collectors with different percentages of nanofluids and porous media.

    In this research, an enhanced solar collector with both porous media and Nanofluid is investigated. In the present study, 0.1–0.5% CuO and Al2O3 concentration were added to the collector fully filled by porous media to achieve maximum energy and exergy efficiencies of solar CCHP systems. All steps of the investigation are shown in Table 6.

    Energy and exergy analyses of parabolic solar collectors and SCCHP systems are shown in Figures 6 and 7.

    Figure 6 Energy and exergy efficiencies of the PTC with porous media and nanofluid.

    Figure 7 Energy and exergy efficiency of the SCCHP.

    Results show that the highest energy and exergy efficiencies are 74.19% and 32.6%, respectively, that is achieved in Step 12 (parabolic collectors with filled porous media and 0.5% Al2O3). In the second step, the maximum energy efficiency of SCCHP systems with fourteen steps of simulation are shown in Figure 7.

    In the second step, where 0.1, −0.6% of the nanofluids were added, it is found that 0.5% leads to the highest energy and exergy efficiency enhancement in solar collectors and SCCHP systems. Using concentrations more than 0.5% leads to sediment in the solar collector’s pipe and a decrease of porosity in the pipe [73]. According to Figure 7, maximum energy and exergy efficiencies of SCCHP are achieved in Step 12. In this step energy efficiency is 54.49% and exergy efficiency is 18.29%. In steps 13 and 14, with increasing concentration of CUO and Al2O3 nanofluid solution in porous materials, decreasing of energy and exergy efficiency of PTC and SCCHP system at the same time happened. This decrease in efficiency is due to the formation of sediment in the porous material. Calculations and simulations have shown that porous materials more than 0.5% nanofluids inside the collector pipe cause sediment and disturb the porosity of porous materials and pressure drop and reduce the coefficient of performance of the cogeneration system. Most experience showed that CUO and AL2O3 nanofluids with less than 0.6% percent solution are used in the investigation on the solar collectors at low temperatures and discharges [74]. One of the important points of this research is that the best ratio of nanofluids in the solar collector with a low temperature is 0.5% (AL2O3 and CUO); with this replacement, the cost of solar collectors and SCCHP cycle is reduced.

    5. Conclusion and Future Directions

    In the present study, ways for increasing the efficiency of solar collectors in order to enhance the efficiency of the SCCHP cycle are examined. The research is aimed at adding both porous materials and nanofluids for estimating the best ratio of nanofluid for enhanced solar collector and protecting sedimentation in porous media. By adding porous materials (copper foam with porosity of 95%) and 0.5% nanofluids together, high efficiency in solar parabolic collectors can be achieved. The novelty in this research is the addition of both nanofluids and porous materials and calculating the best ratio for preventing sedimentation and pressure drop in solar collector’s pipe. In this study, it was observed that, by adding 0.5% of AL2O3 nanofluid in working fluids, the energy efficiency of PTC rises to 74.19% and exergy efficiency is grown up to 32.6%. In SCCHP cycle, energy efficiency is 54.49% and exergy efficiency is 18.29%.

    In this research, parabolic solar collectors fully filled by porous media (copper foam with a porosity of 95) are investigated. In the next step, parabolic solar collectors in the SCCHP cycle were simultaneously filled by porous media and different percentages of Al2O3 and CuO nanofluid. At this step, values of 0.1% to 0.6% of each nanofluid were added to the working fluid, and the efficiency of the energy and exergy of the collectors and the SCCHP cycle were determined. In this case, nanofluid and the porous media were used together in the solar collector and maximum efficiency achieved. 0.5% of both nanofluids were used to achieve the biggest efficiency enhancement.

    In the present study, as expected, the highest efficiency is for the parabolic solar collector fully filled by porous material (copper foam with a porosity of 95%) and 0.5% Al2O3. Results of the present study are as follows:(1)The average enhancement of collectors’ efficiency using porous media and nanofluids is 28%.(2)Solutions with 0.1 to 0.5% of nanofluids (CuO and Al2O3) are used to prevent collectors from sediment occurrence in porous media.(3)Collector of solar cogeneration cycles that is enhanced by both porous media and nanofluid has higher efficiency, and the stability of output temperature is more as well.(4)By using 0.6% of the nanofluids in the enhanced parabolic solar collectors with copper porous materials, sedimentation occurs and makes a high-pressure drop in the solar collector’s pipe which causes decrease in energy efficiency.(5)Average enhancement of SCCHP cycle efficiency is enhanced by both porous media and nanofluid 13%.

    Nomenclature

    :Solar radiation
    a:Heat transfer augmentation coefficient
    A:Solar collector area
    Bf:Basic fluid
    :Specific heat capacity of the nanofluid
    F:Constant of air dilution
    :Thermal conductivity of the nanofluid
    :Thermal conductivity of the basic fluid
    :Viscosity of the nanofluid
    :Viscosity of the basic fluid
    :Collector efficiency
    :Collector energy receives
    :Auxiliary boiler heat
    :Expander energy
    :Gas energy
    :Screw expander work
    :Cooling load, in kilowatts
    :Heating load, in kilowatts
    :Solar radiation energy on collector, in Joule
    :Sanitary hot water load
    Np:Nanoparticle
    :Energy efficiency
    :Heat exchanger efficiency
    :Sun exergy
    :Collector exergy
    :Natural gas exergy
    :Expander exergy
    :Cooling exergy
    :Heating exergy
    :Exergy efficiency
    :Steam mass flow rate
    :Hot water mass flow rate
    :Specific heat capacity of water
    :Power output form by the screw expander
    Tam:Average ambient temperature
    :Density of the mixture.

    Greek symbols

    ρ:Density
    ϕ:Nanoparticles volume fraction
    β:Ratio of the nanolayer thickness.

    Abbreviations

    CCHP:Combined cooling, heating, and power
    EES:Engineering equation solver.

    Data Availability

    For this study, data were generated by CARRIER software for the average electrical, heating, and cooling load of a residential building with 600 m2 in the city of Zahedan, Iran.

    Conflicts of Interest

    The authors declare that they have no conflicts of interest.

    Acknowledgments

    This work was partially supported by the National Natural Science Foundation of China under Contract no. 71761030 and Natural Science Foundation of Inner Mongolia under Contract no. 2019LH07003.

    References

    1. A. Fudholi and K. Sopian, “Review on solar collector for agricultural produce,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 9, no. 1, p. 414, 2018.View at: Publisher Site | Google Scholar
    2. G. Yang and X. Zhai, “Optimization and performance analysis of solar hybrid CCHP systems under different operation strategies,” Applied Thermal Engineering, vol. 133, pp. 327–340, 2018.View at: Publisher Site | Google Scholar
    3. J. Wang, Z. Han, and Z. Guan, “Hybrid solar-assisted combined cooling, heating, and power systems: a review,” Renewable and Sustainable Energy Reviews, vol. 133, p. 110256, 2020.View at: Publisher Site | Google Scholar
    4. Y. Tian and C. Y. Zhao, “A review of solar collectors and thermal energy storage in solar thermal applications,” Applied Energy, vol. 104, pp. 538–553, 2013.View at: Publisher Site | Google Scholar
    5. J. M. Hassan, Q. J. Abdul-Ghafour, and M. F. Mohammed, “CFD simulation of enhancement techniques in flat plate solar water collectors,” Al-Nahrain Journal for Engineering Sciences, vol. 20, no. 3, pp. 751–761, 2017.View at: Google Scholar
    6. M. Jahangiri, O. Nematollahi, A. Haghani, H. A. Raiesi, and A. Alidadi Shamsabadi, “An optimization of energy cost of clean hybrid solar-wind power plants in Iran,” International Journal of Green Energy, vol. 16, no. 15, pp. 1422–1435, 2019.View at: Publisher Site | Google Scholar
    7. I. H. Yılmaz and A. Mwesigye, “Modeling, simulation and performance analysis of parabolic trough solar collectors: a comprehensive review,” Applied Energy, vol. 225, pp. 135–174, 2018.View at: Google Scholar
    8. F. Wang, J. Tan, and Z. Wang, “Heat transfer analysis of porous media receiver with different transport and thermophysical models using mixture as feeding gas,” Energy Conversion and Management, vol. 83, pp. 159–166, 2014.View at: Publisher Site | Google Scholar
    9. H. Zhai, Y. J. Dai, J. Y. Wu, and R. Z. Wang, “Energy and exergy analyses on a novel hybrid solar heating, cooling and power generation system for remote areas,” Applied Energy, vol. 86, no. 9, pp. 1395–1404, 2009.View at: Publisher Site | Google Scholar
    10. M. H. Abbasi, H. Sayyaadi, and M. Tahmasbzadebaie, “A methodology to obtain the foremost type and optimal size of the prime mover of a CCHP system for a large-scale residential application,” Applied Thermal Engineering, vol. 135, pp. 389–405, 2018.View at: Google Scholar
    11. R. Jiang, F. G. F. Qin, X. Yang, S. Huang, and B. Chen, “Performance analysis of a liquid absorption dehumidifier driven by jacket-cooling water of a diesel engine in a CCHP system,” Energy and Buildings, vol. 163, pp. 70–78, 2018.View at: Publisher Site | Google Scholar
    12. F. A. Boyaghchi and M. Chavoshi, “Monthly assessments of exergetic, economic and environmental criteria and optimization of a solar micro-CCHP based on DORC,” Solar Energy, vol. 166, pp. 351–370, 2018.View at: Publisher Site | Google Scholar
    13. F. A. Boyaghchi and M. Chavoshi, “Multi-criteria optimization of a micro solar-geothermal CCHP system applying water/CuO nanofluid based on exergy, exergoeconomic and exergoenvironmental concepts,” Applied Thermal Engineering, vol. 112, pp. 660–675, 2017.View at: Publisher Site | Google Scholar
    14. B. Su, W. Han, Y. Chen, Z. Wang, W. Qu, and H. Jin, “Performance optimization of a solar assisted CCHP based on biogas reforming,” Energy Conversion and Management, vol. 171, pp. 604–617, 2018.View at: Publisher Site | Google Scholar
    15. F. A. Al-Sulaiman, F. Hamdullahpur, and I. Dincer, “Performance assessment of a novel system using parabolic trough solar collectors for combined cooling, heating, and power production,” Renewable Energy, vol. 48, pp. 161–172, 2012.View at: Publisher Site | Google Scholar
    16. J. Wang, Y. Dai, L. Gao, and S. Ma, “A new combined cooling, heating and power system driven by solar energy,” Renewable Energy, vol. 34, no. 12, pp. 2780–2788, 2009.View at: Publisher Site | Google Scholar
    17. Y.-Y. Jing, H. Bai, J.-J. Wang, and L. Liu, “Life cycle assessment of a solar combined cooling heating and power system in different operation strategies,” Applied Energy, vol. 92, pp. 843–853, 2012.View at: Publisher Site | Google Scholar
    18. J.-J. Wang, Y.-Y. Jing, and C.-F. Zhang, “Optimization of capacity and operation for CCHP system by genetic algorithm,” Applied Energy, vol. 87, no. 4, pp. 1325–1335, 2010.View at: Publisher Site | Google Scholar
    19. L. Ali, “LDA–GA–SVM: improved hepatocellular carcinoma prediction through dimensionality reduction and genetically optimized support vector machine,” Neural Computing and Applications, vol. 87, pp. 1–10, 2020.View at: Google Scholar
    20. S. Hirasawa, R. Tsubota, T. Kawanami, and K. Shirai, “Reduction of heat loss from solar thermal collector by diminishing natural convection with high-porosity porous medium,” Solar Energy, vol. 97, pp. 305–313, 2013.View at: Publisher Site | Google Scholar
    21. E. Bellos, C. Tzivanidis, and Z. Said, “A systematic parametric thermal analysis of nanofluid-based parabolic trough solar collectors,” Sustainable Energy Technologies and Assessments, vol. 39, p. 100714, 2020.View at: Publisher Site | Google Scholar
    22. H. J. Jouybari, S. Saedodin, A. Zamzamian, M. E. Nimvari, and S. Wongwises, “Effects of porous material and nanoparticles on the thermal performance of a flat plate solar collector: an experimental study,” Renewable Energy, vol. 114, pp. 1407–1418, 2017.View at: Publisher Site | Google Scholar
    23. J. Subramani, P. K. Nagarajan, S. Wongwises, S. A. El-Agouz, and R. Sathyamurthy, “Experimental study on the thermal performance and heat transfer characteristics of solar parabolic trough collector using Al2O3 nanofluids,” Environmental Progress & Sustainable Energy, vol. 37, no. 3, pp. 1149–1159, 2018.View at: Publisher Site | Google Scholar
    24. E. Shojaeizadeh, F. Veysi, and A. Kamandi, “Exergy efficiency investigation and optimization of an Al2O3-water nanofluid based Flat-plate solar collector,” Energy and Buildings, vol. 101, pp. 12–23, 2015.View at: Publisher Site | Google Scholar
    25. A. K. Tiwari, P. Ghosh, and J. Sarkar, “Solar water heating using nanofluids–a comprehensive overview and environmental impact analysis,” International Journal of Emerging Technology and Advanced Engineering, vol. 3, no. 3, pp. 221–224, 2013.View at: Google Scholar
    26. D. R. Rajendran, E. Ganapathy Sundaram, P. Jawahar, V. Sivakumar, O. Mahian, and E. Bellos, “Review on influencing parameters in the performance of concentrated solar power collector based on materials, heat transfer fluids and design,” Journal of Thermal Analysis and Calorimetry, vol. 140, no. 1, pp. 33–51, 2020.View at: Publisher Site | Google Scholar
    27. M. Feizbahr, C. Kok Keong, F. Rostami, and M. Shahrokhi, “Wave energy dissipation using perforated and non perforated piles,” International Journal of Engineering, vol. 31, no. 2, pp. 212–219, 2018.View at: Google Scholar
    28. K. Khanafer and K. Vafai, “A critical synthesis of thermophysical characteristics of nanofluids,” International Journal of Heat and Mass Transfer, vol. 54, no. 19-20, pp. 4410–4428, 2011.View at: Publisher Site | Google Scholar
    29. K. Farhana, K. Kadirgama, M. M. Rahman et al., “Improvement in the performance of solar collectors with nanofluids – a state-of-the-art review,” Nano-Structures & Nano-Objects, vol. 18, p. 100276, 2019.View at: Publisher Site | Google Scholar
    30. M. Turkyilmazoglu, “Condensation of laminar film over curved vertical walls using single and two-phase nanofluid models,” European Journal of Mechanics-B/Fluids, vol. 65, pp. 184–191, 2017.View at: Publisher Site | Google Scholar
    31. X. Zhang, J. Wang, T. Wang, R. Jiang, J. Xu, and L. Zhao, “Robust feature learning for adversarial defense via hierarchical feature alignment,” Information Sciences, vol. 2020, 2020.View at: Google Scholar
    32. X. Zhang, T. Wang, W. Luo, and P. Huang, “Multi-level fusion and attention-guided CNN for image dehazing,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 1, 2020.View at: Publisher Site | Google Scholar
    33. X. Zhang, M. Fan, D. Wang, P. Zhou, and D. Tao, “Top-k feature selection framework using robust 0-1 integer programming,” IEEE Transactions on Neural Networks and Learning Systems, vol. 1, pp. 1–15, 2020.View at: Publisher Site | Google Scholar
    34. X. Zhang, D. Wang, Z. Zhou, and Y. Ma, “Robust low-rank tensor recovery with rectification and alignment,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 1, pp. 238–255, 2019.View at: Google Scholar
    35. X. Zhang, R. Jiang, T. Wang, and J. Wang, “Recursive neural network for video deblurring,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 1, 2020.View at: Publisher Site | Google Scholar
    36. X. Zhang, T. Wang, J. Wang, G. Tang, and L. Zhao, “Pyramid channel-based feature attention network for image dehazing,” Computer Vision and Image Understanding, vol. 1, 2020.View at: Google Scholar
    37. M. Mirmozaffari, “Machine learning algorithms based on an optimization model,” 2020.View at: Google Scholar
    38. M. Mirmozaffari, M. Yazdani, A. Boskabadi, H. Ahady Dolatsara, K. Kabirifar, and N. Amiri Golilarz, “A novel machine learning approach combined with optimization models for eco-efficiency evaluation,” Applied Sciences, vol. 10, no. 15, p. 5210, 2020.View at: Publisher Site | Google Scholar
    39. M. Vosoogha and A. Addeh, “An intelligent power prediction method for wind energy generation based on optimized fuzzy system,” Computational Research Progress in Applied Science & Engineering (CRPASE), vol. 5, pp. 34–43, 2019.View at: Google Scholar
    40. A. Javadi, N. Mikaeilvand, and H. Hosseinzdeh, “Presenting a new method to solve partial differential equations using a group search optimizer method (GSO),” Computational Research Progress in Applied Science and Engineering, vol. 4, no. 1, pp. 22–26, 2018.View at: Google Scholar
    41. F. J. Golrokh, Gohar Azeem, and A. Hasan, “Eco-efficiency evaluation in cement industries: DEA malmquist productivity index using optimization models,” ENG Transactions, vol. 1, pp. 1–8, 2020.View at: