Electric Arc Furnace Steelmaking with Submerged Mixed Injection 9819946018, 9789819946013

Citation preview

Guangsheng Wei Rong Zhu

Electric Arc Furnace Steelmaking with Submerged Mixed Injection

Electric Arc Furnace Steelmaking with Submerged Mixed Injection

Guangsheng Wei · Rong Zhu

Electric Arc Furnace Steelmaking with Submerged Mixed Injection

Guangsheng Wei School of Metallurgical and Ecological Engineering University of Science and Technology Beijing Beijing, China

Rong Zhu School of Metallurgical and Ecological Engineering University of Science and Technology Beijing Beijing, China

ISBN 978-981-99-4601-3 ISBN 978-981-99-4602-0 (eBook) https://doi.org/10.1007/978-981-99-4602-0 Jointly published with Metallurgical Industry Press The print edition is not for sale in China (Mainland). Customers from China (Mainland) please order the print book from: Metallurgical Industry Press. © Metallurgical Industry Press 2024 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publishers, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publishers nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publishers remain neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Preface

Electric arc furnace (EAF) steelmaking is one of the world’s main steelmaking methods. It uses scrap steel as the main raw material and has the characteristics of short process, energy saving and environmental protection. In China, it is an important driving force for the iron and steel industry to realize green manufacturing. In July 2018, the Ministry of Industry and Information Technology stated that the country will study policies to promote the conversion of “long-process” steelmaking to “short-process” steelmaking. The Iron and Steel Industry Adjustment and Upgrading Plan (2016–2020) also pointed out that, in accordance with the concept of green recycling, emphasis is placed on the development of short-process electric arc furnace steelmaking using scrap steel as raw materials. Therefore, accelerating the technological innovation of EAF steelmaking and improving the level of EAF steelmaking operation will play an important role in promoting the structural adjustment, transformation and upgrading of the iron and steel industry in China. In China, EAF steelmaking not only produces ordinary rods and wires, but also is the main process of high-quality special steel smelting. In recent years, with the development of the national economy, the demand for high-quality special steel has increased, and the requirements for the quality of EAF steel products have become increasingly strict. However, using scrap steel as its main raw material, electric arc furnace (EAF) steelmaking is highly volatile in the melting down phosphorus content. Restricted by the furnace structure of eccentric bottom tapping (EBT) EAF, the metallurgical reaction kinetics is weak, which is bad for dephosphorization. The carbon and oxygen reaction is considerably insufficient due to the lack of carbon source by scrap charging, which leads to the lack of CO bubbles in the molten bath. And intensifying oxygen supplying would result in the serious peroxidation of molten steel. Therefore, the new process of submerged mixed injection was first proposed and applied in EAF steelmaking, which could speed up the smelting rhythm and improve the quality of molten steel by delivering lime or carbon powder into the molten steel directly. The mechanism of submerged O2 –CaO and carbon powder injection was analyzed. The molten slag particle demonstrates excellent dephosphorizing capacity for its P2 O5 content which is more than 10% and up to 29.6%. The convective mass v

vi

Preface

transfer of carbon and convective heat transfer can be strengthened by submerged carbon powder injection, which could accelerate the melting of scrap. The materials and energy change of EAF steelmaking with submerged gas-solid injection were investigated, and with the quantity of carbon powder increasing, the powder consumption decreases and the gas consumption and the off-gas generation increase. The impact characteristics of submerged gas-solid injection were investigated, and a theoretical model was built and modified to depict the axis trajectory of submerged gas-solid jet in the liquid bath. With powder injection, the kinetic energy of gas jet was strengthened and the impact penetration depth was increased. The impact characteristics of oven wall oxygen supply were also investigated, and the k value of oxygen gas jet was first proposed. A calculation model was built and modified to predict the jet impact zone depth and volume. The fluid flow characteristics of EAF molten bath with submerged gas-solid injection were also studied. The submerged injection can improve the fluid flow in the lower part of molten bath and accelerated the material and energy transfer. The oven wall oxygen supply can strengthen the stirring in the upper part of molten bath and improve the reaction kinetics conditions in the molten steel-slag interface area. Finally, the B mode of submerged injector and the F mode of oven wall oxygen lance were confirmed as the best match scheme for EAF steelmaking. Based on the above theoretical analysis and experimental research, the industrial application research of EAF steelmaking with submerged gas-solid injection was carried out. With submerged oxygen injection, the composition and temperature of molten steel became more balanced, the content of FeO and TFe in molten slag decreased by 5.28 and 4.63%, respectively, the consumption of ferrous charges and oxygen decreased by 9.2 kg/t and 5.3 Nm3 /t, respectively, the endpoint phosphorus content decreased by 3.8%, and the final carbon-oxygen equilibrium decreased from 0.00318 to 0.00252. On the basis of the above submerged oxygen injection, the carbon powder injection was adopted. In this case, all the final carbon contents were above 0.10% and the content of FeO in molten slag decreased by 0.9%, which further reduced the oxidation of molten steel and increased the metal yield. Finally, the recent innovations and advances of injection metallurgy in EAF steelmaking have been reviewed. Significant research efforts have been made on CO2 utilization in EAF steelmaking processes as coolant gas, stirring gas, carrier gas as well as reacting medium. Coherent jets with CO2 and O2 mixed injection can effectively limit the evaporation and oxidation of iron and reduce the dust generated by reducing fire-spot temperature. Submerged O2 injection with CO2 can strengthen the molten bath stirring and increase the furnace life. Submerged powder injection with CO2 is helpful for dephosphorization, nitrogen removal and liquid steel cleanliness improvement with lower electricity and lime consumption. Bottom blowing CO2 in EAF steelmaking can facilitate the carbon-oxygen reaction reaching equilibrium, decrease the content of nitrogen in molten steel and improve the quality of molten steel due to its superior molten bath stirring capacity. As a greenhouse gas, CO2 is promising for industrial application in EAF steelmaking. Therefore, a possible future process for cyclic utilization of CO2 in EAF-LF steelmaking process was introduced. This CO2 recycling could be effective in mitigating greenhouse gas evolutions within

Preface

vii

the steelmaking shop and of great advantages to green environmental protection as well. The results of this book will provide basic data support for the industrial application of EAF steelmaking with submerged mixed injection, which will promote the development of EAF steelmaking. Beijing, China

Guangsheng Wei Rong Zhu

Acknowledgements

The publication of this book was financially supported by State Key Laboratory of Advanced Metallurgy, School of Metallurgical and Ecological Engineering, University of Science and Technology Beijing. I would like to express my sincere gratitude to Prof. Runzao Liu, Dr. Kai Dong, Mr. Meichen Wang, Mr. Tianping Tang, Dr. Xuetao Wu, Dr. Ting Cheng, Prof. Jingshe Li, Prof. Shufeng Yang, Prof. Xuefeng She, Prof. Chengbin Shi, Mr. Hong Wang, Mr. Shuigen Song and Mr. Yonggang Liu. This work would never have reached completion without their help. I would like to express my thanks for the support by the National Nature Science Foundation of China (No. 51734003 and No. 52004023).

ix

Contents

1 General Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Electric Arc Furnace (EAF) Steelmaking . . . . . . . . . . . . . . . . . . . . . . . 1.1.1 Metallurgical Function of EAF . . . . . . . . . . . . . . . . . . . . . . . . . 1.1.2 Development of EAF Steelmaking . . . . . . . . . . . . . . . . . . . . . . 1.2 Injection Methods in EAF Steelmaking . . . . . . . . . . . . . . . . . . . . . . . . 1.2.1 Oxygen Supplying Technology . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.2 Oxygen Burner Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.3 Coherent Supersonic Jet Technology . . . . . . . . . . . . . . . . . . . . 1.2.4 Bottom-Blowing Technology . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.5 Slag Foaming with Carbon Powder Injection . . . . . . . . . . . . . 1.3 Technologies of Gas–Solid Injection in Steelmaking Process . . . . . . 1.3.1 Free Gas–Solid Injection Outside of the Molten Bath . . . . . . 1.3.2 Submerged Gas–Solid Injection Within Molten Bath . . . . . . 1.4 New Technology of Submerged Gas–Solid Injection in EAF Steelmaking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.1 Technological Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.2 Key Technical Problems to Be Solved . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Mechanism of EAF Steelmaking with Submerged Mixed Injection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Efficient Dephosphorization with Submerged O2 -CaO Injection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Analysis of Dephosphorization Mechanism . . . . . . . . . . . . . . 2.2.2 Comparison of Dephosphorization Effect . . . . . . . . . . . . . . . . 2.3 Improving Scrap Melting by Carburization with Submerged Carbon Powder Injection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Experimental Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.2 Experimental Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 1 1 1 4 5 7 8 9 10 10 11 12 14 15 16 17 19 19 19 19 21 22 22 24 25

xi

xii

Contents

2.4 Materials and Energy Balance Model with Submerged Carbon Powder Injection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.1 Variation Tendency of the Molten Slag and the Off-Gas Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.2 Variation Tendency of the Power Consumption . . . . . . . . . . . 2.4.3 Effective Utilization and Burning Loss of Carbon Powder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.4 Effect of CO Post-combustion on the Power Consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

35 35 37 38 39 40 41

3 Impact Characteristics of Submerged Gas Injection . . . . . . . . . . . . . . . 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Numerical Model and Water Model Experiment . . . . . . . . . . . . . . . . 3.2.1 Numerical Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.2 Water Model Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.3 Model Validation and Error Analysis . . . . . . . . . . . . . . . . . . . . 3.3 Theoretical Modeling for Submerged Gas Jet . . . . . . . . . . . . . . . . . . . 3.3.1 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.2 Theoretical Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.1 Analysis of the Theoretical Model . . . . . . . . . . . . . . . . . . . . . . 3.4.2 Analysis of the Water Model Experiment . . . . . . . . . . . . . . . . 3.4.3 Analysis of the Numerical (CFD) Simulation . . . . . . . . . . . . . 3.4.4 Erosion of the Refractory Around the Submerged Injector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

43 43 44 44 49 50 50 51 51 54 54 56 59

4 Impact Characteristics of Coherent Supersonic Jet . . . . . . . . . . . . . . . . 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Numerical Modeling and Water Experiment . . . . . . . . . . . . . . . . . . . . 4.2.1 Numerical Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2 Water Model Experiment and CFD Model Validation . . . . . . 4.3 Hybrid Modeling for Jet Impact Characteristics . . . . . . . . . . . . . . . . . 4.3.1 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.2 Theoretical Modeling of the Jet Penetration Depth . . . . . . . . 4.3.3 Theoretical Modeling of the Jet Impact Zone Volume . . . . . 4.3.4 Hybrid Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.1 Analysis of the k Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.2 Modification of hp Calculation . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.3 Validation of V impact-zone Calculation . . . . . . . . . . . . . . . . . . . . 4.4.4 Analysis of the Jet Penetration Depth . . . . . . . . . . . . . . . . . . . 4.4.5 Analysis of the Jet Impact Zone Volume . . . . . . . . . . . . . . . . .

67 67 68 69 73 75 75 76 77 80 80 80 82 85 87 88

61 63 64

Contents

xiii

4.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

88 90

5 Modeling and Arrangement of Submerged Nozzles . . . . . . . . . . . . . . . . 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Water Model Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Numerical Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.1 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.2 Governing Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.3 Simulation Details . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.1 Model Validation and Orthogonal Test Results . . . . . . . . . . . 5.4.2 Analysis of Water Model Experiment . . . . . . . . . . . . . . . . . . . 5.4.3 Analysis of Numerical Simulation . . . . . . . . . . . . . . . . . . . . . . 5.4.4 Industrial Application Research . . . . . . . . . . . . . . . . . . . . . . . . 5.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

93 93 94 97 97 98 99 101 101 103 105 109 112 114

6 Combined Blowing and Industrial Application . . . . . . . . . . . . . . . . . . . . 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Water Model Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.1 Model Instruments and Orthogonal Test Scheme . . . . . . . . . . 6.2.2 Analysis of the Water Model Experiment Results . . . . . . . . . 6.3 Numerical Simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.1 Numerical Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.2 Analysis of Numerical Simulation Results . . . . . . . . . . . . . . . 6.4 Industrial Application Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.1 Industrial Application Scheme . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.2 Analysis of the Industrial Application Results . . . . . . . . . . . . 6.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

115 115 116 116 118 120 120 122 126 126 126 131 132

7 Innovations of Injection Metallurgy in EAF Steelmaking . . . . . . . . . . . 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Coherent Jets with COMI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3 Submerged O2 Injection with CO2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4 Submerged Powder Injection with CO2 . . . . . . . . . . . . . . . . . . . . . . . . 7.5 Bottom Blowing CO2 in EAF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.6 Bottom Blowing CO2 in LF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.7 Cyclic Utilization of CO2 in EAF-LF Steelmaking Process . . . . . . . 7.8 Summary and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

133 133 134 136 137 139 142 144 146 146

Chapter 1

General Introduction

1.1 Electric Arc Furnace (EAF) Steelmaking 1.1.1 Metallurgical Function of EAF EAF steelmaking uses scrap steel as the main raw material, which has the characteristics of short process, low energy consumption, and low carbon emissions [1, 2]. The energy consumption of the short process of EAF is about 200 kgce/t and the CO2 emission is about 600 kg/t, which are less than 1/3 of the long process of converter (> 600 kgce/t, > 2000 kg/t), and the advantages of green environmental protection are significant [3, 4]. The modern EAF steelmaking process mainly includes melting, oxidation, heating, and necessary refining (dephosphorization, desulfurization) and other links. With the increasing improvement of technologies, such as the enhanced input of electric energy and chemical energy, the smelting cycle of EAFs has been greatly shortened, and the production efficiency has been greatly improved [5].

1.1.2 Development of EAF Steelmaking The world’s first experimental EAF was born in 1879; In 1890, the EAF realized industrial application for the first time; In 1909, the United States built the first 15t three-phase EAF, the world’s first EAF with a circular furnace shell; In 1936, Germany successfully manufactured the first furnace cover rotary EAF. Beginning in the 1960s, ultra-high power EAFs began to rise and gradually became the mainstream of EAF steelmaking. At the same time, furnace wall oxygen blowing assisted smelting was gradually adopted, and EAF steelmaking technology entered a period of rapid development. In the 1980s, large-scale ultra-high power DC EAFs began to appear, which further promoted the development of EAF short-process steelmaking. At the same time, with the maturity of ordinary supersonic oxygen lance, cluster oxygen © Metallurgical Industry Press 2024 G. Wei and R. Zhu, Electric Arc Furnace Steelmaking with Submerged Mixed Injection, https://doi.org/10.1007/978-981-99-4602-0_1

1

2

1 General Introduction

lance and oxygen burner technology, EAF smelting technology is also becoming mature and perfect [6, 7]. Since the twenty-first century, the crude steel output of the world’s major steelproducing countries has increased steadily, and the output of EAF steel has also increased simultaneously, and the proportion of EAF steel in some countries has also increased. Table 1.1 shows the output of EAF steel in the world. In 2017, the output of EAF steel accounted for about 28% of the global crude steel output. If China is not included, the proportion of the world’s electric furnace steel is more than 46%. Figure 1.1 shows the comparison of the ratio of EAF steel in various regions of the world. The proportion of EAF steel in the United States is as high as 68.4%, the proportion of EAF steel in the European Union is over 40%, and the proportion of EAF steel in Japan and South Korea is 25–30%. EAF steelmaking using scrap steel as raw material is the development trend of the international steel industry. In China, EAF industry has gone through several stages of development, and has gradually established a modern EAF steelmaking industry and EAF manufacturing industry. Table 1.2 shows the statistics of EAF steel output in China. In 2003, the total output of EAF steel reached 39.06 million tons, but the trend of the proportion of EAF steel in China’s total crude steel output has decreased since 2003. This is due to the rapid development of the real estate industry, the substantial increase in Table 1.1 World electric furnace steel production and proportion Years

Total steel output, 10,000 tons

Total output of electric furnace steel, 10,000 tons

EAF steel ratio (%)

2000

84,131

28,352

33.70

2001

84,296

29,588

35.10

2002

89,791

30,439

33.90

2003

96,124

32,682

34.00

2004

104,858

34,813

33.20

2005

112,634

35,818

31.80

2006

124,054

38,742

31.23

2007

134,300

40,900

30.45

2008

133,970

40,994

30.60

2009

123,480

34,450

27.90

2010

143,280

42,080

29.37

2011

153,720

45,280

29.46

2012

155,950

44,660

28.64

2013

164,930

45,240

27.43

2014

166,150

48,700

29.30

2015

162,100

40,680

25.1

2016

162,830

41,190

25.3

2017

168,820

47,270

28.0

1.1 Electric Arc Furnace (EAF) Steelmaking

3

Fig. 1.1 The proportion of EAF steel in various regions of the world in 2017 [8]

demand for construction steel, the relative shortage of scrap steel compared with foreign countries, the tight domestic power supply, and the combined constraints of factors such as the cost of steel per ton. However, in general, EAF steel output is increasing every year, showing an upward trend. In 2007, EAF steel output reached 58.43 million tons, surpassing the United States to become the world’s largest EAF steel producer, and its total amount exceeds India and Germany. The total crude steel output of the three South Korean countries. At present, 92% of the iron required for steel production in China comes from ore, and only 8% comes from scrap steel. This is mainly due to insufficient storage of scrap steel resources and imperfect recycling system. According to the statistics of the China Iron and Steel Association, the actual consumption of steel in China from 1949 to 2010 was about 4.96 billion tons. After deducting the consumption of scrap steel in steelmaking and foundry industries, the steel savings amounted to about 4.568 billion tons. During the “13th Five-Year Plan” period, China’s steel savings volume will reach 11 billion tons. With the rapid growth of steel reserves, China’s social scrap will be released substantially. According to the China Report Hall, China’s scrap steel resources generated approximately 67.9 million tons in 2008, approximately 80.4 million tons in 2010, and exceeded 130 million tons in 2013, which would rank first in the world. Figure 1.2 shows the statistical forecast of China’s social steel storage and production. It is estimated that the self-produced scrap will reach 200 million tons in 2020 and exceed 250 million tons in 2025. With the improvement of the scrap steel recycling system and the scrap steel processing, recycling and distribution industry chain that meets the needs of the steel industry, and the gradual release of domestic scrap steel resources in the future, it will bring major development opportunities for EAF short-process steelmaking with scrap steel as the main raw material and broad market prospects.

4

1 General Introduction

Table 1.2 EAF steel output and proportion in China Years

Total steel output, 10,000 tons

Total output of electric furnace steel, 10,000 tons

EAF steel ratio (%)

2000

12,849

2020

15.72

2001

15,163

2401

15.83

2002

18,225

3040

16.68

2003

22,234

3906

17.57

2004

27,471

4167

15.17

2005

35,579

4179

11.75

2006

42,102

4420

10.50

2007

48,997

5843

11.93

2008

51,200

6333

12.37

2009

56,800

5576

9.82

2010

63,870

6630

10.68

2011

69,400

6800

9.80

2012

71,654

6480

9.04

2013

77,904

7200

9.24

2014

82,270

7100

8.63

2015

80,383

4903

6.10

2016

80,840

4204

5.20

2017

83,170

7485

9.00

Fig. 1.2 China’s social steel storage and production statistics forecast in 2015–2030 [9]

1.2 Injection Methods in EAF Steelmaking In the process of EAF steelmaking, high-efficiency input of chemical energy (oxygen, fuel, etc.) into the EAF bath directly affects the quality, energy consumption and production operation rate, which is the key to the EAF steelmaking. Therefore, various forms and functions of injection methods have been developed and applied.

1.2 Injection Methods in EAF Steelmaking

5

1.2.1 Oxygen Supplying Technology 1.2.1.1

Furnace Door Oxygen Supply Technology

In order to accelerate the melting of scrap steel, the traditional EAF operation adopts the manual oxygen blowing method from the furnace door, that is, the operator holds the oxygen blowing tube to cut the scrap steel from the furnace door, or inserts the oxygen blowing tube into the molten bath to accelerate the melting of the scrap and the decarburization. With the gradual increase in oxygen consumption, the manual oxygen blowing method cannot meet the production needs; and the labor conditions of manual oxygen blowing are poor, unsafe, and unstable. Modern EAF steelmaking has developed an EAF door oxygen lance, which can be used in the control room just as shown in Fig. 1.3. Due to the need for foaming slag, while blowing oxygen into the furnace, carbon powder must be injected into the furnace, and the furnace door carbon oxygen lance is developed accordingly. EAF door oxygen blowing equipment is divided into two categories according to water cooling methods [10], one is water-cooled furnace door carbon oxygen lance, the other is consumable furnace door carbon oxygen lance. When the water-cooled furnace door carbon oxygen lance works in the furnace, the horizontal and vertical angles can be adjusted to flexibly realize the functions of fluxing scrap and foaming slag. However, due to the injector adopts the water-cooling method, the water-cooled furnace door carbon–oxygen injector cannot be inserted into the molten bath when it is extended into the furnace to avoid explosion, and it cannot contact with the scrap steel in the furnace, otherwise it will affect the life of the injector. The consumable furnace door carbon oxygen lance is directly inserted into the molten bath with three outer coating CAIN steel pipes driven by a manipulator. It can also be used for cutting scrap steel and boosting. The injector is consumed while working and has a large range of activities in the furnace. Fig. 1.3 Furnace door carbon oxygen lance

6

1 General Introduction

The water-cooled furnace door carbon oxygen lance has the advantages of high oxygen utilization rate, good foaming slag effect, stable decarburization and dephosphorization effects, and high degree of automation. However, it is mainly used in conjunction with an oxygen burner when the temperature of the steel material is low. It cannot be used for continuous oxygen blowing and the oxygen blowing depth is difficult to control. It cannot be in contact with molten steel during operation, which has certain limitations. The consumable furnace door carbon oxygen lance can start cutting scrap earlier in the furnace, and there is a large space for movement in the furnace, and there is no need to worry about water leakage accidents from the watercooled furnace door carbon oxygen lance, but the operation process requires labor at intervals Connect the oxygen blowing tube.

1.2.1.2

Furnace Wall Oxygen Supply Technology

The purpose of supplying oxygen to the furnace wall of the EAF is to eliminate the cold zone in the furnace and to ensure the balanced melting of the charge. The modularized control of the furnace wall is used to inject pure oxygen to increase the specific power input of the EAF steelmaking and improve production efficiency [11]. Figure 1.4 shows the oxygen supply module of the furnace wall. The furnace wall oxygen lance mainly has the functions of decarburization, fluxing, secondary combustion and foaming slag. Compared with the traditional installation method, the installation method of the furnace wall oxygen lance is closer to the molten bath, and the distance from the jet to the molten bath is shortened by 40– 50% compared with the traditional installation method, which can greatly improve Fig. 1.4 Oxygen supply module for furnace wall

1.2 Injection Methods in EAF Steelmaking

7

the decarburization of the molten bath. Multi-point injection can be realized in the furnace, and the amount of oxygen and carbon powder can be accurately controlled.

1.2.1.3

EBT Oxygen Supply Technology

In order to achieve slag-free tapping, modern EAFs use Eccentric Bottom Tapping (EBT) technology, which not only reduces the amount of slag in the tapping process, but also shortens the smelting cycle and reduces the tapping temperature drop. But at the same time, the EBT zone has become one of the cold zones in the EAF, causing problems such as a slower melting rate of scrap steel in this zone and a large difference between the composition of the molten bath. Installing an EBT oxygen lance can solve the EBT cold zone problem, as shown in Fig. 1.5. The EBT oxygen lance can promote the melting of scrap steel in the EBT area, and completely solve the problems of the scrap steel in the EBT area that has not melted during tapping and the tap hole cannot be opened. After the molten bath appears, it will increase the molten bath in the EBT area. Meanwhile, the temperature and the composition of molten bath will be more uniform.

1.2.2 Oxygen Burner Technology EAF steelmaking has generally adopted oxygen burner technology to ensure the simultaneous melting of the charge [12]. At the same time, the oxygen burner can also strengthen the secondary combustion of CO, effectively shorten the smelting time and improve the production efficiency of the EAF. At present, depending on the fuel used, oxygen burners mainly include oil-oxygen burners, coal-oxygen burners, Fig. 1.5 Installation of EBT oxygen lance

8

1 General Introduction

gas burners and other forms. The fuels used include diesel, heavy oil, pulverized coal and natural gas. The oxygen burner is mainly used in the melting period, because the heat generated can be mainly transferred to the scrap through radiation and forced convection. Both of these heat transfer methods mainly depend on the temperature difference between the scrap and the flame and the surface area of the scrap. Therefore, the efficiency of the burner is highest at the beginning of the melting of each basket of scrap, when the flame is surrounded by relatively cold scrap. As the temperature of scrap steel increases and the surface of scrap steel shrinks, the efficiency of the burner continues to decrease. In order to achieve reasonable efficiency, the burner should be stopped after about 50% of the melting period is completed. After that, due to the low efficiency, even if the oxygen burner continues to be used, it will not achieve the effect of boosting and saving electricity, but only increase oxygen and fuel consumption.

1.2.3 Coherent Supersonic Jet Technology The jet distance of the traditional supersonic oxygen lance is short and relatively dispersed, and the oxygen jet has relatively small impact on the molten bath, which is easy to cause splashing, and the effective utilization rate of oxygen in the furnace is low. Coherent supersonic jet technology is a new type of oxygen injection technology, which can solve the above-mentioned defects of traditional supersonic oxygen lance [13]. As shown in Fig. 1.6, the coherent supersonic jet sets an annular protective airflow (produced by the combustion of gas and oxygen) around the main oxygen jet, which extends the length of the supersonic core section of the main oxygen jet. The oxygen jet core section increases the range by 40–80% compared with the traditional supersonic method, which can better inject oxygen to the depth of the molten bath. The oxygen jet injected into the molten pool is finally dispersed into bubbles, which significantly improves the kinetic conditions of the steelmaking reaction and improves the oxygen utilization rate. After the EAF steelmaking adopts the technology, the smelting cycle is shortened by more than 10 min, and the electricity consumption per is reduced by more than 50 kWh/t. To optimize the process parameters, lots of scholars have carried out serizes of research work, including the effects of different ambient temperature, different fuel gas types and flow rates, different oxygen temperatures and flow rates and different lance structures on the jet characteristics. However, in terms of the characteristics of the jet impingement and stirring bath, although domestic and foreign scholars have done a lot of research on the ordinary supersonic jet [14], there is few report on the characteristics of the coherent supersonic jet impingement and stirring bath.

1.2 Injection Methods in EAF Steelmaking

9

Fig. 1.6 Coherent supersonic jet technology for EAF steelmaking

1.2.4 Bottom-Blowing Technology The bottom-blowing technology of EAF steelmaking is shown in Fig. 1.7. Through the gas supply element (nozzle or breathable brick) at the bottom of EAF, Ar, N2 , CO2 and other media are injected into the molten bath to strengthen the molten bath stirring, which would improve the metallurgical reaction dynamics conditions, accelerate the mass and heat transfer speed, improve the uniformity of the temperature and composition, shorten the smelting time, and accelerate the production rhythm [15]. At present, the representative bottom-blowing technologies used in EAF steelmaking at home and abroad include the DPP bottom blowing technology developed by RHI AG and the safe and long-life bottom blowing technology developed by the University of Science and Technology Beijing (USTB). Among them, the safe and long-life bottom blowing technology developed by USTB has systematically studied the fluid flow characteristics of the EAF molten bath under different furnace types and different bottom blowing arrangements, and determined the best bottom blowing arrangement. And this technology has been applied in the EAFs of different capacities and furnace types in domestic EAF steelmaking enterprises, and the bottom blowing life is more than 1000 heats. Fig. 1.7 Bottom blowing in EAF

10

1 General Introduction

1.2.5 Slag Foaming with Carbon Powder Injection In order to shorten smelting time and increase productivity, modern EAF steelmaking uses higher secondary voltage and power factor for long arc smelting operations, increasing the input of active power and increasing the melting rate of the charge. However, the strong heat flow of the arc radiates to the furnace wall, which increases the heat load on the furnace wall and increases the melting loss and heat loss of refractory materials. In order to make the heat of the arc enter the molten steel as much as possible, modern EAF steelmaking widely uses carbon powder injection foaming slag technology [16]. In EAF steelmaking, carbon powder or silicon carbide powder is injected into the molten bath with oxygen injection, which would form a strong carbon–oxygen reaction, and a large amount of CO gas is formed in the slag layer. Usually, the foaming slag makes the thickness of the slag reach 2.5–3.0 times of the length of the arc, which can completely shield the arc, reduce the radiation of the arc to the furnace roof and furnace wall, extend the life of the arc furnace body, and improve the thermal efficiency. The foaming slag technology is suitable for large-capacity and ultra-high-power EAFs, and the effect is more prominent in DC EAFs with longer arcs. The foaming slag can increase the heat transfer efficiency of the arc to the molten bath from 30 to 60%, shorten the smelting cycle by 10–14%, reduce the smelting power consumption by about 22% and reduce the electrode consumption by about 2 kg/t. However, in the process of carbon powder injection in the traditional EAF, the carbon powder airflow is affected by the limitation of the injection speed and the disturbance of the high temperature flue gas flow in the furnace. The speed of carbon powder injection decays quickly, and the jet impact force is insufficient.

1.3 Technologies of Gas–Solid Injection in Steelmaking Process The steelmaking process, including Converters, EAFs, refining furnaces and other steelmaking furnaces, generally adopts gas–solid injection technology for smelting, which can inject powder required into the steelmaking bath for smelting using Ar or other carrier gas by the gas–solid injection system. According to the different injection methods, the gas–solid injection smelting technology in the steelmaking process can be basically divided into two types: Free gas–solid injection outside of the molten bath and submerged gas–solid injection within molten bath.

1.3 Technologies of Gas–Solid Injection in Steelmaking Process

11

1.3.1 Free Gas–Solid Injection Outside of the Molten Bath As shown in Fig. 1.8, free gas–solid injection outside of the molten bath is widely used in the field of EAF steelmaking [17], including EAF wall carbon powder injection, furnace door carbon powder injections and EAF wall lime powder injection, etc. The converter top powder injection smelting technology includes KG-LI technology in which oxygen and lime powder are blown into the furnace through a top-blowing oxygen lance, and stainless steel smelting technology in which chrome ore powder is injected by the top lance for smelting reduction and direct alloying [18]. The KG-LI technology has been industrially tested in the second steelmaking plant of Chiba Steel Works, Japan. Kawasaki Steel is the first steel conglomerate to develop and apply the converter smelting reduction method to prepare stainless steel and has successfully realized the industrialization of this technology. The RH-KTB (RH-Kawasaki Top blowing) process [19] (Fig. 1.9a) uses a top injector to inject powder into the vacuum chamber for desulfurization. In this process, the top injector is not in contact with the molten steel, which is a non-contact powder injection. The injector is not corroded by the molten steel, no refractory material is lost, and the spraying is stable.

Fig. 1.8 Schematic diagrams of gas–solid injection in EAF steelmaking

12

1 General Introduction

Fig. 1.9 RH refining powder spraying process

1.3.2 Submerged Gas–Solid Injection Within Molten Bath The injection metallurgy methods used on the ladle are mainly the French IRSID method, the German TN method, and the Swedish SL method [20]. Among them, the IRSID method was first proposed by the French Iron and Steel Research Institute in 1963. As shown in Fig. 1.10a, the SL method is to inject refined powder or alloy powder into the molten bath by submerged injector, which was developed by Scandinavian Lancers AB, Sweden. In Fig. 1.10b, the V-KIP (Vacuum-Kimizu Inject Process) method [21] is to inject powder continuously into the ladle from the top of the ladle. Because the powder injection is carried out under vacuum conditions, the ability of this method for dehydrogenation, desiliconization, deoxidation, and desulfurization is greatly improved. Figure 1.9b, c are respectively RH-PB method and RH-Injection method. Both methods belong to submerged gas–solid injection within molten bath, and the injector is in direct contact with the molten steel. The RH-Injection method directly inserts the injector into the ladle molten bath; the RH-PB method is equipped with a powder injection device at the bottom of the vacuum chamber to inject the powder directly into the molten steel. As shown in Fig. 1.11, the converter bottom powder injection process mainly includes Q-BOP, K-BOP, KMS and KOBM. For Q-BOP and K-BOP, lime powder was injected through the bottom nozzles by oxygen. And for KMS, carbon powder was directly injected into the molten bath by the bottom nozzles to provide a heat source, thereby increasing the ratio of converter smelting scrap. For KOBM converter smelting in Canada Dofasco Steel Company, all the lime needed is blown through the bottom blowing nozzles in the form of powder. The maximum injection rate of lime powder is 5 kg/(t min) and the bottom blowing oxygen intensity is 1.0 Nm3 / (t min).

1.3 Technologies of Gas–Solid Injection in Steelmaking Process

13

Fig. 1.10 Ladle refining powder spraying process Fig. 1.11 Powder injection at the bottom of converter

As shown in Fig. 1.12, Northeastern University proposed the L-BPI (LadleBottom Powder Injection) technology. This technology replaces argon-blowing venting bricks with powder-spraying elements, and injects powder into the molten steel from the bottom of the ladle, achieving a desulfurization rate of over 85%, and the desulfurization time can be shortened to less than 20 min.

14

1 General Introduction

Fig. 1.12 Powder injection at the bottom of ladle furnace

1.4 New Technology of Submerged Gas–Solid Injection in EAF Steelmaking In recent years, based on the development of multi-functional oxygen supplying technology for EAF steelmaking, the technology of submerged mixed injection have been proposed and developed by the Metallurgical Lance Research Center of USTB to improve the efficiency of oxygen utilization and the metallurgical reaction kinetic conditions [22]. Just as shown in Fig. 1.13, the technology of submerged mixed injection includes submerged oxygen injection, submerged O2 -CaO injection, and submerged carbon powder injection. The process is achieved through the following steps: 1. Directly inject pure O2 or gas–solid powder gas jet into the molten bath through the injection element installed at the lower side of EAF; System for carbon powder injection Electric Arc Furnace

Air

System for lime powder injection O2

Submerged injector

Carrier gas & Carbon powder Carrier gas & Lime powder

Ar / N2

Gas for preventing clogging

Gas for cooling injector CH4 /Ar/ N2

Fig. 1.13 Schematic diagram of submerged mixed injection smelting system in EAF

1.4 New Technology of Submerged Gas–Solid Injection in EAF Steelmaking

15

2. Develop matching injection equipment and control systems, detect and control the medium injection rate accurately, and realize automatic control; 3. Study the jet behavior of submerged mixed injection to optimize the process parameters, further increase the jet impact depth and improve utilization efficiency; 4. Transform the oxygen supply system of EAF steelmaking to meet the new process requirements, improve the molten steel quality, reduce the production energy consumption, and speed up the smelting rhythm.

1.4.1 Technological Advantages The submerged mixed injection technology in the EAF steelmaking is different from traditional EAF smelting methods, which is a major innovation. At the same time, the powder particles can effectively extend the penetration depth of the gas jet, further strengthen the stirring of the molten bath, and accelerate the chemical reaction speed. This process has the following advantages.

1.4.1.1

Strengthen Molten Bath Stirring and Stabilize the Endpoint Control

The EAF steelmaking furnace is large and the molten bath is shallow. It has always been difficult to directly inject a large flow of gas into the EAF molten bath for stirring. The reaction kinetic conditions of the molten bath are poor, and the temperature and composition of the molten steel in various regions are uneven. The existing oxygen supply method of EAF steelmaking decays rapidly, and the utilization efficiency is difficult to predict, which brings great technical difficulties to the process control. The submerged mixed injection can realize the large flow of gas injection within the molten bath and greatly improves the EAF bath stirring. The utilization efficiency of related media is stable, which reduces the composition and temperature fluctuations and operation deviation, laying the foundation for the stable endpoint control of EAF smelting.

1.4.1.2

Advantages of Metallurgical Effect

With submerged oxygen injection, the oxygen utilization efficiency is maintained above 98% [23], which significantly improves the dephosphorization dynamics conditions and promotes the interface reaction of steel-slag. With submerged O2 CaO injection, the powder particles are used to increase the penetration depth of the oxygen jet, while reducing the outlet temperature of the submerged injector to protect the relative components, and the CaO particle can achieve efficient dephosphorization and significantly improve the quality of molten steel. With submerged carbon powder

16

1 General Introduction

injection, the utilization efficiency of carbon powder can be improved significantly and better molten bath stirring can accelerate the uniform dispersion of carbon powder in the molten bath, improve the carburizing efficiency and significantly accelerates the melting speed of scrap.

1.4.1.3

Advantages of Application Prospect

The submerged mixed injection is a new EAF smelting method, which has an important influence on the innovation of the EAF steelmaking process. This technology has different working modes according to the multiple charge structures. (1) For the charge structure with molten iron, submerged O2 injection or submerged O2 -CaO injection is adopted to strengthen dephosphorization and decarbonization and speed up the smelting rhythm. (2) For the charge structure dominated by scrap steel, submerged O2 -CaO injection is adopted in the early stage of smelting to accelerate the melting of scrap steel by increasing the carbon content of the molten bath. Submerged O2 injection or submerged O2 -CaO injection is adopted to strengthen decarburization and improve dephosphorization. When the EAF steelmaking is approaching the end of smelting, the injection system can directly inject a large flow of Ar into the molten bath to deeply remove [N], [H] and inclusions in the molten steel. Therefore, this technology is of great significance in shortening the smelting cycle, saving energy and reducing consumption, improving the quality of molten steel and increasing production efficiency.

1.4.2 Key Technical Problems to Be Solved As a new technology of EAF steelmaking, the submerged mixed injection technology must solve the following key problems: 1. Research on the mechanism of EAF steelmaking with submerged mixed injection, including the dephosphorization mechanism of submerged O2 -CaO injection, the smelting mechanism of submerged carbon powder injection. Study the law of material and energy changes with submerged mixed injection, which would provide a data basis for the optimization of the smelting process. 2. Research on the characteristics of submerged mixed injection, grasp the law of jet impact under different injection conditions, and provide theoretical guidance for the optimization of gas–solid injection parameters in the EAF steelmaking bath. 3. The construction of the EAF steelmaking oxygen injection system combining furnace wall coherent supersonic jet and submerged mixed injection, which could improve the molten bath stirring efficiently.

References

17

4. Improve the injection system equipment design, formulate the relevant smelting processes and carry out the intermediate and industrial-scale trials of submerged mixed injection. This is the technical guarantee for the application and promotion of this new technology. Therefore, understanding the relevant characteristics of submerged mixed injection in EAF steelmaking is of great significance to develop the EAF Smelting system combining submerged mixed injection and coherent supersonic jet.

References 1. Miller FP, Vandome AF, Mcbrewster J (2010) Electric arc furnace. Alphascript Publishing, Germany, pp 312–391 2. Mori R (2006) Electric arc furnace steelmaking method (trans: Zhu G). Metallurgical Industry Press, Beijing, pp 178–189 3. Toulouevski YN, Zinurov IY (2010) Innovation in electric arc furnaces. Springer Berlin Heidelberg, Germany, pp 211–235 4. Fruenhan R (2006) Oxygen versus EAF steelmaking in the 21st century. Trans Indian Inst Met 59(5):607–617 5. Zhu R (2018) Modern electric arc furnace steelmaking oxygen theory and technology. Metallurgical Industry Press, pp 351–372 6. Xu K, Hong X (2005) Review of the short process of electric furnace and some problems in the development. China Metall 15(7):1–8 7. Zhu R, He C, Liu R, Li J (2010) The development of electric arc furnace steelmaking equipment technology. China Metall 20(04):8–16 8. Association WS (2018) Steel statistical yearbook. World Steel Association, pp 23–24 9. Chinese Academy of Engineering Consulting Project (2015) Strategic research on a powerful country in ferrous metal mineral resources. pp 43–91 10. Memoli F, Mapelli C, Ravanelli PE (2004) Simulation of oxygen penetration and decarburisation in EAF using supersonic injection system. ISIJ Int 44(8):1342–1349 11. Memoli F, Mapelli C, Ravanelli P et al (2004) Evaluation of the energy developed by a multipoint sidewall burner-injection system during the refining period in a EAF. ISIJ Int 44(9):1511–1516 12. Yang Y (2017) Modification of full premixed oxygen burner on the wall of electric furnace. Metall Mater 37(05):48–50 13. Mathur P (2004) Coherent jets in steelmaking: principles and learnings. Praxair Met Technol 47(3):184–205 14. Alam M, Irons G, Brooks G et al (2011) Inclined jetting and splashing in electric arc furnace steelmaking. ISIJ Int 51(9):1439–1447 15. Wei G, Zhu R, Dong K et al (2016) Research and analysis on the physical and chemical properties of molten bath with bottom-blowing in EAF steelmaking process. Metall Mater Trans B 47(5):3066–3079 16. Morales RD, Lule GR, Lopez F et al (2007) The slag foaming practice in EAF and its influence on the steelmaking shop productivity. ISIJ Int 35(9):1054–1062 17. Larry W, Philip J (2005) The future of lime for steelmaking. Iron Steel Technol Assoc Iron Steel Technol 48(3):87–92 18. Nozaki N (2008) Bottom blowing converter method (trans: Zhang J, Zhang B). Metallurgical Industry Press, Beijing, pp 44–82 19. Wei J, Yu N, Fan Y et al (2002) Study on flow and mixing characteristics of molten steel in RH and RH-KTB refining processes. J Shanghai Univ (Engl Ed) 6(2):167–175

18

1 General Introduction

20. Liu L (2001) The development of refining technology outside the furnace. Steelmaking 17(4):1– 7 21. Zhang R (1986) V-KIP method ladle refining treatment. Steelmaking 23(5):82–85 22. Wei G, Zhu R et al (2018) Technological innovations of carbon dioxide injection in EAF-LF steelmaking. JOM 70(6):969–976 23. Liao Y, Tao H, Tang X et al (2015) Optimization of smelting process with bottom side blowing and oxygen supply of 50t electric furnace. Jiangxi Metall 35(03):15–16, 25

Chapter 2

Mechanism of EAF Steelmaking with Submerged Mixed Injection

2.1 Introduction Submerged mixed injection is a new EAF smelting method, including submerged gas injection and submerged gas–solid injection, which contains O2 -CaO injection and carbon powder injection. The technology of submerged O2 -CaO injection for dephosphorization is to inject lime powder directly into EAF molten bath by using submerged nozzle. The rapid dephosphorization of EAF steelmaking can be realized by the “gas–solid-slag-metal” multiphase reaction system in EAF molten bath. The technology of submerged carbon powder injection is to inject carbon powder directly into the molten bath by submerged nozzle, which can accelerate the carburization of molten bath, promote the melting of scrap steel and improve the quality of molten steel. This chapter mainly analyzes the dephosphorization mechanism of submerged lime powder injection and the carburization mechanism of carbon powder injection, studies the material and energy variation law of submerged mixed injection in EAF steelmaking, and focuses on the analysis of the variation relationship between gas–solid injection parameters and smelting power consumption, gas consumption, furnace gas composition and volume.

2.2 Efficient Dephosphorization with Submerged O2 -CaO Injection 2.2.1 Analysis of Dephosphorization Mechanism Phosphorus is a harmful element in most steel grades and limits the development of high-quality steel. However, for EAF steelmaking, the phosphorus content of initial molten steel fluctuates greatly due to the complex raw material structures, the carbon © Metallurgical Industry Press 2024 G. Wei and R. Zhu, Electric Arc Furnace Steelmaking with Submerged Mixed Injection, https://doi.org/10.1007/978-981-99-4602-0_2

19

20

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

content of molten steel is low, the viscosity of molten steel is high and the flow speed of molten bath is slow due to the limitation of EAF structure, which leads to great difficulty in dephosphorization. The new technology of submerged O2 -CaO injection provides a new high efficient dephosphorization method for EAF steelmaking. During the EAF steelmaking process with submerged O2 -CaO injection, the dephosphorization reaction in Eq. (2.1) occurring during lime powder floatation is a “transient reaction” [1, 2]. As shown in Fig. 2.1, the submerged injection oxygen reacts with the liquid metal to form a fire point zone, and the lime powder blown with oxygen reacts with the large amount of FeO generated near the fire point zone to form Ca-ferrate particles. The particles oxidize phosphorus during the floating process to become slag particles composed of xCaO·P2 O5 , but the large amount of SiO2 that forms during the desilication period may cause a decrease in (%P2 O5 ) of the slag particles. At the same time, xCaO·P2 O5 aggregates with the suspended lowbasicity slag particles formed during desilication and entrapment of the top slag and continues to float and separate from the molten steel for further dephosphorization. On the other hand, the top slag with a high basicity also has a certain dephosphorization ability, which can be combined with the submerged O2 -CaO injection process to remove phosphorus form the molten steel. 2[P] + 5(FeO) + x(CaO) = (xCaO · P2 O5 ) + 5[Fe]

Fig. 2.1 Dephosphorization mechanism of submerged O2 -CaO injection

(2.1)

2.2 Efficient Dephosphorization with Submerged O2 -CaO Injection

21

2.2.2 Comparison of Dephosphorization Effect Tables 2.1 and 2.2 list the composition of molten slag particles with submerged O2 -CaO injection and the composition of molten slag with the traditional EAF steelmaking process. As shown in Table 2.1, the content of P2 O5 in traditional EAF slag is only 2–4%, while the content of P2 O5 in molten slag with submerged O2 -CaO injection is more than 10% and the highest is 29.6%. The dephosphorization effect of molten slag with submerged O2 -CaO injection is better than that of traditional EAF slag. The molten slag produced in the process of submerged O2 -CaO injection has strong dephosphorization ability, which provides conditions for rapid and efficient dephosphorization in EAF steelmaking. As far as the data of T.Fe in slag is concerned, the content of T.Fe in slag of traditional EAF steelmaking is basically distributed in 20–30%, or even higher. The traditional EAF steelmaking uses repeated slag-flowing and slag-making for dephosphorization operation, the content of T.Fe in slag is high, which would result in high consumption of iron and steel materials and increased production cost. However, for EAF steelmaking with submerged O2 -CaO injection, the content of T.Fe in the molten slag is only about 1.0%. The deep dephosphorization is realized and the content of T.Fe in the slag is very low, which can significantly reduce the iron loss, further reduce the consumption of iron and steel materials and production cost, and provide a premise for the realization of low-cost deep dephosphorization in EAF steelmaking. On the other hand, speeding up the fluid flow in EAF bath can strengthen the material and energy transfer between slag and steel, and improve the metallurgical reactions such as dephosphorization. However, in order to prevent liquid steel from peroxidation, the oxygen supply intensity of EAF wall is limited and due to the Table 2.1 Composition of molten slag particles with submerged O2 -CaO injection/% [3] Label

CaO

P2 O5

SiO2

T.Fe

MgO

Slag particle 1

48.4

29.6

10.9

1.0

2.8

Slag particle 2

58.6

19.3

13.2

0.9

0.8

Slag particle 3

60.4

17.3

14.8

0.9

0.4

Slag particle 4

66.3

10.4

14.7

0.9

0.3

Table 2.2 Composition of molten slag with traditional EAF steelmaking/% Label

CaO

P2 O5

SiO2

T.Fe

MgO

Slag 1

38.8

2.8

11.4

28.2

3.9

Slag 2

41.6

3.7

12.9

26.7

5.4

Slag 3

37.5

3.3

10.9

30.1

3.8

Slag 4

44.2

2.3

11.5

25.7

5.3

22

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

special flat shallow bath structure, the bottom blowing rate of EAF is also limited (the bottom blowing rate of EAF is generally controlled below 100 NL/min). Therefore, it is difficult to improve the stirring intensity of EAF steelmaking bath. The technology of submerged O2 -CaO injection can realize large flow gas injection in EAF molten bath (the flow rate of injection gas is generally 300–600 Nm3 /h). At the same time, good bath stirring can enhance the uniform dispersion of lime powder in molten steel and further improve the utilization efficiency of lime powder.

2.3 Improving Scrap Melting by Carburization with Submerged Carbon Powder Injection In order to study the mechanism for improving scrap melting by molten bath carburization with the submerged carbon powder injection technology, the induction furnace experiments were carried out to study the characteristics of a scrap steel bar with different carbon contents and different bottom-blowing gas flow rates [4]. The melting process of the steel bar was analyzed and the relative rate-determining step was clarified. Meanwhile, the effect of the molten bath carbon content and the molten bath fluid flow on the steel scrap melting was studied. Finally, the convective mass transfer coefficient and the convective heat transfer coefficient between the steel bar and the molten metal were also obtained on the basis of the experimental data.

2.3.1 Experimental Installation Figure 2.2 shows the experimental platform and it consists of a medium frequency induction furnace, a buffer vessel with N2 cylinders, a fixing device, a Pt-Rh-Pt thermocouple, an infrared radiation thermometer and a computer. The capacity of the induction furnace is 150 kg and a bottom blowing element was installed at the bottom of the medium frequency induction furnace, which could provide two bottom blowing nozzles with the diameter being 1.0 mm. The bottom-blowing medium was N2 and its flow rate can be controlled by the gas flow meter. The fixing device was employed to clamp and fix the steel bar, which was used to analyze the melting characteristics of steel scrap in molten metal and Fig. 2.3 shows the steel bar used in this study. The Pt-Rh-Pt thermocouple was employed to monitor the center temperature of the steel bar during its melting process. The infrared radiation thermometer, FLUKE EF1RH, was employed to measure the temperature of molten metal online and its measurement temperature range is 1000–3200 °C. In this study, the temperature of the molten metal was controlled by adjusting the input power of the medium frequency induction furnace.

2.3 Improving Scrap Melting by Carburization with Submerged Carbon … Pt-Rh-Pt Thermocouple

Signal converter

Fixing device

Signal converter

Experimental steel bar

23

Thermodetector

Gas flow meter Buffer vessel

Computer Valves

N2

N2

N2

Bottom blowing nozzles

Induction coil

Medium frequency induction furnace

Fig. 2.2 Schematic diagram of experimental platform with induction furnace

Fig. 2.3 The steel bar used

Figure 2.4 demonstrates the melting process of steel bar during the induction furnace experiment and the experimental operation is as follows: Fig. 2.4 The melting process of steel bar during the induction furnace experiment

24

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

1. According to the experimental schemes, the metal charge was prepared well and put into the induction furnace. Then, turn on electricity and the melting process started. 2. When the temperature of the molten bath in the induction furnace reached the required temperature, adjust the input power of the induction furnace to maintain the stable temperature of the molten bath. 3. Insert the steel bar into the molten bath and after the designed soaking time, the steel bar was taken out for quench cooling. 4. Then the weight and the average diameter of the steel bar was measured. Five diameters were measured along the steel bar’s length and the average value of these five diameters was the average diameter of the steel bar after melting. 5. The steel bar’s center temperature was measured by a Pt-Rh-Pt thermocouple and the temperature data was recorded by the computer.

2.3.2 Experimental Schemes In the present study, two kinds of steel bar (45 carbon steel) were used to analyze the effect of the carbon content and fluid flow of molten bath on the melting of steel scrap and their diameters are, respectively, 20 and 50 mm with the same length being 150 mm. Table 2.3 shows the composition of the steel bar used in the experiment. Label Φ20 and Label Φ50 represent the steel bar with the diameter being 20 mm and 50 mm, respectively. The steel bar with the diameter being 20 mm was employed to investigate the weight and diameter change of the steel bar during the melting process. In this study, a Pt-Rh-Pt thermocouple was inserted into the thermometer hole of the steel bar. The external diameter of the thermocouple is 8 mm and as a result, the steel bar with the diameter being 50 mm was used to monitor the steel bar center temperature change during its melting process. The thermometer hole was drilled at the center of the steel bar and its inner diameter and depth were 10 mm and 80 mm, respectively. Three kinds of molten bath with different carbon content were formed to investigate the effect of the carbon content on the melting of steel scrap. Round bar steel Q235 and pig iron were applied to form the molten bath with the designed carbon content. The molten metal was sampled to detect the chemical composition after the formation of the molten bath. Table 2.4 lists the key parameters of molten bath formed in the experiment. Label [% 4.03], Label [% 3.01] and Label [% 2.10] represent the molten bath with the carbon content being 4.030%, 3.010% and 2.100%. Therefore, in this paper, Label [% 4.03], Label [% 3.01] and Label [% 2.10] can also Table 2.3 The composition of the steel bar used in the experiment Label

Diameter (mm)

w(C) (%)

w(Si) (%)

w(Mn) (%)

w(P) (%)

w(S) (%)

Φ20

20

0.450

0.210

0.550

0.030

0.031

Φ50

50

0.460

0.240

0.540

0.035

0.032

2.3 Improving Scrap Melting by Carburization with Submerged Carbon …

25

Table 2.4 The composition of molten bath with different carbon content Item

Label

w(C) (%)

w(Si) (%)

w(Mn) (%)

w(P) (%)

w(S) (%)

Temperature (°C)

1

[% 4.03]

4.030

0.640

0.310

0.036

0.015

1400

2

[% 3.06]

3.010

0.620

0.320

0.037

0.016

1400

3

[% 2.10]

2.100

0.630

0.330

0.037

0.015

1400

Table 2.5 The composition of molten bath with different bottom-blowing gas flow rate Item Gas flow rate, w(C) (%) w(Si) (%) w(Mn) (%) w(P) (%) w(S) (%) Temperature L/min (°C) 1

3.0

3.010

0.620

0.320

0.037

0.016

1400

2

5.0

3.030

0.580

0.360

0.033

0.014

1400

3

7.0

2.980

0.610

0.350

0.034

0.017

1400

be defined as high carbon molten bath, medium carbon molten bath and low carbon molten bath, respectively. To investigate the effect of the molten bath fluid flow on the melting of steel scrap, three kinds of bottom-blowing modes with different gas flow rates were adopted and the bottom-blowing medium is N2 . Due to the stirring effect of bottom blowing gas on the molten bath, the larger the bottom-blowing gas flow rate, the faster the molten bath fluid flows. Similarly, the molten metal was also sampled to detect the chemical composition after the formation of the molten bath. Table 2.5 lists the key parameters of molten bath with different bottom-blowing gas flow rates in the experiment. It can be seen that the carbon content of these three cases was around 3.0%. That is, the experiments to study the effect of the molten bath fluid flow on the melting of steel scrap were carried out with the medium carbon molten bath.

2.3.3 Results and Discussion 2.3.3.1

Analysis of the Melting Process of Steel Scrap

Figure 2.5 shows the shape change of the steel bar during the induction furnace experiment. Normally, the melting process of steel scrap consists of three stages and they are, respectively, the iron coagulating stage, the carburizing and melting stage and the full boiling stage. The iron coagulating stage occurs at the initial stage after the steel bar’s immersion into the molten bath. In this stage, due to the large difference between the steel bar and the molten bath, some molten metal freezes on the surface of the steel bar and therefore, the iron coagulating layer occurs. With the heating time increasing, the temperature of the steel bar increases and as a result, the iron coagulating layer melts gradually. Finally, the surface of the steel bar returns to

26

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

Fig. 2.5 Shape change of the steel bar during the induction furnace experiment

the original shape. The time of this stage is real short and it is about 20 s and 40 s for the steel bar with its diameter being 20 mm and 50 mm, respectively. The steel bar is heated continuously and the iron coagulating stage is followed by the carburizing and melting stage. Figure 2.6 depicts the mechanism of the steel bar in the molten bath during the carburizing and melting stage. In this stage, carbon in the molten metal carburizes to the surface of steel bar and hence, the carburized layer can be formed. According to the relative data in the iron-carbon phase diagram, a 1.0% increase in the carbon content of iron-carbon alloy would decrease the melting point of iron-carbon alloy by 78.5 °C [5, 6]. Therefore, when the melting point of the carburized layer is lower than the molten bath temperature, the carburized layer would be melted by the surrounding flowing molten metal. After that, the new surface of the steel bar appears and then, the carburizing and melting process of the steel bar surface repeats successively. With the temperature of the molten bath increasing and the diameter of steel bar decreasing, the temperature of the molten bath is above the melting point of the steel bar and as a result, the steel bar would directly melt without carburization. This stage is defined as the full boiling stage. In this stage, the melting rate of the steel bar is about 18.0 g/s, which is faster than the previous two stages. The melting speed of the steel bar is controlled by the carburizing and melting stage and the rate-determining step of this stage is the carburization process. Figure 2.7 shows the microstructure of the steel bar after soaking for 40 s. It can be found that the metallurgical structure is cast iron, acicular martensite and the original structure of steel bar in turn from outside to inside. The cast iron structure layer is formed by the molten metal adhered on the surface of the steel bar when the steel bar is extracted from the molten bath and the acicular martensite layer is formed by the

2.3 Improving Scrap Melting by Carburization with Submerged Carbon … Fig. 2.6 Diagram of the steel bar melting mechanism in the molten metal during the carburizing and melting stage

27

Steel bar Carburized layer Molten Metal Surface carburization Melting point depression

New interface appearance Carburized layer melting

Fig. 2.7 The microstructure of the steel bar in the molten bath with the carbon content being 4.030%

carburization effect. As reported [7, 8], the carburization process is endothermic and higher temperature is in favor of the carburizing reaction and can accelerate the melting rate of the steel bar. Therefore, as key parameters of the steel scrap melting, the convective mass transfer coefficient of carbon and the convective heat transfer coefficient between the steel bar and the molten metal were studied and analyzed.

2.3.3.2

Effect of the Molten Bath Carbon Content on the Melting of Steel Bar

The steel bars with the diameter being 20 mm were respectively inserted into the high carbon molten bath, the medium carbon molten bath and the low carbon molten bath. The weight and the diameter of the steel bar during the melting process were recorded.

28

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

Figure 2.8 shows the weight change of the steel bar in the three molten bath with different carbon content. In each case, it can be found that with the soaking time increasing, the weight of the steel bar decreases and the attenuation rate of the steel bar weight is faster and faster. For instance, in the molten bath with the carbon content being 4.030%, the slopes of the linear regression are 1.19, 2.41, 2.98, 3.23 and 3.36 when the time is in the ranges of 0–30, 30–40, 40–55, 55–65 and 65–70 s. Figure 2.9 shows the diameter change of the steel bar in the molten bath with different carbon content. Similar with the variation tendency of the steel bar weight, the diameter of the steel bar decreases with the soaking time increasing and the attenuation rate of the steel bar diameter is faster and faster. For instance, in the molten bath with the carbon content being 4.030%, the slopes of the linear regression are 0.046, 0.099, 0.133, 0.162 and 0.184 when the time is in the ranges of 0–30, 30–40, 40–55, 55–65 and 65–70 s. There are two aspects to account for this phenomenon. Firstly, with the soaking time increasing, the temperature of the steel bar increases and the carburizing reaction is accelerated. Secondly, the ratio of surface area to volume becomes larger with the diameter decreasing during the melting process and as a result, the carbon concentration in unit volume would become larger with the same carburizing quality. Therefore, the melting speed of the steel bar is faster and faster. From Fig. 2.8, it can be found that the weight of the steel bar is lower with larger carbon content. When the time is in the range of 0–30 s, the mass melting rate is 0.79, 0.96 and 1.19 g/s with the molten bath carbon content being 2.100, 3.010 and 4.030% and when the time is in the range of 40–55 s, the mass melting rate is 2.14, 2.66 and 2.98 g/s with the molten bath carbon content being 2.100, 3.010 and 4.030%. And the same regularity also occurs in the other time ranges. Meanwhile, from Fig. 2.9, it also can be found that the diameter of the steel bar is lower with larger carbon content. When the time is in the range of 0–30 s, the diameter reduction rate is 0.030, 0.037 and 0.046 mm/s with the molten bath carbon content being 2.100, 3.010 and 4.030% and when the time is in the range of 40–55 s, the diameter reduction rate is 0.091, 0.114 and 0.133 mm/s with the molten bath carbon content being 2.100, 3.010 and 4.030%. And the same regularity also occurs in the other time ranges. Fig. 2.8 Weight change of the steel bar in molten bath with different carbon content

2.3 Improving Scrap Melting by Carburization with Submerged Carbon …

29

Fig. 2.9 Diameter change of the steel bar in molten bath with different carbon content

That means the molten bath carbon content has a significant impact on the steel bar melting. The larger the molten bath carbon content, the faster the melting of the steel bar. The main reason is that the carburizing reaction can be accelerated by the larger carbon concentration gradient between the molten metal and the steel bar. Figure 2.10 compares the microstructure of the steel bar in molten bath with the carbon content being 4.030 and 2.100%. The thickness of the acicular martensite layer with high carbon molten bath is 100 µm while that with low is 75 µm, which also demonstrates that the larger carbon concentration gradient could benefit the transfer of carbon from the molten bath to the steel bar. Figure 2.11 shows the center temperature change of the steel bar in the molten bath with different carbon content. It can be seen that the larger the molten bath carbon content, the higher the steel bar center temperature. The heating rate with the larger molten bath carbon content is faster than that with lower molten bath carbon content. This also demonstrates that the higher carbon content could accelerate the melting of the steel scrap.

2.3.3.3

Effect of the Molten Bath Fluid Flow on the Melting of Steel Bar

The steel bars with the diameter being 20 mm were respectively inserted into the medium carbon molten bath with different bottom-blowing gas flow rates. The weight and the diameter of the steel bar during the melting process were recorded. Figure 2.12 shows the weight change of the steel bar in molten bath with different bottom-blowing gas flow rates. In each case, it can be found that with the soaking time increasing, the weight of the steel bar decreases and the attenuation rate of the steel bar weight is faster and faster. For instance, in the molten bath with the bottomblowing gas flow rate being 5.0 L/min, the slopes of the linear regression are 1.05, 2.06, 2.87, 3.21 and 3.41 when the time is in the ranges of 0–30, 30–40, 40–55, 55–65 and 65–70 s. Figure 2.13 shows the diameter change of the steel bar in molten bath

30

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

Fig. 2.10 Microstructure of the steel bar in molten bath with different carbon content. a [% C] = 4.03; b [% C] = 2.10 Fig. 2.11 Center temperature change of the steel bar in molten bath with different carbon content

with different bottom-blowing gas flow rates. Similar with the variation tendency of the steel bar weight, the diameter of the steel bar decreases with the soaking time increasing and the attenuation rate of the steel bar diameter is faster and faster. For instance, in the molten bath with the bottom-blowing gas flow rate being 5.0 L/min,

2.3 Improving Scrap Melting by Carburization with Submerged Carbon …

31

the slopes of the linear regression are 0.040, 0.084, 0.125, 0.156 and 0.180 when the time is in the ranges of 0–30, 30–40, 40–55, 55–65 and 65–70 s. Based on the data shown in Fig. 2.12, the weight of the steel bar is lower with larger bottom-blowing gas flow rate. When the time is in the range of 0–30 s, the mass melting rate is 0.96, 0.105 and 1.06 g/s with the bottom-blowing gas flow rate being 3.0, 5.0 and 7.0 L/min and when the time is in the range of 40–55 s, the mass melting rate is 2.66, 2.87 and 3.02 g/s with the bottom-blowing gas flow rate being 3.0, 5.0 and 7.0 L/min. And the same phenomenon also can be found in the other time ranges. Meanwhile, as shown in Fig. 2.13, the diameter of the steel bar is lower with larger bottom-blowing gas flow rate. When the time is in the range of 0–30 s, the diameter reduction rate is 0.037, 0.040 and 0.041 mm/s with the bottom-blowing gas flow rate being 3.0, 5.0 and 7.0 L/min and when the time is in the range of 40–55 s, the diameter reduction rate is 0.114, 0.125 and 0.133 mm/s with the bottom-blowing Fig. 2.12 Weight change of the steel bar in molten bath with different bottom-blowing gas flow rates

Fig. 2.13 Diameter change of the steel bar in molten bath with different bottom-blowing gas flow rates

32

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

Fig. 2.14 Center temperature change of the steel bar in molten bath with different bottom-blowing gas flow rates

gas flow rate being 3.0, 5.0 and 7.0 L/min. Similar phenomenon also can be found in the other time ranges. In the induction furnace experiment, the molten bath stirring can be strengthened by the bottom-blowing gas and larger bottom-blowing gas flow rate means faster molten bath fluid flow. On the basis of the experimental data, it can be concluded that the molten bath fluid flow also demonstrates a significant impact on the steel bar melting. The faster the molten bath fluid flow, the faster the melting of the steel bar. On one hand, the convective heat transfer coefficient between the steel bar and the molten metal is tightly associated with the velocity of the surrounding fluid flow. The faster molten bath fluid flow would enlarge the convective heat transfer coefficient between the steel bar and the molten metal and therefore, the heat transfer from the molten metal to the steel bar is strengthened. On the other hand, the melting process of the carburized layer into the molten bath can also be enhanced by the molten bath with faster flow rate. Figure 2.14 shows the center temperature change of the steel bar in the molten bath with different bottom-blowing gas flow rate. It also can be found that the center temperature of the steel bar is higher with larger bottom-blowing gas flow rate. The heating rate with the larger bottom-blowing gas flow rate is faster than that with lower bottom-blowing gas flow rate. The change tendency of the steel bar center temperature further indicates that improving the stirring intensity of the molten bath would benefit the melting of the steel scrap.

2.3.3.4

The Convective Mass Transfer Coefficient of Carbon Between the Steel Bar and the Molten Metal

During the melting process of the steel bar in the molten metal, there is the carburizing reaction in the liquid–solid interface. The transfer of carbon on the liquid–solid interface can be presented by the following expression.

2.3 Improving Scrap Melting by Carburization with Submerged Carbon … Table 2.6 Results of the convective mass transfer coefficient of carbon between steel bar and molten metal

33

Label

K, × 10–5 m/s

Label

K, × 10–5 m/s

[% C] = 4.03

8.46

3.0 L/min

8.25

[% C] = 3.01

8.32

5.0 L/min

9.15

[% C] = 2.10

8.27

7.0 L/min

9.69

Average

8.35

Average

9.03

       ∂w Csi ∞ i i i K w(Cl ) − w(Cl ) = w(Cl ) − w(Cs ) v − D ∂ xi

(2.2)

where, w(Cl∞ ), w(Cli ) and w(Csi ) are the mass fraction of carbon in the molten metal, liquid phase of carburized layer and solid phase of carburized layer, respectively; K is the convective mass transfer coefficient of carbon between the steel bar and the i molten metal; v is the migration rate of the liquid–solid interface, m/s; w(Cl ) and i ∂w C ( s ) is the carburizing w(C i ) can be obtained by the iron-carbon diagram. D ∂ xi

s

quantity in the solid phase and it can be neglected because its value is very small. Therefore, Eq. (2.3) can be obtained and K can be calculated by Eq. (2.4).     K w(Cl∞ ) − w(Cli ) = w(Cl∞ ) − w(Cli ) v K =

  w(Cli ) − w(Csi ) v w(Cl∞ ) − w(Cli )

(2.3)

(2.4)

In this study, the temperature of molten bath is 1400 °C. According to the ironcarbon diagram, the value of w(Cli ) and w(Csi ) are, respectively, 2.05% and 0.6%. Table 2.6 listed the calculation result of K based on the experimental results. It can be found that the convective mass transfer coefficient of carbon between the steel bar and the molten metal is in the range of 8.0 × 10–5 m/s to 10.0 × 10–5 m/s. According to the research about the convective mass transfer coefficient of carbon K, the value of K is in the range of 1.8 × 10–5 m/s to 12.3 × 10–5 m/s [9–11]. Obviously, the value of K obtained in this study is consistent with the result of previous studies.

2.3.3.5

The Convective Heat Transfer Coefficient Between the Steel Bar and the Molten Metal

The fluid flow of the molten bath is the forced convection by the action of bottomblowing and the heat transfer between the molten bath and the steel bar is very intense. With the increasing of the fluid flow velocity, the Reynolds number increases, which would have a great impact on the boundary layer thickness. As the previous study reported, the calculation for the convective heat transfer between the steel bar and the molten metal is as follows [12].

34

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

Q = α · F · Δt

(2.5)

q = α · Δt

(2.6)

where, α is the convective heat transfer coefficient; Q is the heat of convection; q is the heat flux; F is the heat exchange area and Δt is the temperature difference of the thermal boundary layer. Based on the experimental data, the value of α is obtained by measuring the mass melting rate of the steel bar. During the melting of the steel bar in the molten bath, the heat supplied by the molten metal is the sum of the heat for the melting of steel bar and the heat transferred into the solid steel bar. Hence, the calculation formula of heat balance on the convective heat transfer surface at one point can be obtain as the following equation shows.

∂ts α · Δt = ρ H v − λ ∂ xi

 (2.7)

where, v is the migration rate of the

liquid–solid interface; ρ is the density of steel ∂ts bar; H is the heat of fusion; λ ∂ xi is the heat absorption of the solid steel bar and in this study, it can be neglected because its value is very small. According the previous research data [13, 14], H is 274.4 J/g and Δt is 7 °C with the molten bath temperature being 1400 °C. Therefore, the value of α can be obtained on the basis of the experimental results. As Table 2.7 lists, the value α of the three cases is 18,959, 20,736 and 22,555, respectively. It can be concluded that the value of α obtained in this study is consistent with the result of previous studies shown in Table 2.8. Table 2.7 Results of the convective heat transfer coefficient between steel bar and molten metal

Table 2.8 The convective heat transfer coefficient measured by previous studies [4, 7, 10]

Label (L/ min)

Steel bar mass melting rate (g/s)

Average heat transfer area (m2 )

A (W/ (m2 ·K))

3.0

2.95

0.0061

18,959

5.0

3.18

0.0060

20,736

7.0

3.42

0.0059

22,555

Researcher

A (W/(m2 ·K))

Yang

17,000–33,000

H. W. Hartog

40,000–55,000

H. Gaye

17,500–25,000

R. D. Pehlke

25,120

N. Hiroyuki

33,450

2.4 Materials and Energy Balance Model with Submerged Carbon Powder …

35

2.4 Materials and Energy Balance Model with Submerged Carbon Powder Injection On the basis of the material balance, the energy balance and the chemical equilibrium, the materials and energy balance model of EAF steelmaking with submerged carbon powder injection (SCPI) were built, which were tested in the case of a commercial EAF, as shown by Fig. 2.15 [15, 16]. The results of the theoretical calculation were in accord with the results of the industrial production listed in Table 2.9, which could demonstrate that the materials and energy balance model was reasonable. These two models were used to investigate the effect of the gas–solid parameters on the key technical parameters of EAF steelmaking.

2.4.1 Variation Tendency of the Molten Slag and the Off-Gas Volume Figure 2.16 shows the effect of the carbon powder quantity on the molten slag in EAF steelmaking with SCPI. The quantity of the molten slag increases with the carbon powder quantity increasing. The quantity of the molten slag is 82.33 kg/t when the carbon powder quantity is 10 kg/t and that increases to 89.54 kg/t when the carbon powder quantity is 40 kg/t. The main reason is that the ash content in the carbon powder is about 12.4% and almost all of the ash are removed into the molten slag during the steelmaking process. As for the smelting off-gas, the addition of carbon powder and CO2 influences the off-gas volume. Just as shown in Fig. 2.17, with the carbon powder quantity increasing, the volume of the off-gas increases noticeably. When the carbon powder quantity changed from 10 to 40 kg/t, the total volume of off-gas increased from

Fig. 2.15 The materials and energy balance system model of EAF steelmaking. a Material balance model; b energy balance model

36

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

Table 2.9 Comparison between the measured averaged values and the calculated averaged values Label 001 002 003 004 005 006

Carbon powder consumption (kg/t)

Power consumption (kWh/t)

O2 consumption (Nm3 /t)

Calculated

12.0

380.9

27.7

Measured

11.7

385.5

28.3

Calculated

16.0

369.1

33.1

Measured

16.3

265.3

34.2

Calculated

20.0

358.5

36.3

Measured

20.1

256.2

35.2

Calculated

23.0

354.7

39.5

Measured

22.7

357.6

42.6

Calculated

28.0

350.8

44.9

Measured

27.6

350.2

43.6

Calculated

32.0

348.3

49.2

Measured

32.3

343.2

47.3

Fig. 2.16 The relationship between the quantity of slag and the quantity of carbon powder

58.87 to 147.50 Nm3 /t, the volume of CO2 changes from 22.98 to 61.54 Nm3 /t and the volume of CO changes from 10.47 to 28.31 Nm3 /t. It can be found that the volume of CO2 increased by a large margin. With the assumed CO post-combustion ratio, a large amount of CO2 was formed due to the CO combustion and meanwhile, the powder carrier gas is also a main source of CO2 . The volume and composition change law of the off-gas in EAF steelmaking with SCPI can offer the reference basis data to optimize the cooperative operation between the smelting system and the dust-removal system in EAF steelmaking.

2.4 Materials and Energy Balance Model with Submerged Carbon Powder …

37

Fig. 2.17 The relationship between power consumption and off-gas volume and composition

2.4.2 Variation Tendency of the Power Consumption Based on the above analysis, the SCPI technology can accelerate the melting process of scrap by enhance the carburization of molten steel. Furthermore, a great deal of heat generated by the carbon and oxygen reaction can also accelerate the smelting rhythm and reduce the power consumption. Figure 2.18 shows the variation tendency of the power consumption with the quantity of carbon powder in EAF steelmaking with SCPI. On one hand, it can be found that the power consumption decreases with the quantity of carbon powder increasing. Based on the calculation results, a 1 kg/t increase in the quantity of carbon injected into the molten steel by SCPI would decrease the power consumption by about 3–4 kWh/t. On the other hand, the decrease of power consumption is gradually reduced with the quantity of carbon powder increasing. The average decrease speed of the power consumption per ton steel is about 3.6 kWh per kg carbon powder when the quantity of carbon powder is less than 20 kg/t and that would become 0.7 kWh per kg carbon powder when the quantity of carbon powder in the range of 30–40 kg/t. That is, the effective utilization efficiency of carbon powder decreases with the quantity of carbon powder increasing. With large injection rate of the carbon powder, the instantaneous ability in dissolving carbon powder of the molten bath is limited and hence, a part of carbon powder escapes directly from the molten steel into the furnace tank before they are captured by the molten steel. Furthermore, the more the carbon powder injected, the more the carbon escaping into the furnace tank. Hence, the effective utilization efficiency of carbon powder decreases with more carbon powder. Therefore, the injection parameters of the SCPI technology should be adjusted and optimized according to the actual smelting conditions during the EAF steelmaking process, such as the burden design, the smelting stage and so on.

38

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

Fig. 2.18 The relationship between the power consumption and the quantity of carbon powder

2.4.3 Effective Utilization and Burning Loss of Carbon Powder Figure 2.19 shows the relationship between the carbon powder burning loss and the quantity of carbon powder. It can be found that when the quantity of carbon powder is less than 20 kg/t, the burning loss rate of carbon powder with O2 is almost invariable and the burning loss rate of carbon powder with CO2 decreases gradually with the increasing quantity of carbon powder. In this range, with the carbon powder quantity increasing, the proportion of carbon powder melting into the molten steel decreases and correspondingly, the power consumption decreases. When the quantity of carbon powder is in the range of 20 to 40 kg/t, the burning loss rate of carbon powder with O2 is increases gradually and the burning loss rate of carbon powder with CO2 decreases slightly with the quantity of carbon powder increasing. It demonstrates that with the quantity of carbon powder increasing, more and more carbon powder run out from the molten bath into the furnace tank without carburization and this part carbon powder was burnt by O2 in the furnace tank directly. However, compared with the carbon and oxygen reaction in the molten steel during the steelmaking process, a few part of heat generated by the combustion reaction between carbon powder and O2 in the furnace tank can utilized by the molten bath. Therefore, the decreased speed of the power consumption is gradually reduced. Based on the industrial application research, the melting down carbon contents with different carbon powder quantity were measured and the corresponding effective utilization ratio of carbon powder melting into the molten steel was also calculated as shown in Figs. 2.20 and 2.21. When the quantity of carbon powder is less than 20 kg/t, the melting down carbon content of the EAF molten bath is proportional to the quantity of carbon powder and the effective utilization ratio of carbon powder increases stably. When the quantity of carbon powder is in the range of 20–40 kg/t, it can be found that with the quantity of carbon powder increasing, the melting down

2.4 Materials and Energy Balance Model with Submerged Carbon Powder …

39

Fig. 2.19 The relationship between the carbon powder burning loss and the quantity of carbon powder

Fig. 2.20 The melting down carbon content with different carbon powder quantity

carbon content increases more and more slowly and the effective utilization ratio of carbon powder decreases correspondingly.

2.4.4 Effect of CO Post-combustion on the Power Consumption In EAF steelmaking, a large amount of CO is generated during the smelting process and these CO gas contains massive chemical latent heat, which is useful to reduce the smelting energy consumption [17, 18]. Actually, it is significant to enhance the CO post-combustion and improve the energy utilization rate of CO post-combustion. Figure 2.22a, b show the relationship between the power consumption and the CO post-combustion ratio and the relationship between the power consumption and the energy utilization rate of CO post-combustion, respectively. In Fig. 2.22a, the energy

40

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

Fig. 2.21 The rate of carbon powder into molten steel with different carbon powder quantity

utilization rate of CO post-combustion is assumed as 42%. And with the CO postcombustion ratio increasing, the power consumption decreases. A 1% increase in the CO post-combustion ratio would decrease the power consumption by 0.76 kWh/ t. In Fig. 2.22b, the CO post-combustion ratio is assumed as 55%. And with the energy utilization rate of CO post-combustion increasing, the power consumption decreases. A 1% increase in the energy utilization rate of CO post-combustion would decrease the power consumption by 1.6 kWh/t. In EAF steelmaking with SCPI, the quantity of carbon powder injected into the EAF steelmaking system is far more than that in traditional EAF steelmaking. Hence, the utilization and optimization of CO post-combustion is of greater importance. Generally, the CO post-combustion includes CO post-combustion in foaming slag and CO post-combustion in free space [19, 20]. With CO post-combustion in free space, the heat generated by the reaction between O2 and CO in above the molten bath is transformed to the molten slag layer by radiation and convection and then the heat is transformed from the molten slag layer to the molten steel. In this heat transfer process, the energy utilization rate is only about 20–40%. With CO post-combustion in foaming slag, the heat generated is directly transformed to the molten steel from the molten slag layer, which improves the efficiency of energy utilization noticeably. Therefore, in EAF steelmaking with SCPI, it is very important to enhance the CO post-combustion, especially the CO post-combustion in foaming slag, to increase the energy utilization rate and reduce the power consumption.

2.5 Conclusions 1. The molten slag particles produced in the process of submerged O2 -CaO injection in EAF steelmaking have excellent dephosphorization ability. For the molten slag particles with submerged O2 -CaO injection, the content of P2 O5 is more than

References

41

Fig. 2.22 The relationship between the power consumption and the CO post-combustion parameters. a CO post-combustion ratio; b energy utilization rate of CO post-combustion

10%, the highest is 29.6%, and the content of T.Fe is only 1.0%, which provides conditions for low cost and rapid dephosphorization in EAF steelmaking. 2. In EAF steelmaking with SCPI, both the molten bath carbon content and the molten bath fluid flow have a significant effect on the scrap melting. The larger the molten bath carbon content, the faster the molten bath fluid flow, the faster the melting of the scrap. The carburizing reaction can be accelerated by the larger carbon concentration gradient and the faster molten bath fluid flow can strengthen the heat transfer from the molten metal to the scrap by enlarging the convective heat transfer coefficient. 3. The materials and energy balance model of EAF steelmaking with SCPI was in accord with the industrial application. And based on the results of these two models, it can be found that with the carbon powder quantity increasing, the power consumption decreases, the quantity of the molten slag increases and the off-gas volume increases. When the quantity of carbon powder is more than 20 kg/ t, the effective utilization efficiency of carbon powder decreases with the quantity of carbon powder increasing. The CO post-combustion is useful to reduce the smelting energy consumption in EAF steelmaking. Hence, with SCPI, it is very important to enhance the CO post-combustion, especially the CO postcombustion in foaming slag, to increase the energy utilization rate and reduce the power consumption.

References 1. Miyata M, Tamura T, Higuchi Y (2017) Development of hot metal dephosphorization with lime powder top blowing: part 1. Low blowing rate. ISIJ Int 57(2):1751–1755 2. Miyata M, Tamura T, Higuchi Y (2017) Development of hot metal dephosphorization with lime powder top blowing: part 2. High blowing rate. ISIJ Int 57(4):17561

42

2 Mechanism of EAF Steelmaking with Submerged Mixed Injection

3. Nozaki T, Takeuchi S, Haida O, Emi T, Morishita H, Sudo F (1983) Mechanism of hot metal dephosphorization by injecting lime base fluxes with oxygen into bottom blown converter. Trans Iron Steel Inst Jpn 23(6):513–521 4. Wei G, Zhu R, Tang T, Dong K (2019) Study on the melting characteristics of steel scrap in molten steel. Ironmaking Steelmaking 46(7):609–617 5. Matsuura K, Itoh Y, Narita T (1993) A solid-liquid diffusion couple study of a peritectic reaction in iron-carbon system. ISIJ Int 33(2):583–587 6. Kim YU, Pehlke R (1974) Mass transfer during dissolution of a solid into liquid in the ironcarbon system. Met Trans 5(23):2527–2532 7. Gaye H, Wanin M, Gugliermina P, Schittly P (1986) Vitesse de fusion des ferrailles au convertisseur. Modèle théorique et expérimentation industrielle. Rev Mét 82(3):793–806 8. Shukla AK, Deo B, Millman S, Snoeijer B, Overbosch A (2010) An insight into the mechanism and kinetics of reactions in BOF steelmaking: theory versus practice. Steel Res Int 81(5):940– 948 9. Wright JK (1989) Steel dissolution in quiescent and gas stirred Fe/C melts. Metall Mater Trans B 20(2):363–374 10. Mineo K, Susumu M (1967) Dissolution of steel cylinder into liquid Fe-C alloy. Tetsu-toHagane 53(2):983–997 11. Kazumi M, Hiroyuki N (1969) Study on the rate of scrap melting in the steelmaking process. Tetsu-to-Hagane 55(5):347–354 12. Yang W, Jiang X, Lin L, Peng X, Wang M, Shi X (2017) Thermal simulation experiments on scrap melting. Iron Steel 52(3):27–35 13. Yamamoto T, Ujisawa Y, Ishida H, Takatani K (2007) Operation and design of scrap melting process of the packed bed type. ISIJ Int 39(2):705–714 14. Pehlke RD (1980) BOF steelmaking introduction. Metalluryical Industry Press, Beijing, China, pp 105–110 15. Opitz F, Treffinger P, Wöllenstein J (2017) Modeling of radiative heat transfer in an electric arc furnace. Metall Mater Trans B 48(5):1–15 16. Logar V, Dovžan D, Škrjanc I (2012) Modeling and validation of an electric arc furnace: part 1, heat and mass transfer. ISIJ Int 52(3):402–412 17. Tang X, Kirschen M, Abel M et al (2003) Modelling of EAF off-gas post combustion in dedusting systems using CFD methods. Steel Res Int 74(4):201–210 18. Krassnig HJ, Kleimt B, Voj L et al (2008) EAF post-combustion control by on-line laser-based off-gas measurement. Arch Metall Mater 53(2):455–462 19. Kirschen M, Velikorodov V, Pfeifer H (2006) Mathematical modelling of heat transfer in dedusting plants and comparison to off-gas measurements at electric arc furnaces. Energy 31(14):2926–2939 20. Kim DS, Jung HJ, Kim YH et al (2015) Optimisation of oxygen injection in shaft EAF through fluid flow simulation and practical evaluation. Ironmaking Steelmaking 41(5):321–328

Chapter 3

Impact Characteristics of Submerged Gas Injection

3.1 Introduction Intensifying oxygen supply is an important part of the smelting method used in modern EAF steelmaking. This reduces the overall power consumption by enhancing the input of chemical energy into the EAF molten bath, which can also accelerate the fluid flow in the molten bath and improve the metallurgical reactions [1–4]. As a new technology, submerged mixed injection can deliver oxygen into the molten steel directly [5, 6] and it is necessary to know its impact influence on the molten bath. The impact characteristics of oxygen jets have been widely studied, though they mainly focus on the oxygen jet above the molten bath, such as coherent or conventional supersonic jets. Theoretical modeling, water model experiments, and numerical simulations are widely used to study the impact cavity of coherent and conventional supersonic jets under different operating conditions [7–11]. However, in EAF steelmaking, few studies have reported on impact characteristics of the submerged gas jet in an EAF. G. A. Irons analyzed the mechanism of bubble formation at nozzles in pig iron and built a mathematical description model of the initial bubble size; however, the effect of bubble break-up was not taken into account [12]. Bottin carried out a water model experimental study of gas penetration depth for a submerged side-blown equipment with a lower gas flow rate and developed a novel empirical equation using a MATLAB digital image processing algorithm [13]. Oryall used a water model and an electro resistivity probe to measure the gas volume fraction and bubble frequency generated by a submerged gas jet to focus on the mechanisms of the jet breaking up and bubble formation [14]. Hoefele analyzed the results of the water model experiment to determine the dynamics of a gas jet discharging horizontally into a the liquid bath of a nickel converter [15]. Zhan built a Computational Fluid Dynamics (CFD) model to analyze gas mixing behavior in a side-blown metallic bath [16]. To some extent, the submerged gas jet behavior in EAF steelmaking is similar to the Ar-O2 jet behavior in an argon-oxygen-decarburization (AOD) furnace [17– 23]. Tilliander developed a three-dimensional, three-phase model of gas injection in an AOD converter to investigate the flow field of the molten bath [17]. Andersson © Metallurgical Industry Press 2024 G. Wei and R. Zhu, Electric Arc Furnace Steelmaking with Submerged Mixed Injection, https://doi.org/10.1007/978-981-99-4602-0_3

43

44

3 Impact Characteristics of Submerged Gas Injection

studied the flow field of an AOD converter with six tuyeres from an existing mathematical model that solved the transport equations in three dimensions as well as for three phases [18]. Tilliander developed a fundamental mathematical model of gas injection in the AOD converter process and compared the results of the predicted and experimentally determined gas plumes and other model predictions [19]. Wei proposed a mathematical model for molten steel flow in the combined side and top blowing AOD refining process [20]. Odenthal investigated the fluid flow characteristics in the AOD converter by a combination of operational measurements, physical simulation on a downscaled water model, and numerical simulation using volume of fluid (VOF) model [21]. However, these research results in the AOD converter are not adapted to reflect the impact characteristics of the submerged gas jet in EAF steelmaking. The furnace structure, the submerged depth, and the operating parameters of the submerged gas jet in an EAF are very different from those in an AOD. Therefore, it is necessary to study the impact characteristics of the submerged gas jet in EAF steelmaking systemically, which would benefit the optimization of process parameters. This chapter is focused on analyzing the impact characteristics of the submerged gas jet in EAF steelmaking. The validity of the numerical model was first examined by water model experiments. A theoretical model was built to depict the axis trajectory of the submerged gas jet in the liquid bath. The effect of the installed angle and the gas flow rate on the horizontal and vertical penetration distances of the submerged gas jet was analyzed by combining the results of the numerical simulation and the water model experiment. Moreover, the effect of operating parameters on the erosion of the refractory material around the submerged injector was also analyzed.

3.2 Numerical Model and Water Model Experiment The objective of this section is to reveal, by numerical simulation and water model experiment, the fundamental sub-phenomena of submerged injection under steelmaking conditions. The VOF model was adopted in the numerical simulation.

3.2.1 Numerical Model 3.2.1.1

Assumptions

1. The flow of liquid and oxygen are three-dimensional, transient, and nonisothermal. 2. The gas phase is a compressible Newtonian fluid whereas the molten steel is an incompressible Newtonian fluid. 3. A non-slip condition has been applied to all of the walls, and a standard wall function has been used to model the mean velocities near the wall.

3.2 Numerical Model and Water Model Experiment

45

4. The injection medium only consists of oxygen, and other chemical reactions need not be considered. 3.2.1.2

Governing Equations

During the numerical simulation, the VOF model was used. The relative continuity, momentum, and energy conservation equations were as follows [24, 25]: Continuity equation: ∂ρ ∂ + (ρu i ) = 0 ∂t ∂ xi

(3.1)

Momentum conservation equation:   ∂ ∂p ∂  ∂  τij − ρu i' u 'j ρu i u j = − + (ρu i ) + ∂t ∂x j ∂ xi ∂x j

(3.2)

Energy conservation equation:  ∂ ∂  ∂ qi + c p ρu i' T ' (ρ Eu i + pu i ) = − (ρ E) + ∂t ∂ xi ∂ xi  ∂  τij u j − ρu i' u 'j u j + ∂ xi

(3.3)

where ρ is the fluid density; u i and u j are, respectively, the mean velocity components in the i-th and j-th direction; u i' and u 'j are, respectively, the fluctuating velocity components in the i-th and j-th direction; T ' is the fluctuating temperature; c p is the specific heat at constant pressure of gas; E is the total energy; qi is the transfer rate of heat conduction; τij is the stress tensor of the surface molecular viscous force. τij and the Reynolds stress (−ρu i' u 'j ) can be calculated by the following equations:   ∂u j ∂u i 2 ∂u k τij = μ + − δij ∂x j ∂ xi 3 ∂ xk     ∂u j ∂u i 2 ∂u k − δi j ρk + μt −ρu i' u 'j = μt + ∂x j ∂ xi 3 ∂ xk

(3.4) (3.5)

where μ, μt , and k are, respectively, the molecular viscous force, the turbulent viscosity, and the turbulent kinetic energy of the fluid. The two-equation standard k-ε turbulence model was used to model the Reynolds stress, −ρu i' u 'j , to mathematically close Eq. (3.5). The following equations were used to calculate the turbulent kinetic energy k and its dissipation rate ε. ∂u j ∂(ρk) ∂(ρku i ) ∂ μt ∂k + = (μ + ) − ρu i' u 'j − ρε ∂t ∂ xi ∂x j σk ∂ xi ∂ xi

(3.6)

46

3 Impact Characteristics of Submerged Gas Injection

∂(ρε) ∂(ρεu i ) ∂ μt ∂ε = (μ + ) + ∂t ∂ xi ∂x j σε ∂ xi ∂u j ε ε2 − C1ε ρu i' u 'j − C2ε ρ k ∂ xi k

(3.7)

The turbulent viscosity μt (Pa·s) was computed by combing k and ε as follows: μt = ρCμ

ε2 k

(3.8)

In Eqs. (3.6)–(3.8), C 1ε , C 2ε , C 3ε , σ k , σ ε , and C μ are constants, and their values are 1.44, 1.92, 0.8, 1.0, 1.3, and 0.09, as suggested by Launder [26]. Studies have been carried out to modify the standard k-ε turbulence model based on a large temperature gradient. Heinz modified the standard k-ε turbulence model by considering the compressibility of fluid [27], and Abdol-Hamid set C μ as a function of local stagnation temperature by analyzing the effect of the temperature gradient on the turbulence model [28]. Based on the results of Abdol-Hamid’s model and a large number of experiments, Alam proposed a new method to modify the standard k-ε turbulence model that could eliminate the influence of the temperature gradient on the accuracy of the k-ε turbulence model [29]. The results of Alam’s modified model were in good accord with the experimental results reported [30]. In this section, the oxygen is directly injected into the molten steel, causing a large temperature gradient. Therefore, Alam’s modified model was used in numerical modeling. In the modified k-ε turbulence model, the value of C μ was modified by dividing its standard value (0.09) by the variable C T to achieve the desired decrease in turbulent viscosity at the shear layer, which in turn reduced the growth rate of mixing, as Eqs. (3.9)–(3.12) show. Cμ =  CT = 1 +

0.09 CT 1.2Tg0.6

(3.9)

1 + f (Mτ ) / 2k Mτ = a   2 H (Mτ − Mτ0 ) f (Mτ ) = Mτ2 − Mτ0

(3.10)

(3.11) (3.12)

where M τ is the turbulence Mach number; a is the speed of sound; H is the Heaviside function; and M τ0 = 0.1.

3.2 Numerical Model and Water Model Experiment

3.2.1.3

47

Computation Methodology

Figure 3.1 shows the geometric representation and the representative grids adopted in the numerical simulations. The numerical calculations were performed using hexahedral structured grids, and the grids near the submerged injector were refined to obtain meaningful numerical results. The installed angle of the submerged gas jet is defined as the angle between the submerged injector central axis and the horizontal direction. Table 3.1 shows the parameters used in the present numerical simulation while Table 3.2 shows the physical properties of the relevant fluids. Based on its industrial application, the diameter of the submerged injector exit is 12 mm, and three kinds of gas flows rates were studied. In addition, four installed angles were adopted to investigate the effect of the installed angle on the impact characteristics of the submerged gas jet, shown in Fig. 3.1. Three grid refinement levels were examined: coarse grid (321,562 cells), medium grid (496,076 cells), and fine grid (567,820 cells). Figure 3.2 depicts the jet axial velocity distribution using the three grid levels. The average percentage variation

Fig. 3.1 Detailed grid arrangements of the numerical model

Table 3.1 Parameters used in the numerical simulation

Items

Numerical simulation

The submerged injector diameter (mm)

12

Gas flow rate (Nm3 /h)

200, 400, 600

Angle between the jet and the horizontal direction (°)

0, 5, 10, 15

48

3 Impact Characteristics of Submerged Gas Injection

Table 3.2 Physical properties of the fluids Items

Molten steel

Molten slag

Oxygen

Water

Viscosity (kg/(m·s))

0.0065

0.35

1.92 × 10–5

7.98 × 10–3

Density (kg/m3 )

7200

3000

1.29

997

Thermal conductivity (W/(m·K))

15

1.2

0.0246

0.0242

Specific heat (J/(kg·K))

670

1200

919.31

4182

Surface tension (N/m)

1.6

0.4

–

0.071

Temperature (K)

1873

1873

298

298

Fig. 3.2 Axial velocity of the coherent supersonic jet at three grid refinement levels

of the axial velocity profile calculated with the coarse and medium grid levels was approximately 1.6 pct, with a maximum deviation of 4.5 pct. Between the medium and fine grid levels, the variation could be considered negligible (less than 1 pct). To maximize the computational time and efficiency, the results obtained with the medium grid were used for analysis and discussion. A non-slip condition was applied to the walls, and the standard wall function was adopted. A mass-flow inlet boundary condition was adopted at the inlet of submerged injector, and a pressure outlet boundary condition was used. The initial gauge pressure of the inlet and the outlet were, respectively, 500,000 Pa and 101,325 Pa. An initial time step of 10–5 s followed by an adaptive method were chosen for the time steps with a global courant number of 0.5 to improve the efficiency of calculation and reduce the operation time. During the numerical simulation, the calculations were conducted in a transient solution mode, and the pressure–velocity coupling scheme was achieved with the PISO algorithm. A body-force-weighted discretization scheme was used for pressure, while geometric reconstruction was used for the interface interpolation method. The momentum and mass equations were solved with second order upwind schemes. In this section, the numerical simulation was considered to be convergent in all cases when the residuals of energy and other dependent variables were, respectively, less than 10–6 and 10–3 .

3.2 Numerical Model and Water Model Experiment

49

3.2.2 Water Model Experiment As illustrated in the experimental platform in Fig. 3.3, a water model experiment was carried out to study the impact characteristics of the submerged gas jet and to validate the CFD model built in Sect. 3.2.1. Based on the similarity of kinetics and geometry, the physical model was built at a scale of 1:3. In this section, a modified Froude number was used to maintain the similarity between the two systems and the equations are as follows: Fr ' = Fr1' , that is Q = Qm

/

L Lm

ρg u 2 ρgm u 2m = ρl gL ρlm gL m

(3.13)

ρgm ρl × ρlm ρg

(3.14)

5 ×

where um and u are the gas velocities of the water model and the CFD model; Qm and Q are the gas flows of the water model and the CFD model; L m and L are the characteristic lengths of the water model and the CFD model; ρ lm and ρ l are the liquid density of the water model and the CFD model; ρ gm and ρ g are the gas density of the water model and the CFD model; and g is gravity acceleration. Table 3.3 shows relative parameters used in the water model experiment. The diameter of the submerged injector exit is 4 mm. In the water experiment, water and air were employed to represent molten steel and oxygen, respectively. Four schemes of gas flow rates and four schemes of installed angles were adopted to study their effects on the submerged gas jet behavior. Additionally, the effect of submerged depth was also investigated, which is defined as the distance from the submerged injector exit to the liquid bath surface. A high speed camera was employed to capture the instantaneous motion and the transient behavior of the submerged gas jet in the liquid bath at 25 fps (frames per second). Based on the experiment setup, the contour profiles of the submerged gas jet in the liquid bath were captured. The horizontal penetration distance and the vertical penetration depth of the submerged gas jet were measured at different experimental schemes as listed in Table 3.3. Meanwhile, three repetition experiments have been done to ensure the reliability of results.

Fig. 3.3 Experimental instruments used in the water model experiment

50 Table 3.3 Parameters used in the water model experiment

3 Impact Characteristics of Submerged Gas Injection

Items

Water model experiment

The submerged injector diameter (mm)

4

Gas flow rate (Nm3 /h)

3, 5, 8, 14

Angle between the jet and the horizontal direction (°)

0, 5, 10, 15

Submerged depth (mm)

100, 200

3.2.3 Model Validation and Error Analysis The water model was simulated by this section’s numerical approach using the same conditions as adopted in the water model experiment to validate the CFD model. The submerged depth was 200 mm, and the installed angle of the submerged injector was 0°. Table 3.4 compares the horizontal penetration distances of the submerged gas jets at different gas flow rates from the numerical simulations and the water model experiments. The results of the numerical simulations are very close to the measured data from the water model experiments. The error between the experimental data and the numerical results is no greater than 3.53 pct, with the average error being 2.75 pct. Hence, it can be concluded that the numerical simulation results are reasonably in accord with the experimental measurements.

3.3 Theoretical Modeling for Submerged Gas Jet The trajectory of the submerged gas jet was mathematically modeled by theoretical analysis and mathematical derivation to investigate the impact characteristics of the submerged gas jet in the liquid bath. Figure 3.4 shows the geometric model of the submerged gas jet, which was built based on the water model experiment phenomena and past studies [31]. Table 3.4 Comparison of the horizontal penetration distance between the water model experiment and the numerical simulation

Gas flow rate (Nm3 /h)

3

5

8

Water model experiment (mm)

85

135

200

Numerical simulations (mm)

82

138

205

Error (%)

3.53

2.22

2.50

3.3 Theoretical Modeling for Submerged Gas Jet

51

Fig. 3.4 Geometric model of the submerged gas jet for theoretical modeling based on the physical phenomenon

3.3.1 Assumptions 1. The submerged gas jet is regarded as a stable and incompressible fluid. 2. There is only mass and momentum exchange between the jet and the surrounding liquid. 3. The momentum of the submerged gas jet is conserved in the horizontal direction. 4. The transient variation of the momentum of the submerged gas jet in the vertical direction is equal to the buoyancy of the jet.

3.3.2 Theoretical Modeling As shown in Fig. 3.4, the microelement between the horizontal distance x and x + dx was considered a computing primitive and is vertical to the jet central axis. Therefore, the Eq. (3.15) can be obtained according to the law of mass conservation. Cg ρg Av = ρg A0 v0

(3.15)

where A and A0 are the jet sectional areas at the horizontal distance x and the injector exit, respectively; v and v0 are the jet velocities at the horizontal distance x and the injector exit, respectively; ρ g is the gas velocity. The submerged gas jet in the molten bath would become a gas–liquid jet by the effects of jet entrainment. Therefore, the following equation can be obtained: Cg + CL = 1

(3.16)

where C g and C L are the gas and liquid volume fractions in the jet at the horizontal distance x, respectively. In the horizontal direction, the momentum of the submerged gas jet is conserved, and the following equations can be obtained:

52

3 Impact Characteristics of Submerged Gas Injection

m x vx = m x0 vx0

(3.17)

ρm Av 2 cos θ = ρg A0 v02 cos α

(3.18)

where vx and vx0 are the velocity components in x-axis direction at the horizontal distance x and the submerged injector exit, respectively; α is the installed angle of the submerged injector; θ is the angle between the jet axis and the horizontal direction at the horizontal distance x; ρ m is the gas–liquid jet density and can be calculated by the gas density ρ g and the liquid density ρ L as shown in Eq. (3.19). ρm = Cg ρg + CL ρL

(3.19)

Therefore, the following equation can be obtained by combining Eqs. (3.15)– (3.18): ρL − ρg vx ρL − ρm = ρm ρg v0 cos α

(3.20)

In the vertical direction, the momentum of the submerged gas jet is conserved, and the following equation can be obtained:   d mv y = Fb

(3.21)

where m is the mass of the computing primitive at the horizontal distance x; vy is the velocity component in the y direction at the hjetorizontal distance x; F b is the buoyancy of the computing primitive at the horizontal distance x and can be calculated by Eq. (3.22). Fb = (ρL − ρm )g Ads

(3.22)

Equation (3.23) can be obtained by combining Eqs. (3.21) and (3.22). mdv y + v y dm = (ρL − ρm )g Ads

(3.23)

where ds is the axis length of the computing primitive at the horizontal distance x which can be calculated by the following equation. ds =

dx cos θ

(3.24)

Therefore, Eq. (3.25) can be obtained by combining Eqs. (3.23) and (3.24). dv y d(ln m) (ρl − ρm )g + vy − =0 dx dx ρm v cos θ

(3.25)

3.3 Theoretical Modeling for Submerged Gas Jet

53

Then, the following equation can be obtained by combining Eqs. (3.20) and (3.25).   ρL − ρg dv y d(ln m) g + vy − =0 dx dx ρg v0 cos α

(3.26)

In the present study, the following dimensionless variables were assumed. xr = vr =

vy x y vx , yr = , vxr = , v yr = d0 d0 v0 v0

ρg v02 v m  , mr = , Fr ' =  v0 ρg A0 v0 ρL − ρg gd0

where d 0 is the injector exit diameter and Fr ' is the modified Froude. Based on Eqs. (3.26) and (3.17), the following equations can be obtained by incorporating these dimensionless variables. dv yr d(ln m r ) 1 =0 + v yr − ' dxr dxr Fr cos α mr =

cos α vxr

(3.27) (3.28)

Equation (3.29) can be obtained through mathematical transformations in combining Eqs. (3.27)–(3.28). d2 yr 1 = dxr2 Fr ' vxr cos θ

(3.29)

The relationship between vxr and x r is as follows as reported by Han. vxr =

K cos α xr + K

(3.30)

where K is a constant. Therefore, the following equation can be obtained by combining Eqs. (3.29)– (3.30): d2 yr 1 xr + K = dxr2 K Fr ' cos2 α The boundary conditions of Eq. (3.31) are listed as follows: xr = 0, yr = 0,

dyr = tan α dxr

(3.31)

54

3 Impact Characteristics of Submerged Gas Injection

Therefore, the formula of the axis of the submerged gas jet can be obtained according to the above boundary conditions.

1 3 K 2 sin 2α 1 ' x + xr + K Fr xr yr = K Fr ' cos2 α 6 r 2 2

(3.32)

Based on the installed angle, the gas flow rate, the submerged depth, and the submerged injector exit diameter, the axis of the submerged gas jet in the liquid bath can be described by Eq. (3.32).

3.4 Results and Discussion 3.4.1 Analysis of the Theoretical Model Comparing data from both the theoretical model and the water model experiment, the value of the constant, K, is 0.335. Figure 3.5 shows that the results of the theoretical model with K = 0.335 are in accordance with the results of the water model experiment, which demonstrates that the theoretical model results concur reasonably closely with the measurements. In EAF steelmaking, the depth from the submerged injector exit to the molten steel surface is 500–700 mm, and the submerged injector exit diameter is about 12 mm. Hence, the dimensionless quantity of the submerged depth, yr , is in the range of approximately 40–60. Figure 3.6 shows the effects of the installed angle on the axis trajectory of the submerged gas jet, obtained by the theoretical model when the gas flow rate is set to 500 Nm3 /h. Both the horizontal and vertical penetration distances are affected by Fig. 3.5 Comparison between the measured and theoretical values of the horizontal penetration depth by applying the constant K

3.4 Results and Discussion

55

the installed angle. The vertical penetration distance increases as the installed angle increases. Figure 3.7 magnifies the corresponding diagram labelled Part A in Fig. 3.6. Figure 3.7a, b show the variation of the jet axis trajectories for the ranges 0° ≤ α ≤ 20° and α ≥ 20°, respectively. For 0° ≤ α ≤ 20°, the horizontal penetration distance increases as α increases. The trend is more obvious in the range of 0° to 10°. When α ≥ 20°, the horizontal penetration distance decreases as α increases, and the attenuation rate increases as α increases. Furthermore, the effect of the installed angle on the erosion of the surrounding refractory material was also taken into account. Figure 3.8 shows the axis trajectory of the submerged gas jet with different gas flow rates when the installed angle of the submerged injector (α) is 10°. As the gas flow rate increases, both the horizontal and vertical penetration distances increase. With the same installed angle and the same injector exit diameter, the jet velocity at the submerged injector exit increases by a larger gas flow rate. Therefore, both the Fig. 3.6 The axis trajectory of the submerged gas jet at different installed angles from the theoretical model

Fig. 3.7 Part A enlarged diagram from Fig. 3.6: the axis trajectory of the submerged gas jet at different installed angles. a 0° ≤ α ≤ 20°; b α ≥ 20°

56

3 Impact Characteristics of Submerged Gas Injection

Fig. 3.8 The axis trajectory of the submerged gas jet at different gas flow rates

horizontal and the vertical penetration depths of the submerged gas jet are larger. This phenomenon is well in accord with the results of the numerical simulations and water model experiments.

3.4.2 Analysis of the Water Model Experiment Figure 3.9 shows photos of the submerged gas jet at different installed angles and different gas flow rates in the water model experiment. Each row depicts the jet behaviors under the same installed angle with increasing gas flow rates, and each column depicts the jet behaviors at the same gas flow rate under increasing installed angles. When the gas flow rate is 3 Nm3 /h, the jet velocity at the injector exit is small and many gas bubbles are formed at the outlet of the injector. These bubbles float up under the effect of buoyancy. As the gas flow rate increases, the diameter of the gas bubble increases, and, as a result, the larger bubbles make the liquid splash more severely. When the gas flow rate is more than 5 Nm3 /h, the shape of the gas in the liquid bath changes into a jet state from a bubble state. Furthermore, both the horizontal and vertical penetration distances increase, and the mixing intensity of liquid bath is improved. When the gas flow rate is 14 Nm3 /h, there are a few large bubbles, as shown in Fig. 3.9. With a higher gas flow rate, the exiting gas jet can maintain a certain velocity within a certain distance, which can increase the penetration depth in the horizontal and vertical directions. Based on the energy conservation law, the energy is transferred from the gas jet to the surrounding liquid, and because of the entrainment of the gas jet, a gas–liquid jet is formed. As a result, the jet velocity gradually attenuates, and the gas floats up to the liquid surface in the form of small bubbles. Figures 3.10 and 3.11 show the penetration distance of submerged gas jet in the horizontal direction and the vertical direction, respectively, in the water model experiments. At a given installed angle, both the horizontal and the vertical penetration

3.4 Results and Discussion

57

Fig. 3.9 Photos of the submerged gas jet at different installed angles and gas flow rates in the water model experiment

distances increase as the gas flow rate increases. Meanwhile, as the installed angle increases, the horizontal penetration distance decreases while the vertical penetration distance increases, which is in accord with the results of both the theoretical model and the numerical simulation. Therefore, optimizing the installed angle is important to improve mixing of the molten bath in both the horizontal and vertical penetration distances. Figure 3.12 shows photos of the submerged gas jet with the submerged depths varied at 100 and 200 mm in the water model experiment. The jet behaviors in these two cases appear substantially similar. However, the bottom row of Fig. 3.12 shows that the splashing height of the liquid is lower with the larger submerged depth. Accordingly, the trajectory of the gas jet is longer in the horizontal direction, and more kinetic energy from the gas jet is transferred to the liquid bath. Therefore, larger circulation and stronger local eddies can be formed, which is helpful to improve the mixing intensity and the kinetic conditions of the liquid bath.

58

3 Impact Characteristics of Submerged Gas Injection

Fig. 3.10 The horizontal penetration distance of the submerged gas jet in water model experiments

Fig. 3.11 The vertical penetration distance of the submerged gas jet in water model experiments

Fig. 3.12 Photos of submerged gas jet with different submerged depths (y) and installed angles in the water model experiment

3.4 Results and Discussion

59

3.4.3 Analysis of the Numerical (CFD) Simulation Figure 3.13 depicts the velocity contour profiles of the submerged gas jet under different conditions. Each row shows the jet velocity distribution in the liquid steel bath under different installed angles different gas flow rates. The phenomena captured by the numerical simulations is consistent with the phenomena observed in the water model experiments. In the area near the injector exit, the gas jet has a large initial velocity, and due to its inertial force, the gas jet can generate a horizontal and vertical penetration depth at any set of conditions. Intense energy transfer occurs between the gas jet and the liquid steel because of the resistance and buoyancy of the liquid steel. As a result, the velocity of the gas jet attenuates gradually, and the gas floats up and escapes into the ambient atmosphere. At a given installed angle, the gas jet’s initial velocity increases as the gas flow rate increases, and accordingly, the energy transfer between the gas jet and the surrounding liquid is more intense. Therefore, a larger horizontal and vertical penetration distance can be obtained. At a given gas flow rate, as installed angle increases, the jet’s velocity component in the horizontal direction decreases while the jet’s velocity component in the vertical direction increases. Therefore, with a larger installed angle, the horizontal penetration distance is shorter while the vertical penetration distance is longer. Figures 3.14 and 3.15 show the penetration distances of the submerged gas jet in the horizontal and vertical directions, respectively, which were obtained by the

Fig. 3.13 The velocity contour profile of a submerged gas jet with different installed angles and gas jet flow rates. a 200 Nm3 /h; b 400 Nm3 /h; c 600 Nm3 /h

60

3 Impact Characteristics of Submerged Gas Injection

numerical simulations. As Fig. 3.14 shows, when the installed angle is 0°, the horizontal penetration distance increases from 280 to 535 mm when the gas flow rate increases from 200 to 600 Nm3 /h. Meanwhile, when the gas flow rate is 600 Nm3 / h, the horizontal penetration distance decreases from 535 to 308 mm as the installed angle increases from 0° to 15°. As Fig. 3.15 shows, when the installed angle is 15°, the vertical penetration distance increases from 92 to 247 mm when the gas flow rate increases from 200 to 600 Nm3 /h. Meanwhile, when the gas flow rate is 600 Nm3 / h, the horizontal penetration distance increases from 58 to 247 mm as the installed angle increases from 0° to 15°. It can be concluded that the installed angle has a significant impact on the impact characteristics of the submerged gas jet in the liquid bath. Fig. 3.14 The horizontal penetration distance of the submerged gas jet in the numerical simulation

Fig. 3.15 The vertical penetration distance of the submerged gas jet in the numerical simulation

3.4 Results and Discussion

61

3.4.4 Erosion of the Refractory Around the Submerged Injector The advantages of S-COMI on the metallurgical effects in EAF steelmaking are of significance and include stronger molten bath mixing and better dephosphorizing capacity. However, the life of the refractory material is a key technical index for new technology. As previous studies have reported, to consider erosion of the refractory material, the return stroke model and the gas etching model were proposed to analyze the erosion mechanism of the injection element [32, 33]. In the return stroke model, when the bubble leaves the injector, the liquid enters the space of the previous bubble, which could scour the surface of the injector and cause smooth erosion surface. In the gas etching model, when the bubbles close to the injector break up, the surrounding liquid generates an instantaneous high attack to the injector and as a result, the element erodes. In this case, some small pits can be found on the surface of the element. In this section, the erosion mechanism of the refractory around the submerged injector was investigated in both the water model experiment and the numerical simulation. As Fig. 3.16 shows, boric acid elements were used to monitor the erosion rate under different operating conditions. Figure 3.16a–c show the initial boric acid element, the installed boric acid element, and the used boric acid elements, respectively, from the water model experiment. Figure 3.17 shows the relationship between the erosion rate of the boric acid element and the gas flow rate in the water model experiment. When the gas flow rate is less than 4 Nm3 /h, the erosion rate of the boric acid element increases as the gas flow rate increases; in this process, the return stroke model plays a stronger role. When the gas flow rate is larger than 6 Nm3 /h, the erosion rate of the boric acid element also increases as the gas flow rate increases; in this process, however, the gas etching model plays a stronger role. In the range of 4 to 6 Nm3 /h, the erosion of the boric acid element first decreases and then increases as the gas flow rate increases. During this process, the blown gas state changes from the bubble state to the jet state, causing the effect of the return stroke model to be weakened while the effect of the gas etching model is slightly strengthened. Therefore, due to the comprehensive

Fig. 3.16 The boric acid element used in the water model experiment. a Initial boric acid element; b the installed boric acid element; c the used boric acid element

62

3 Impact Characteristics of Submerged Gas Injection

Fig. 3.17 The erosion rate of the boric acid element at different gas flow rates and installed angles

effect of the return stroke model and the gas etching model, the erosion rate of the boric element decreases in this range. In this section, the effect of the installed angle on the erosion of the boric acid element was studied. The injector element was divided into two parts: the upper part and the underpart. In each part, the radial length was measured at four positions, and the averaged value was adopted to analyze the erosion rate of the boric acid element. Figure 3.18 shows the relationship between the erosion rate of the boric acid element in the radial direction by installed angle. As the installed angle increases, the erosion rate of the upper part of the element decreases while that of the underpart increases. The main reason is that with the increased installed angle, the velocity component in the vertical direction increases, and the fluid flow accelerates. Therefore, according to the gas etching model, the erosion rate of the element increases. The wall shear stress distribution around the exit of submerged injector also can be obtained in the numerical simulation. The wall shear stress can reflect the erosion intensity of the element. The larger the wall shear stress, the faster of the erosion Fig. 3.18 The erosion rate of the boric acid element in radial direction with different installed gas jet angles

3.5 Conclusions

63

Fig. 3.19 The wall shear stress distribution around the exit of submerged injector at different installed angles and gas flow rates

rate. Figure 3.19 shows the wall shear stress around the submerged injector exit under different operating conditions. At the installed angle of 10°, Fig. 3.19 clearly shows the wall shear stress of the upper part is larger than that of the underpart, which is in accord with the results of the water model experiment shown in Fig. 3.18. Furthermore, as the gas flow rate increases, the wall shear stress increases. In addition, at a given gas flow rate, the wall shear stress of the underpart of the element increases as the installed angle increases, which is also in accord with the results obtained by the water model experiment.

3.5 Conclusions In this chapter, a three-dimensional, three-phase numerical (CFD) model, an experimental model, and a theoretical model of a submerged gas jet were developed to investigate the impact characteristics of submerged gas injection in EAF steelmaking. Water model experiments were carried out to validate the numerical model. This section investigated the influence of the installed angle and the gas flow rate on the horizontal and vertical penetration distances of the submerged gas jet in addition to

64

3 Impact Characteristics of Submerged Gas Injection

studying the erosion of the refractory around the submerged injector. The conclusions are as follows: 1. The applicability of the numerical model is validated by the reasonably good agreement between the results of the numerical simulation and the experimental data from the water model. 2. A theoretical model was built to depict the axis trajectory of the submerged gas jet in the liquid bath. The dimensionless quantity of the submerged depth, yr , is in the range of 40–60 approximately in EAF steelmaking. In this range, the horizontal penetration distance increases as the installed angle, α, increases from 0° ≤ α ≤ 20°; the trend is more apparent in the range from 0° to 10°. When α is larger than 20°, the horizontal penetration distance decreases as α increases, and the attenuation rate increases as α increases. 3. Based on the results of the water model experiment and the numerical simulation, we conclude that the gas flow rate and the installed angle exert significant influence on the behavior of the submerged gas jet in the liquid bath. As the gas flow rate increases, the horizontal and vertical penetration distances increase. Meanwhile, as the installed angle increases, the horizontal penetration distance increases while the vertical penetration distance decreases. 4. With the submerged gas jet, the gas etching model plays a stronger role in the erosion of the element around the submerged injector exit. As the installed angle increases, the velocity component in the vertical direction increases, the fluid flow is accelerated in that direction, and the wall shear stress on the underpart of the element increases. Therefore, the erosion rate of the upper part of the element decreases while that of the underpart increases. The measurements of the water model experiment were in good accord with the levels of erosion found in the numerical simulation.

References 1. Lee B, Sohn I (2014) Review of innovative energy savings technology for the electric arc furnace. JOM 66(9):1581–1594 2. Wei G, Zhu R, Dong K, Ma G, Cheng T (2016) Research and analysis on the physical and chemical properties of molten bath with bottom-blowing in EAF steelmaking process. Metall Mater Trans B 47(3):3066–3079 3. Memoli F, Mapelli C, Ravanelli P et al (2007) Simulation of oxygen penetration and decarburisation in EAF using supersonic injection system. ISIJ Int 44(8):1342–1349 4. Ma G, Zhu R, Dong K, Li Z, Liu R, Yang L, Wei G (2016) Development and application of electric arc furnace combined blowing technology. Ironmaking Steelmaking 34(8):594–599 5. Wei G, Zhu R, Wu X, Dong K, Yang L, Liu R (2018) Technological innovations of carbon dioxide injection in EAF-LF steelmaking. JOM 70(6):969–976 6. Wei G, Zhu R, Cheng T, Dong K, Yang L, Tang T, Wu X (2018) Effect of main gas composition on flow field characteristics of supersonic coherent jets with CO2 and O2 mixed injection (COMI) at steelmaking temperature. ISIJ Int 58(5):842–851 7. Muñoz-Esparza D, Buchlin J, Myrillas K, Berger R (2012) Numerical investigation of impinging gas jets onto deformable liquid layers. Appl Math Model 36(6):2687–2700

References

65

8. Alam M, Irons G, Brooks G, Fontana A, Naser J (2011) Inclined jetting and splashing in electric arc furnace steelmaking. ISIJ Int 51(9):1439–1447 9. Wei G, Zhu R, Cheng T, Dong K, Yang L, Wu X (2018) Study on the impact characteristics of coherent supersonic jet and conventional supersonic jet in EAF steelmaking process. Metall Mater Trans B 49(3):361–374 10. Lee M, Whitney V, Molloy N (2010) Jet-liquid interaction in a steelmaking electric arc furnace. Scand J Metall 30(5):330–336 11. Solórzano-López J, Zenit R, Ramírez-Argáez M (2011) Mathematical and physical simulation of the interaction between a gas jet and a liquid free surface. Appl Math Model 35(10):4991– 5005 12. Irons G, Guthrie R (1978) Bubble formation at nozzles in pig iron. Metall Mater Trans B 9(3):101–110 13. Ma J, Zhou P, Cheng W (2016) Dimensional analysis and experimental study of gas penetration depth model for submerged side-blown equipment. Exp Thermal Fluid Sci 75(3):220–227 14. Oryall G, Brimacombe J (1976) The physical behavior of a gas jet injected horizontally into liquid metal. Metall Trans B 7(3):391–403 15. Hoefele E, Brimacombe J (1979) Flow regimes in submerged gas injection. Metall Trans B 10(3):631–648 16. Zhan S, Lai C, Hsiao T (2003) CFD analysis of gas stirring behavior in side-blown metallic bath. Cent South Univ Technol 32(2):148–151 17. Tilliander A, Jonsson LTI, Jönsson PG (2014) A three-dimensional three-phase model of gas injection in AOD converters. Steel Res Int 85(3):376–387 18. Andersson N, Tilliander A, Jonsson L et al (2012) An in-depth model-based analysis of decarburization in the AOD process. Steel Res Int 83(11):1039–1052 19. Tilliander A, Jonsson T, Jonsson P (2007) Fundamental mathematical modeling of gas injection in AOD converters. ISIJ Int 44(44):326–333 20. Wei J, He Y, Shi G (2011) Mathematical modeling of fluid flow in bath during combined side and top blowing AOD refining process of stainless steel: application of the model and results. Steel Res Int 82(6):693–702 21. Odenthal H, Thiedemann U, Falkenreck U et al (2010) Simulation of fluid flow and oscillation of the argon oxygen decarburization (AOD) process. Metall Mater Trans B 41(2):396–413 22. Wei J, Ma J, Fan Y et al (2013) Water modelling study of fluid flow and mixing characteristics in bath during AOD process. Ironmaking Steelmaking 26(5):363–371 23. Visuri V, Järvinen M, Kärnä A et al (2017) A mathematical model for reactions during topblowing in the AOD process: validation and results. Metall Mater Trans B 48(3):1868–1884 24. Hirt C, Nichols B (1981) Volume of fluid (VOF) method for the dynamics of free boundaries. J Comput Phys 39(1):201–225 25. Li Q, Li M, Kuang S, Zou Z (2015) Numerical simulation of the interaction between supersonic oxygen jets and molten slag-metal bath in steelmaking BOF process. Metall Mater Trans B 46(3):1494–1509 26. Launder B, Spalding D (1972) Lectures in mathematical model of turbulence. Academic Press, London, pp 124–129 27. Heinz S (2003) A model for the reduction of the turbulent energy redistribution by compressibility. Phys Fluids 15(3):3580–3583 28. Abdol-Hamid KS, Pao SP, Massey SJ et al (2004) Temperature corrected turbulence model for high temperature jet flow. Trans ASME-I-J Fluids Eng 126(5):844–850 29. Alam M, Naser J, Brooks G (2010) Computational fluid dynamics simulation of supersonic oxygen jet behavior at steelmaking temperature. Metall Mater Trans B 41(3):636–645 30. Sumi I, Kishimoto Y, Kikuchi Y et al (2006) Effect of high-temperature field on supersonic oxygen jet behavior. ISIJ Int 46(9):1312–1317 31. Gulawani S, Deshpande S, Joshi J (2007) Submerged gas jet into a liquid bath: a review. Ind Eng Chem Res 46(10):3188–3218 32. Ma G (2016) The application fundamental research on combined blowing of 70t EAF in Xining special steel. University of Science and Technology Beijing, Beijing, pp 100–108

66

3 Impact Characteristics of Submerged Gas Injection

33. Aoki T (2009) The mechanism of the back-attack phenomenon on a bottom blowing tuyere investigated in model experiments (related theory and technology). Tetsu-to-Hagane 76(11):1996–2003

Chapter 4

Impact Characteristics of Coherent Supersonic Jet

4.1 Introduction Supplying oxygen into the molten bath via impinging jets is a main feature of the EAF steelmaking process [1, 2]. In modern EAF steelmaking, both the coherent supersonic jet and the conventional supersonic jet have been widely used [3, 4]. The oxygen jets play an important role in the steelmaking process, because they can control bath stirring, chemical reaction kinetics, energy consumption, foaming slag formation, and bath recirculation by exchanging momentum, heat, and mass with the molten steel and slag [5]. Typically, the jet impact characteristics reflect, to a certain extent, the stirring effectiveness of the oxygen jets on the molten bath. In industrial production, the effect of the oxygen jet on the molten bath is generally controlled by adjusting the lance height and the gas flow rate [6]. The characteristic parameters of the impact cavity generated by the supersonic jet are difficult to measure in actual process conditions. In order to optimize the oxygen-supplying operation, the behavior of the gas-jet/liquid surface interaction has been widely studied using water models and numerical simulations. Molloy [7] studied the oscillatory nature of the impinging jet system and proposed three different mechanisms associated with the impact of the jet onto the liquid surface: dimpling, splashing, and penetrating. Collins [8] carried out a number of experiments in order to investigate the effect of jet momentum, lance angle, and lance height (distance between lance exit and molten steel surface). Nordquist [9] investigated the effect of the lance height and nozzle diameters on the penetration depth through water model experiments. Li and Li [10, 11] studied the cratering process of the impingement of top-blown gas jets on the liquid bath experimentally and theoretically, developed a theoretical model to predict the cavity depth of a two-layer liquid bath impinged by multiple gas jets in a basic oxygen furnace, and analyzed the transferring characteristics of momentum/energy during oxygen jetting into the molten bath by the multi-fluid volume of a fluid model. Bapin [12] analyzed the droplet generation characteristics in a top-blowing steelmaking converter and found that the ambient furnace temperature had a significant effect on droplet generation by affecting the jet © Metallurgical Industry Press 2024 G. Wei and R. Zhu, Electric Arc Furnace Steelmaking with Submerged Mixed Injection, https://doi.org/10.1007/978-981-99-4602-0_4

67

68

4 Impact Characteristics of Coherent Supersonic Jet

velocity attenuation. Sabah and Brooks [13, 14] carried out a cold model experiment to investigate the splash distribution in oxygen steelmaking and analyzed the power of the air jet around gas injection using an energy balance approach. Li and Harris [15, 16] studied gas dispersion phenomena and modeled the gas bubble volume of a narrow slot in a liquid by a water model experiment and theoretical analysis. Mikael Ersson [17] built a fundamental mathematical model of a top-blown bath and studied the flow field and surface deformation caused by an impinging jet. Muñoz-Esparza [18] developed a three-dimensional numerical model by the volume of fluid approach (VOF), which can present the flapping motion of the jet and cavity oscillations. Alam [19] investigated the relationship between the critical depth of penetration and the lance angle in EAF steelmaking by theoretical and water models. Solórzano-López [20] observed the interaction between a gas jet and a liquid free surface by mathematical and physical simulations, and compared different calculated expressions of the cavity depth. Lee [21] thought that the cavities formed by an inclined gas jet oscillated owing to the wave generated by the gas jet and the wave propagated along the base of the cavity. In addition, the fluid flow characteristics of the conventional supersonic jet and the coherent supersonic jet were thoroughly studied and widely reported [22–24]. However, the impact characteristics of the coherent supersonic jet have not been investigated systemically, and the previously reported mathematical models for the conventional supersonic or subsonic jet are not adapted to calculate the penetration depth of the coherent supersonic jet. The chapter is focused on analyzing the impact characteristics of the coherent and conventional supersonic jets in EAF steelmaking. The validity of the computational fluid dynamics (CFD) model was first examined by water model experiments. A hybrid model was built to calculate the penetration depth and the impact zone volume of the coherent and conventional supersonic jets by integrating the theoretical model and the CFD model. The hybrid model results showed good consistency with the CFD model data. Moreover, the effects of lance height on the jet penetration depth and the jet impact zone volume were also analyzed.

4.2 Numerical Modeling and Water Experiment The objective of the present study is to reveal the fundamental subphenomena in the molten bath induced by the coherent and conventional supersonic jets inside an EAF under steelmaking conditions, using the VOF approach and water model experiments. The validity of the VOF model was first examined, and it was then used to investigate the jet impact characteristics.

4.2 Numerical Modeling and Water Experiment

69

4.2.1 Numerical Modeling 4.2.1.1

Assumptions

1. The flow of liquid as well as oxygen was three-dimensional, transient, and nonisothermal. 2. The gas phase was regarded as compressible Newtonian fluid, whereas the liquids were regarded as incompressible Newtonian fluids. 3. A nonslip condition was applied to all of the walls and a standard wall function was used to model the mean velocities close to the wall. 4. Chemical reactions in the molten bath were not taken into consideration. 4.2.1.2

Governing Equations

The numerical simulations were conducted by integrating the Navier–Stokes equations by the Reynolds averaging method [25]. The momentum and energy conservation equations of the Navier–Stokes equations were written in a conservative form as follows: Equation of continuity:   n    1 ∂ m ji − m i j (αi ρi ) + ∇ · (αi ρi νi ) = Sαi + ρi ∂t i=1

(4.1)

where ρ i and α i are the density and volume fraction of the i-th phase; vi is the velocity component in the direction i, m/s; mij is the mass flowing from the i-th phase to the j-th phase, kg; mji is the mass flowing from the j-th phase to the i-th phase, kg; t is time, s; and S αi is the custom source. Momentum conservation equation:    ∂ (ρ v→) + ∇ · (ρ v→v→) = −∇ p + ∇ · μ ∇ v→ + ∇ v→T + ρ g→ + F ∂t

(4.2)

where ρ is the gas density, kg/m3 ; v is the instantaneous velocity of the fluid, m/ s; t is time, s; p is static pressure, Pa; μ is dynamic viscosity, Pa·s; g is the gravity acceleration, m/s2 ; and F is the other outside volume force, N/m3 . The standard k-ε turbulence model was adopted. The turbulence kinetic energy k (m2 /s2 ) and dissipation rate ε (m2 /s3 ) are, respectively, determined by the following transport equations. ∂(ρk) ∂(ρkvi ) ∂ + = ∂t ∂ xi ∂x j



 μt ∂k μ+ · σk ∂ xi

+ G k + G b − ρε − Y M + Sk

(4.3)

70

4 Impact Characteristics of Coherent Supersonic Jet

∂(ρε) ∂(ρεvi ) ∂ + = ∂t ∂ xi ∂x j



 μt ∂ε μ+ · σε ∂ xi

ε ε2 + C1ε (G k + C3ε G b ) − C2ε ρ + Sε k k

(4.4)

where ρ is the gas density, kg/m3 ; t is time, s; vi is the flow velocity of the fluid in the direction i, m/s; Gk is the turbulent kinetic energy generated by the average velocity, J; Gb is the turbulent kinetic energy generated by buoyancy, J; Y M is the turbulent dissipation rate generated by compressible turbulence pulsation; and S k and S ε are custom sources. The turbulent viscosity μt (Pa·s) was computed by combing k and ε as follows: μt = ρCμ

ε2 k

(4.5)

where C 1ε , C 2ε , C 3ε , σ k , σ ε , and C μ are constants and their values are 1.44, 1.92, 0.8, 1.0, 0.9, and 0.09, as suggested by Launder [26]. The energy equation, also shared among the phases, is as follows. ∂(ρ E) + ∇ · [u(ρ E + p)] = ∇ · (keff ∇T ) + Sh ∂t

(4.6)

In the VOF model, the energy E can be determined by the mass average method, as shown by the following equation.

n i=1 αi ρi E i E= n i=1 αi ρi

(4.7)

where E i for each phase is based on the specific heat of the phase and the shared temperature. The effective thermal conductivity k eff and ρ are shared by the phases. The source term S h contains contributions from radiation, as well as any other volumetric heat sources [27]. A CFD model, illustrated in Fig. 4.1, was built to investigate the jet impact characteristics. In Fig. 4.1, H 0 is the lance height, m; hg is the slag layer height, m; hl is the penetration depth in the molten steel, m; H is the axial free distance from the nozzle exit, m, and H = H 0 − hg ; and De is the diameter of the jet exit, mm. Accordingly, Fig. 4.2 shows the geometric representation and the representative grids used in the simulations. The numerical calculations were performed using hexahedral structured grids and the grids near the interface between different phases were refined to obtain more accurate numerical results. Three grid levels were examined: coarse grid (276,386 cells), medium grid (432,960 cells), and fine grid (563,286 cells). Figure 4.3 depicts the axial velocity distribution of the coherent supersonic jet at the jet centerline using the three grid levels. The average percentage variation of the axial velocity profile calculated with the coarse and medium grid levels was approximately 2.0 pct, with a maximum deviation of 4.2 pct. Between the medium

4.2 Numerical Modeling and Water Experiment

71

Fig. 4.1 Shape of the jet impinging cavity

and fine grid levels, the variation could be negligible (less than 1 pct). Considering the computational time and efficiency, the results obtained with the medium grid were used for analysis and discussion. Table 4.1 shows the parameters used in the numerical simulations and Table 4.2 shows the physical properties of the relevant fluids. Six lance heights were studied in the numerical simulations to analyze the impact characteristics of the coherent and conventional supersonic jets. Figure 4.4 illustrates the geometry of the coherent supersonic jet nozzle adopted in this section. The design Mach number of the supersonic nozzle is 2.0 and its inlet, throat, and exit diameters are, respectively, 40.4, 26.3, and 34.2 mm. The mass flow rates of the main oxygen, shrouding oxygen, and CH4 are, respectively, 0.992 kg/s, 0.238 kg/s, and 0.0595 kg/s for the coherent supersonic jet, and 0.992 kg/s, 0 kg/s, and 0 kg/s for the conventional supersonic jet. A nonslip condition was applied to walls and a standard wall function was adopted. Mass-flow inlet boundary conditions were adopted at the inlets of main oxygen, shrouding oxygen, and CH4 , and a pressure-out boundary condition was used at the Fig. 4.2 Detailed grid arrangements of the CFD model

72

4 Impact Characteristics of Coherent Supersonic Jet

Fig. 4.3 Axial velocity of the coherent supersonic jet with three grid levels

Table 4.1 Parameters used in the present numerical simulations

Items

CFD model

Water model

Molten bath hydraulic diameter D0 (mm)

2600

650

Molten steel depth H s (mm)

1200

300

Slag layer thickness hg (mm)

150

35

Lance height H 0 (mm)

900, 1150, 1650, 2150, 2650, 3150

100, 150, 200

Table 4.2 Physical properties of the fluids Items

Molten steel

Liquid slag

Oxygen

CH4 1.09 ×

Water

0.0065

0.35

1.92 ×

Density (kg/ m3 )

7200

3000

1.29

0.68

997

925

Thermal conductivity (W/(m·K))

15

1.2

0.0246

0.0332

0.0242

0.6

Specific heat (J/(kg·K))

670

1200

919.31

2222

4182

–

Surface tension (N/ m)

1.6

0.4

–

–

0.071

0.029

Temperature (K)

1873

1873

298

298

298

298

10–5

7.98 ×

Oil

Viscosity (kg/(m·s))

10–5

10–3

0.015

4.2 Numerical Modeling and Water Experiment

73

Fig. 4.4 Structure of the coherent supersonic jet

outlet position of the numerical model. The initial gauge pressures of the inlet and outlet gauges are, respectively, 400,000 Pa and 101,325 Pa. The calculations were conducted in a transient solution mode and the pressure– velocity coupling scheme was achieved with the PISO algorithm. The body-forceweighted discretization scheme was used for pressure, whereas geometric reconstruction was used for the interface interpolation method. The momentum and mass equations were solved with second-order upwind schemes. In this section, the numerical simulations in all cases were thought to be convergent when the residuals of energy and other dependent variables were, respectively, less than 10−6 and 10−3 . A time step of 10−5 s was used during the simulations.

4.2.2 Water Model Experiment and CFD Model Validation Based on the setup shown in Fig. 4.5, a water model experiment was carried out to investigate the jet impact characteristics and validate the CFD model built in Sect. 4.2.1. Table 4.1 also lists the dimensions of the water model, which was designed by scaling down the CFD model by a ratio of 1:4. In the experiment, water, oil, and air were used to represent molten steel, liquid slag, and oxygen. The lance can be adjusted by the regulator to change the lance height, and a high-speed camera was employed to capture the instantaneous motion and the transient behavior of the jet impingement in the liquid bath at 25 frames per second. The cavity depth was measured at different lance heights, as listed in Table 4.1. In the water model experiments, methane combustion is very difficult to model and is highly hazardous. Therefore, the lance used in the experiment was a conventional supersonic jet and its design mass flow rate was 0.00366 kg/s with a Mach number of 1.4. To validate the water model, it was simulated by the present CFD approach under the same conditions as adopted in the experiment. Figure 4.6 shows the jet transient impinging process at different lance heights in the water model experiment. Row (a) shows the original pictures captured by the high-speed camera and row (b) shows the

74

4 Impact Characteristics of Coherent Supersonic Jet

Fig. 4.5 Experimental instruments of the water model experiment

corresponding enlarged pictures. Figure 4.7 shows the profiles of the jet impact zone at different lance heights in the relative numerical simulations according to the water model experiment. Figure 4.8 depicts the penetration depth under different lance heights in the water model experiments and numerical simulations. Note that the numerical simulation results are very close to the experimental data under different lance heights and the maximum error is 4.21 pct, whereas the average error is 3.01 pct. Therefore, it can be concluded that the numerical simulation results agree reasonably well with the experimental results.

(a)

Cavity

Cavity

Cavity

(b)

100mm

150mm

200mm

Fig. 4.6 Jet impinging process at different lance heights in the water model experiment. a The original pictures; b the corresponding enlarged pictures

4.3 Hybrid Modeling for Jet Impact Characteristics

75

Fig. 4.7 Profiles of the jet impact zone at different lance heights in the relative numerical simulations Fig. 4.8 Penetration depth at different lance heights in water model experiments and numerical simulations

4.3 Hybrid Modeling for Jet Impact Characteristics The jet penetration depth and jet impact zone volume are important characteristics of the oxygen jet in the EAF steelmaking process. The penetration depth and volume of the impact zone generated by the coherent and conventional supersonic jets were calculated by combining the theoretical model and numerical simulations.

4.3.1 Assumptions 1. The surface tension of the molten bath was neglected and the slag foaming phenomenon during the steelmaking process was not considered. 2. The oxygen jet was regarded as an ideal compressible gas and during the interaction between the oxygen and the molten bath, the oxygen was absorbed by the molten bath completely. 3. The cross-section of the impact zone was regarded as a circle.

76

4 Impact Characteristics of Coherent Supersonic Jet

4.3.2 Theoretical Modeling of the Jet Penetration Depth According to previous research reports [28, 29], the shape of the jet impinging cavity is shown in Fig. 4.1. The pressure of the jet centerline at the bottom gas–liquid interface of the impinging cavity Pl is equal to the static pressure of the molten bath based on the principle of energy conservation, as shown by the following equation. Pl =

1 ρg v 2 = ρsteel gh l + ρslag gh g 2

(4.8)

where ρ g , ρ steel , and ρ slag are the density of the jet, molten steel, and slag, respectively, kg/m3 and v is the jet velocity of the centerline at the bottom of the impinging cavity, m/s. As previously reported [30], the relationship between Pl and the pressure of the jet centerline at the molten bath surface Px can be obtained by Eq. (4.9). 1

Pl = σ Arx2 Px

(4.9)

where Ar x is the Archimedes number of the jet at the molten bath surface; σ is an empirical constant, σ = 1.3–1.5; and Px can be calculated by Eq. (4.10) when the oxygen is absorbed completely by the molten steel during the oxygen jet interaction with the molten bath. Px =

Ix Ax

Ix = ρx vx2 Ax

(4.10) (4.11)

Using the momentum conservation of the oxygen jet, the following equation can be obtained. Ie = ρe ve2 Ae = ρx vx2 Ax = Ix

(4.12)

where, I x and I e are the jet momenta at the molten bath surface and the jet exit, respectively, kg·m/s; Ax and Ae are the jet areas at the molten bath surface and the jet exit area, respectively, m2 ; vx and ve are the jet velocities at the molten bath surface and the jet exit, respectively, m/s; and ρ x and ρ e are the jet densities at the molten bath surface and the jet exit, respectively, kg/m3 . Therefore, hl can be calculated by Eq. (4.14), as deduced from Eq. (4.13). ρe ve2 Ae = ρsteel gh l + ρslag gh g Ax

2 1 ρ v2 ρslag De e e h l = σ Arx2 − hg ρsteel g Dx ρsteel 1

1

Pl = σ Arx2 Px = σ Arx2

(4.13)

(4.14)

4.3 Hybrid Modeling for Jet Impact Characteristics

77

where De is the diameter of the jet exit, mm; Dx is the jet diameter at the molten bath surface, mm, and it can be calculated by Eq. (4.15).

Dx =

ρe ρx

 21

 ve De vx

(4.15)

In this section, the k value was defined and expressed by the following equation.

k=

vx ve



H De

 (4.16)

As a result, the following equation can be obtained combing Eqs. (4.14)–(4.16). 1

h l = σ Arx2



 k De 2 ρslag ρe ve2 ρx − hg ρsteel g ρe H ρsteel

(4.17)

According to a previous report [31], the following expression shows the mathematical formula for Ar x , which indicates the ratio of the liquid buoyancy to the jet inertia force. Arx =

1  ρsteel g Dx ρsteel g ρe 2 H = ρx vx2 ρx vx2 ρx k

(4.18)

Ultimately, hl can be obtained by solving equations containing Eqs. (4.17) and (4.18). 1

hl = σ

(ρe ρx ) 4 ve De (ρsteel g) 1

hp = hl + hg = σ

1 2

(ρe ρx ) 4 ve De (ρsteel g)

1 2





k H0 − h g

k H0 − h g

 21

 21

−

+

ρslag hg ρsteel

ρsteel − ρslag hg ρsteel

(4.19)

(4.20)

Therefore, the theoretical model for calculating the jet penetration depth hp was obtained. It can be seen that the k value and lance height H 0 are the key factors affecting the jet penetration depth for a given oxygen lance. The k value is analyzed systematically in Sect. 4.4.1. To improve the accuracy of the theoretical model, it is of great importance to obtain the exact values of k, ve , ρ x , and ρ e .

4.3.3 Theoretical Modeling of the Jet Impact Zone Volume Based on the previous research reported by Birkhoff and Zarantonello [32], the longitudinal profile of the impinging cavity was assumed as a paraboloid. Figure 4.9 shows

78

4 Impact Characteristics of Coherent Supersonic Jet

z

Fig. 4.9 Diagram of the paraboloid for volume integral

x y

a diagram of the paraboloid and the parametric equation of a revolving paraboloid is as follows. x 2 + y 2 = 2 pz

(4.21)

where p is the characteristic parameter of a paraboloid. Therefore, the volume of the impact zone V impact-zone , surface area of the impact zone S, and cross-sectional area of the impact zone at the molten bath surface Ax can be obtained by integration, according to Fig. 4.1. h∫l +h g

∫h l Vimpact−zone =

2π pzdz + 0

S=

2π pzdz

(4.22)

hl

  3 2 1 2 pp 2 p + 2 h l + h g 2 − pp 2 3 3   Ax = 2π p h g + h l

(4.23) (4.24)

The characteristic parameter p is the key to the V impact-zone calculation. Based on the shape of the jet impinging cavity, as shown in Fig. 4.1, the pressure of the jet centerline at the bottom gas–liquid interface of the impinging cavity Pl is equal to the static pressure of the molten bath according to the principle of energy conservation. This is shown in the following equation, which is the integral form of Eq. (4.8).

4.3 Hybrid Modeling for Jet Impact Characteristics

∫

∫

1 ρg v 2 ds = 2

s

79

∫ ρsteel gdVsteel +

Vsteel



+ πρslag gph g h g + 2h l

ρslag gdVslag = πρsteel gph 2l Vslag



(4.25)

where s is the superficial area of the impact zone, m2 ; V steel and V slag are, respectively, the volumes of the molten steel and liquid slag parts in the impact zone. According to Eqs. (4.8)–(4.12) and (4.26) can be obtained. In addition, Eq. (4.27) can be built by combing Eqs. (4.23), (4.24), and (4.26). Equation (4.29) can be obtained according to the momentum conservation of the oxygen jet. ∫

1 1 σ Arx2 rx v 2 ds = 2

∫

s

1 ρg v 2 ds 2

(4.26)

s Ax

(4.27)

s

∫ s

∫ ∫

Ax

1 σ 2 1 σ 2

1

Arx2 rx v 2 ds 1 2

Arx rx v 2 dAx

=

1 1 nρx v 2 dAx = nρe ve2 Ae 2 2

(4.28)

Ax

Therefore, the characteristic parameter p can be calculated as follows. p=

2h p 6πh p (ρsteel g(h p −h g )2 −ρslag gh 2g +2ρslag gh g h p ) 1

σ Arx2 ρe ve2 Ae

23 +1 −1

(4.29)

Actually, the horizontal shear stress produced during the interaction between the supersonic jet and the molten bath has a significant influence on the impinging cavity. Hence, the characteristic parameter p should be modified. According to a previous study [33], the axial jet free distance from the nozzle exit to the molten bath surface affects the shape of the jet impinging cavity and the characteristic parameter p was modified as follows: p=

λ (H0 − h g )

3 2



2h p 6πh p (ρsteel g(h p −h g )2 −ρslag gh 2g +2ρslag gh g h p ) 1

σ Arx2 ρe ve2 Ae

+1

23

(4.30) −1

where λ is the coefficient of correction. The jet impact zone volume V impact-zone can then be obtained as follows:

80

4 Impact Characteristics of Coherent Supersonic Jet

Fig. 4.10 Flow chart of the hybrid model for jet impact characteristics

Vimpact−zone =

λ 3

(H0 − h g ) 2



2πh 3p 6πh p (ρsteel g(h p −h g )2 −ρslag gh 2g +2ρslag gh g h p ) 1

σ Arx2 ρe ve2 Ae

+1

23

(4.31) −1

4.3.4 Hybrid Modeling A hybrid model was established to describe the impact characteristics of the coherent and conventional supersonic jets. As shown in Fig. 4.10, by combing the theoretical model and numerical simulations, the hybrid model was used to calculate the jet penetration depth hp and jet impact zone volume V impact-zone . The key parameters, k value, ve , ρ x , and ρ e , can be calculated and outputted by the CFD model and the userdefined function (labeled “UDF program” in Fig. 4.10), which can simulate the fluid flow characteristics of the coherent and conventional supersonic jets, and process the related data in the procedure. It can be seen that these four key parameters can be used by the theoretical model to calculate the jet penetration depth hp . Then, hp , k value, ve , ρ x and ρ e can be used to calculate the jet impact zone volume V impact-zone . Therefore, it is of great importance for the V impact-zone calculation that hp be obtained accurately. In Sect. 4.4.2, the theoretical model for hp is modified to obtain more accurate value.

4.4 Results and Discussion 4.4.1 Analysis of the k Value As a newly defined variable, k value is a key parameter to depict the jet impact characteristics. As expressed by Eq. (4.16), k value is the product of a dimensionless quantity of velocity and a dimensionless quantity of free distance. The dimensionless

4.4 Results and Discussion

81

quantity of velocity (vx /ve ) reflects the jet velocity variation with respect to the jet exit velocity at the jet free distance (H/De ), which is also a dimensionless quantity. vx and H have decisive effect on the k value calculation, whereas ve and De are constants that depend on the design parameters of the Laval nozzle under standard conditions. Moreover, the second derivative of k value is equal to the first derivative of vx and this means that the change trend of the scope of k value reflects the variation tendency of vx . Comparing with another approach that uses vx directly to analyze the jet impact characteristics by other researchers [5, 6], these dimensionless numbers in the modeling process can improve the generalization ability of the mathematical formula. Figure 4.11 shows the k value distributions of the coherent and conventional supersonic jets. The k value increases and then declines after exiting the Laval nozzle for both cases. However, there are obvious differences in the specific variation tendency between these two types of jet. For the coherent supersonic jet, the k value increases in proportion to the axial free distance from the nozzle in the region 0 m ≤ H ≤ 1.5 m, and then declines with increasing axial free distance. For the conventional supersonic jet, the k value shows the same variation tendency as for the coherent supersonic jet in the region 0 m ≤ H ≤ 0.5 m, then remains stable in the region 0.5 m ≤ H ≤ 2.0 m and finally declines. It is obvious that the k value of the coherent supersonic jet is more than that of the conventional supersonic jet when H ≥ 0.5 m. For both cases, H is the common dependent variable for the k value calculation. Hence, the k value was studied through analyzing the change in vx . Figure 4.12 shows the axial velocity distributions of the coherent and conventional supersonic jets at the jet centerline. After the exit from the Laval nozzle, the phenomenon of repeated fluctuation of the axial velocity is presented in both cases. This occurs as a result of the incorrect expansion of the supersonic jet. After repeated fluctuations of the supersonic oxygen jet, the axial velocity of the jet reaches a steady Fig. 4.11 k value distributions of the coherent and conventional supersonic jets

82

4 Impact Characteristics of Coherent Supersonic Jet

Fig. 4.12 Axial velocity distributions of the coherent and conventional supersonic jets

state and the potential core forms. The potential core length of the coherent supersonic jet is more than three times larger than that of the conventional supersonic jet. The coherent supersonic jet remains coherent up to 1.5 m before the velocity starts to decrease, whereas the velocity of the conventional supersonic jet decreases gradually after 0.5 m from the nozzle exit plane. The shrouding flow injection and the subsequent combustion of oxygen and CH4 account for the attenuation characteristics of the main supersonic oxygen jet. Figures 4.13a, b show the static temperature and turbulent kinetic energy distributions of the coherent and conventional supersonic jets in the longitudinal section. Obviously, the high-temperature zone, which is the red part in Fig. 4.13I, formed by the shrouding flame shows characteristics of low density. The high temperature reduces the density of the gases surrounding the main supersonic oxygen jet, which in turn reduces the turbulent mixing rate between the main supersonic oxygen jet and the surroundings, as reported by Alam [22]. It can be concluded that the shrouding combustion flame can create a “protective case” of high temperature and low density around the main supersonic oxygen jet. As a result, the potential core length of the jet was prolonged. Above all, the k value change reflects the potential core length of the two kinds of supersonic jet to some extent. It is the shrouding combustion flame that delays the velocity attenuation of the main supersonic jet and extends the potential core length.

4.4.2 Modification of hp Calculation Table 4.3 compares the values of hp calculated by the theoretical model with those obtained from the numerical simulations. For the coherent supersonic jet, the penetration depths obtained by the theoretical model are very close to the CFD results when the lance height H 0 is no more than 2.15 m and the maximum error is less than

4.4 Results and Discussion

83

Fig. 4.13 Static temperature distributions (a) and turbulent kinetic energy distributions (b) of the coherent supersonic jet; I and the conventional supersonic jet; II on longitudinal section

3.92 pct. However, for the lance heights H 0 of 2.65 m and 3.15 m, large deviations between the theoretical model results and the CFD results appear, showing errors of 11.90 pct and 25.79 pct, respectively. Similar trends are found for the traditional supersonic jet; the results of the theoretical model are in good agreements with the CFD results when H 0 is 0.90 m and 1.15 m. Beyond those heights, significant differences between the results arise when H 0 > 1.65 m, and the error is up to 34.16 pct when H 0 = 3.15 m. Overall, it can be concluded that the theoretical model can predict the penetration depth of the supersonic jet well when the lance height is in a certain range, but does not apply to situations in which the oxygen lance height is beyond the range. The difference between the theoretical model results and the CFD results may arise from the uncertainties involved during the theoretical modeling process. In fact, both

84

4 Impact Characteristics of Coherent Supersonic Jet

Table 4.3 Values of hp (penetration depth in the molten steel) at different lance height, m Lance height H 0 (m)

0.90

1.15

1.65

2.15

2.65

3.15

Coherent supersonic jet

CFD result

0.754

0.723

0.546

0.408

0.294

0.221

Theoretical model

0.761

0.749

0.526

0.392

0.329

0.278

Error (%)

0.93

3.60

3.66

3.92

CFD result

0.513

0.434

0.321

0.232

0.179

0.161

Theoretical model

0.531

0.451

0.359

0.298

0.234

0.216

Error (%)

3.51

3.92

Conventional supersonic jet

11.84

28.45

11.90

30.73

25.79

34.16

the molten steel density ρ steel and the molten slag density ρ slag have certain impacts on the buoyancy term of the Archimedes number Ar x . However, the value of Ar x was magnified in the theoretical model, because only the molten steel density ρ steel was adopted to calculate the buoyancy term of the Archimedes number Ar x . Deeper impact cavities can be formed by the high-speed oxygen jet at lower lance positions. In these cases, the molten slag density plays a marginal role in the calculation of Ar x , considering that the molten slag density is only 41.67 pct of the molten steel density, as listed in Table 4.2. Therefore, the results of the theoretical model agree reasonably with the CFD results. However, when the lance position is higher, the impact cavities are obviously shallower and the penetration depth in the molten steel is even less than the thickness of the molten slag layer. In these cases, the molten slag exerts a tremendous influence on the buoyancy term of Ar x . Consequently, the accuracy of the theoretical model will be lower at higher lance positions, simply because the effect of the molten slag density was neglected in the theoretical modeling. Above all, when the lance position is higher, the impact cavity is shallower and the influence of the molten slag density on the buoyancy term of Ar x should be considered. Based on the theoretical model, two modified models for predicting the penetration depth of the coherent and conventional supersonic jets were obtained. The characteristic density ρ C was employed to calculate the buoyancy term of Ar x during the modeling process and the modified models of the supersonic jet penetration depth are presented by the following equations. In modified model 1, the characteristic density ρ C is the arithmetic mean of ρ steel and ρ slag , whereas ρ C is equal to ρ slag in modified model 2. 1  2

 21 ρC ρe ρx 4 ve De ρsteel − ρslag k hp = σ  + hg  21 H − h ρsteel 2 0 g ρsteel g ⎧ Theoretical model ⎨ ρsteel slag ρC = ρsteel +ρ Modified model 1 2 ⎩ Modified model 2 ρslag

(4.32)

(4.33)

4.4 Results and Discussion

85

Fig. 4.14 Penetration depth of the coherent supersonic jet at different lance heights

Figure 4.14 depicts the penetration depth of the coherent supersonic jet for different lance heights obtained from the theoretical model, modified model 1, modified model 2, and the numerical simulations. The CFD results coincide well with the results calculated by the theoretical model and its modifications. It can be concluded that the penetration depth of the coherent supersonic jet can be predicted well by the comprehensive calculation of the theoretical model and the two modified models. For the coherent supersonic jet examined in this section, the theoretical model, modified model 1, and modified model 2 can be, respectively, used to calculate the jet penetration depth when H 0 < 2.0 m, 2.0 m < H 0 < 3.0 m, and H 0 > 3.0 m. Similar situations are presented in Fig. 4.15 and the penetration depth of the conventional supersonic jet can be also well predicted by the three models. For the conventional supersonic jet in this section, the theoretical model, modified model 1, and modified model 2 can be, respectively, applied to calculate the jet penetration depth when H 0 < 1.5 m, 1.5 m < H 0 < 2.0 m, and H 0 > 2.0 m. Compared with the CFD results, the error is no more than 3.92 pct for both the coherent supersonic jet and the conventional supersonic jet. It can be concluded that the supersonic jet penetration depth can be well predicted by combining the theoretical model and the two modified models.

4.4.3 Validation of Vimpact-zone Calculation In the hybrid model, the coefficient of correction λ was calculated based on the values of V impact-zone obtained by the numerical simulations, yielding values of 20.4 and 5.9 for the coherent and conventional supersonic jets, respectively. Table 4.4 lists the values of V impact-zone obtained by the CFD and hybrid models. It can be seen that the CFD results show reasonably good agreement with the hybrid model, with a maximum error of 3.62 pct for the coherent supersonic jet.

86

4 Impact Characteristics of Coherent Supersonic Jet

Fig. 4.15 Penetration depth of the conventional supersonic jet at different lance heights

Table 4.4 Values of V impact-zone calculated by the CFD model and the hybrid model (m3 ) H 0 (m) Coherent supersonic jet

Conventional supersonic jet

0.9

1.15

1.65

2.15

2.65

3.15

CFD model

0.217

0.142

0.147

0.153

0.138

0.131

Hybrid model

0.216

0.137

0.143

0.151

0.133

0.130

Error (%)

0.46

3.52

2.72

1.31

3.62

0.76

CFD model

0.079

0.092

0.103

0.071

0.055

0.032

Hybrid model

0.083

0.096

0.101

0.063

0.046

0.025

Error (%)

5.06

4.34

1.94

7.04

7.27

9.37

Additional simulation experiments were carried out to validate the reliability of the obtained coefficient of correction λ. Table 4.5 compares the values of V impact-zone calculated by the CFD and hybrid models at H 0 = 1.4 m and H 0 = 1.9 m. The comparison between the CFD and hybrid models shows reasonably good agreement between the simulation results and computing results. The maximum errors are 3.45 pct and 5.62 pct for the coherent and conventional supersonic jets. It can be concluded that the hybrid model with the coefficient of correction is reasonable and reliable for predicting the volume of the impact zone generated by these two supersonic jets. Table 4.5 Additional values of V impact-zone calculated by the CFD model and the hybrid model (m3 ) Items

Coherent supersonic jet

Conventional supersonic jet

H 0 (m)

1.4

1.9

1.4

1.9

CFD model

0.145

0.149

0.097

0.089

Hybrid model

0.140

0.144

0.093

0.084

Error (%)

3.45

3.35

4.12

5.62

4.4 Results and Discussion

87

4.4.4 Analysis of the Jet Penetration Depth Figure 4.16 depicts the penetration depth change in the coherent and conventional supersonic jets at different lance heights, as calculated by the numerical simulations. As Fig. 4.16 shows, the effect of the lance height on the jet penetration depth is significant, whereby the penetration depth decreases with increasing lance height. Meanwhile, it can be seen that the penetration depth of the coherent supersonic jet is larger than that of the conventional supersonic jet for the same lance height. This further illustrates the advantages of the coherent supersonic jet in delivering large amounts of oxygen to the liquid melt with a better stirring effect compared to the conventional supersonic jet. Moreover, for the coherent supersonic jet, the slope of the linear regression is 0.124 when the lance height H 0 is in the range of 0.90–1.15 m, whereas that of the conventional supersonic jet is 0.316. The main reason accounting for this phenomenon is that the potential core length of the coherent supersonic jet is up to 1.5 m and that of the conventional supersonic jet is only 0.5 m. The reasons were thoroughly analyzed in Sect. 4.4.1, and it is clear from Fig. 4.12 that the axial velocity of the conventional supersonic jet decays faster than that of the coherent supersonic jet when the axial distance is in the range 0.9–1.15 m. Figure 4.17a, b show the contours of the impingent zone in the molten steel caused by the coherent and conventional supersonic jets at different lance heights, respectively. It is clear that the coherent supersonic jet can generate a deeper impact cavity than the conventional supersonic jet at the same lance height. As previously reported [8], there are three impingement modes resulting from the jetting of gas onto a liquid surface, which are the penetrating mode, splashing mode, and dimpling mode. The contour of the impingement cavity gradually changes from the penetrating mode to the splashing mode, and then to the dimpling mode as the lance height increases. Fig. 4.16 CFD results of the penetration depth at different lance heights

88

4 Impact Characteristics of Coherent Supersonic Jet

Fig. 4.17 Contours of the impact zone in the molten steel. a Coherent supersonic jet; b conventional supersonic jet

4.4.5 Analysis of the Jet Impact Zone Volume In the present study, the influence of the lance height on the volume of the impact zone generated by the coherent and conventional supersonic jets was also studied. As Fig. 4.17 shows, the coherent supersonic jet can generate a deeper impact cavity than the conventional supersonic jet for the same lance height. Figure 4.18 shows the impact zone volume change at different lance heights for these two cases, as calculated by the numerical simulations. It can be seen that the effect of the lance height on V impact-zone is significant, and that the impact zone volumes generated by the coherent supersonic jet are larger than those by the conventional supersonic jet for the same lance height. This further illustrates the advantages of the coherent supersonic jet in delivering large amounts of oxygen to the liquid melt, with a better stirring effect compared with the conventional supersonic jet. For these two supersonic jets, the jet impact zone volume V impact-zone initially increases and then decreases with increasing lance height. In addition, V impact-zone is maximum when H 0 = 900 mm, because the lance is at a low position and the supersonic jet works in the penetrating mode. The change tendency of V impact-zone with H 0 is consistent with the change tendency of the impact zone superficial area.

4.5 Conclusions By integrating theoretical models and numerical simulations, a hybrid model was established and modified to calculate the penetration depth and impact zone volume of coherent and conventional supersonic jets. Water model experiments were carried out to validate the numerical simulations. This section compared the impact characteristics of the coherent and conventional supersonic jets, and investigated the

4.5 Conclusions

89

Fig. 4.18 CFD results of the impact zone volume at different lance heights

influence of the lance height on the jet penetration depth and jet impact zone volume. The conclusions are as follows: 1. The applicability of the numerical simulation model is validated by the reasonably good agreement between the numerical simulation results and the measurements based on the water model experiments. By combining the theoretical model and the CFD model, a hybrid model and its modifications can well predict the penetration depth and impact zone volume of the coherent and conventional supersonic jets. 2. Compared with the conventional supersonic jet, the shrouding combustion flame of the coherent supersonic jet delays the velocity attenuation of the main supersonic jet and extends the potential core length, which accounts for the k value change. 3. The lance height exerts significant influence on the jet penetration depth and jet impact zone volume. The penetration depth hl decreases with increasing lance height, whereas the jet impact zone volume V impact-zone initially increases and then decreases with increasing lance height, which is consistent with the change tendency of the jet impact zone superficial area. In addition, the penetration depth and impact zone volume of the coherent supersonic jet are larger than those of the conventional supersonic jet for the same lance height, which illustrates the advantages of the coherent supersonic jet in delivering large amounts of oxygen to the liquid melt with a better stirring effect compared to the conventional supersonic jet.

90

4 Impact Characteristics of Coherent Supersonic Jet

References 1. Lee B, Sohn I (2014) Review of innovative energy savings technology for the electric arc furnace. JOM 66(9):1581–1594 2. Wei G, Zhu R, Dong K, Ma G, Cheng T (2016) Research and analysis on the physical and chemical properties of molten bath with bottom-blowing in EAF steelmaking process. Metall Mater Trans B 47(5):3066–3079 3. Alam M, Naser J, Brooks G, Fontana A (2010) Computational fluid dynamics modeling of supersonic coherent jets for electric arc furnace steelmaking process. Metall Mater Trans B 41(6):1354–1367 4. Wei G, Zhu R, Cheng T, Zhao F (2016) Numerical simulation of jet behavior and impingement characteristics of preheating shrouded supersonic jets. J Iron Steel Res Int 23(10):997–1006 5. Alam M, Naser J, Brooks G et al (2012) A computational fluid dynamics model of shrouded supersonic jet impingement on a water surface. ISIJ Int 52(6):1026–1035 6. Li Q, Li M, Kuang S et al (2015) Numerical simulation of the interaction between supersonic oxygen jets and molten slag-metal bath in steelmaking BOF process. Metall Mater Trans B 46(3):1494–1509 7. Molloy NA (1970) Impinging jet flow in a two-phase system: the basic flow pattern. Iron Steel Inst 226(3):943–950 8. Collins R, Lubanska H (2002) The depression of liquid surfaces by gas jets. Br J Appl Phys 5(1):22–26 9. Nordquist A, Kumbhat N, Jonsson L et al (2006) The effect of nozzle diameter, lance height and flow rate on penetration depth in a top-blown water model. Steel Res Int 77(2):82–90 10. Li M, Li Q, Kuang S, Zou Z (2016) Determination of cavity dimensions induced by impingement of gas jets onto a liquid bath. Metall Mater Trans B 47(3):116–126 11. Li M, Li Q, Kuang S, Zou Z (2016) Transferring characteristics of momentum/energy during oxygen jetting into the molten bath in BOFs: a computational exploration. Steel Res Int 87(3):288–300 12. Rout B, Brooks G, Subagyo, M Rhamdhani, Z Li (2016) Modeling of droplet generation in a top blowing steelmaking process. Metall Mater Trans B 47(3):3350–3561 13. Sabah S, Brooks G (2015) Splash distribution in oxygen steelmaking. Metall Mater Trans B 46(3):863–872 14. Sabah S, Brooks G (2016) Energy balance around gas injection into oxygen steelmaking. Metall Mater Trans B 47(3):458–466 15. Li R, Wraith AE, Harris R (1994) Gas dispersion phenomena at a narrow slot submerged in a liquid. Chem Eng Sci 49(4):531–540 16. Wraith AE, Li R, Harris R (1995) Gas bubble volume at a narrow slot nozzle in a liquid. Chem Eng Sci 50(3):1057–1058 17. Ersson M, Tilliander A, Jonsson L et al (2008) A mathematical model of an impinging air jet on a water surface. Trans Iron Steel Inst Jpn 48(4):377–384 18. Muñoz-Esparza D, Buchlin JM, Myrillas K et al (2012) Numerical investigation of impinging gas jets onto deformable liquid layers. Appl Math Model 36(6):2687–2700 19. Alam M, Irons G, Brooks G et al (2011) Inclined jetting and splashing in electric arc furnace steelmaking. ISIJ Int 51(9):1439–1447 20. Solórzano-López J, Zenit R, Ramírez-Argáez MA (2011) Mathematical and physical simulation of the interaction between a gas jet and a liquid free surface. Appl Math Model 35(10):4991– 5005 21. Lee M, Whitney V, Molloy N (2010) Jet-liquid interaction in a steelmaking electric arc furnace. Scand J Metall 30(30):330–336 22. Alam M, Naser J, Brooks G (2010) Computational fluid dynamics simulation of supersonic oxygen jet behavior at steelmaking temperature. Metall Mater Trans B 41(3):636–645 23. Sumi I, Kishimoto Y, Kikuchi Y, Igarashi H (2006) Effect of high-temperature field on supersonic oxygen jet behavior. ISIJ Int 46(9):1312–1317

References

91

24. Liu F, Zhu R, Dong K, Hu S (2016) Effect of ambient and oxygen temperature on flow field characteristics of coherent jet. Metall Mater Trans B 47(1):1–16 25. Hirt C, Nichols B (1981) Volume of fluid (VOF) method for the dynamics of free boundaries. J Comput Phys 39(1):201–225 26. Launder B (1972) Spalding: lectures in mathematical model of turbulence. Academic Press, London, pp 124–129 27. Li Q, Li M, Kuang S, Zou Z (2014) Computational study on the behaviours of supersonic jets and their impingement onto molten liquid free surface in BOF steelmaking. Can Metall Q 53(3):340–351 28. Bank B, Chandrasekhara D (1963) Experimental investigation of the penetration of a highvelocity gas jet through a liquid surface. J Fluid Mech 15:13–33 29. Mazumdar D, Guthrie RIL (1995) The physical and mathematical modelling of gas stirred ladle systems. ISIJ Int 35(1):1–20 30. Crowe C, Elger D, Roberson J (2004) Eng Fluid Mech 32(4):124–136 31. Melton L, Malone W (1964) Fluid mechanics research and engineering application in nonNewtonian fluid systems. J Petrol Technol 4(1):56–66 32. Birkhoff G, Zarantonello E (1957) Jets, wakes and cavities. AcademicPress, New York, pp 1–6 33. Cheng T (2016) Fundamental study on flow flied characteristics and impact law of coherent jet. University of Science and Technology Beijing, Beijing, pp 70–78

Chapter 5

Modeling and Arrangement of Submerged Nozzles

5.1 Introduction The molten bath of EAF with eccentric bottom tapping (EBT) is flat and the molten bath stirring intensity is poor due to the limited oxygen supply intensity [1–4]. It is important for EAF steelmaking to strengthen the circulation in the liquid bath in order to enhance the smelting reaction rate [5–8]. A series of studies have been conducted to investigate liquid flow pattern in the EAF. Liu et al. [9] analyzed the effect of bottom-blowing arrangement modes on fluid flow in the EAF using a computational fluid dynamics (CFD) model. Ramirez et al. [10] analyzed the combined effect of electromagnetic forces and bottom-blowing on the fluid flow of a direct current (DC) EAF molten bath by a numerical model. Alam et al. [11] studied the mechanism of lance angle on the critical penetration depth in EAF steelmaking by water model experiments. Ma et al. [12] measured the penetration depth of a submerged slide-blown jet based on a water model and a Matlab digital image processing algorithm. Zhan et al. [13] analyzed the flow field distribution in a liquid bath side-blown in an EAF through a numerical model. Cafeery et al. [14] developed a numerical model to investigate the influence of bubbles emerging from bottom tuyeres on the heat transfer in an EAF molten bath. Li [15] studied the effect of radial distance of bottom-blowing nozzles on the mixing time of an EAF liquid bath experimentally and numerically. In Chap. 4, the impact characteristics of single submerged gas injection was analyzed systematically [8]. However, as a new technology, there are very few studies about the EAF liquid bath flow characteristics with two submerged nozzles. Hence, the EAF liquid bath flowing characteristics with submerged gas injection is required to be studied to optimize the process parameters of EAF steelmaking before this new method is put into industrial application. In this chapter, the EAF liquid bath flowing characteristics under submerged gas injection were investigated by the methods of water modelling and numerical modelling. The validity of the numerical model of EAF with submerged gas injection was first examined using the measured data. The effect of the horizontal arrangement mode, gas flow rate, vertical dip angle and submerged depth of the nozzle on the liquid © Metallurgical Industry Press 2024 G. Wei and R. Zhu, Electric Arc Furnace Steelmaking with Submerged Mixed Injection, https://doi.org/10.1007/978-981-99-4602-0_5

93

94

5 Modeling and Arrangement of Submerged Nozzles

bath flow in the EAF was analyzed by orthogonal experiments. The molten bath flow field with different horizontal arrangement modes of two submerged nozzles were also monitored and analyzed. In addition, the influence of the horizontal arrangement mode was also studied by industrial-scale experiments.

5.2 Water Model Experiment An EAF physical model was built by scaling down the prototype model in 1:4 based on the similarity law of dynamics and geometry. Figure 5.1 shows the relative physical model parameters. In this study, water was used to represent molten steel and compressed air was used for oxygen. Equations (5.1)–(5.3) list the equations for modified Froude number calculation, which was adopted to ensure the dynamic similarity [16, 17]. Tables 5.1 and 5.2, respectively, lists the relative parameters in this study. ρg ρg1 u2 u2 × = 1 × gd ρl − ρg gd1 ρl1 − ρg1 /  / ρl − ρg ρg1 ρl − ρg ρg1 d 5 Q = × × = M5 × × Q1 d1 ρl1 − ρg1 ρl1 ρl1 − ρg1 ρl1   A B π D21 × 360 + D22 × 360 + 2 × L1 d= π Fr , = Fr1, , that is,

(5.1)

(5.2)

(5.3)

where, d and d 1 are the diameter of jet nozzles of prototype and physical model respectively and can be calculated by Eq. (5.3); ρ l and ρ l1 are the liquid density of prototype and physical model; Q and Q1 are the gas flows of prototype and physical model; ρ g and ρ g1 are the gas density of prototype and physical model; u and u1 are, the gas velocities of prototype and physical model; g is the gravity acceleration and M is the geometric similarity ratio. In addition, α is the vertical dip angle of the submerged nozzle and Ds is the submerged depth of the submerged nozzle, which represents the distance between the submerged nozzle exit and the molten steel level. Figure 5.2 shows a schematic of the water model, which includes a plexiglass model, air source system, bus-bars, submerged nozzles, valves, conductometers (DJS-0.1), DJ800 data acquisition system and a computer. In the present experiment, the compressed air was injected into the water bath through the submerged nozzles. To measure the mixing time, the conductivity electrodes were adopted to detect the change in the tracer element concentration in the liquid bath and, as Fig. 5.2 shows, two conductivity electrodes were installed at Point A and B in the water model. Point A and Point B were located at the bottom of the liquid bath and they are far away from the submerged nozzles. Therefore, the measured values at Point A and Point B can reflect the mixed intensity of the liquid bath.

5.2 Water Model Experiment

95

Fig. 5.1 EAF geometry and submerged nozzle installation Table 5.1 Relative parameters of the geometry model

Items

Water model

EAF prototype

D1 (mm)

312

1248

D2 (mm)

1175

3700

H 1 (mm)

425

1700

H 2 (mm)

313

1252

H 3 (mm)

150

600

L (mm)

611

5774

L 1 (mm)

611

2444

A (°)

233

233

B (°)

110

110

Submerged nozzle diameter (mm)

3

Depth of molten steel (mm)

300

12 1200

Thickness of slag layer (mm) – Depth of molten bath (mm) Table 5.2 Gas flow rate of single submerged nozzle

150

300

1350

Items

Gas flow rate (Nm3 /h)

EAF prototype

200

400

600

Water model

2.5

5.0

7.5

96

5 Modeling and Arrangement of Submerged Nozzles

Fig. 5.2 Water model experiment platform

In addition, the tracer used in this study was KCl solution that changed the electronic conductivity of the water bath. The time from the KCl solution injection to the time when the tracer concentration at the monitor points reached 99% of the mean tracer concentration in the bath was considered as the mixing time. In this study, the mixing time was obtained by the following equation. Tmix =

TA + TB 2

(5.4)

where, T mix is the mixing time of the complete liquid bath; T A and T B are, respectively, the mixing time measured by the conductivity electrodes at Point A and Point B. Figure 5.3 shows the different horizontal arrangement modes of the two submerged nozzles in an EAF, which are Mode A, Mode B and Mode C, respectively. In Mode A, the central axes of the two submerged nozzles point to the EAF center. In Mode B, both of the central axes of the two submerged nozzles rotate 20° in clockwise and in Mode C, the central axes of the two submerged nozzles rotate 20° in the opposite directions. In the present study, an orthogonal test scheme [18] was designed to investigate the influence of the horizontal arrangement mode, gas flow rate, vertical dip angle and submerged depth of the nozzle on the flow field of the EAF liquid bath. As listed in Table 5.3, the experiment was designed according to four factors and the three levels orthogonal table.

5.3 Numerical Modeling

97

Fig. 5.3 Horizontal arrangement modes of submerged nozzles in an EAF

Table 5.3 Orthogonal test scheme in the present water model experiment Item

Factors Gas flow rate (Nm3 / h)

Horizontal arrangement mode

1

2.5

A

5

30

2

2.5

B

10

45

3

2.5

C

15

60

4

5.0

A

15

45

5

5.0

B

5

60

6

5.0

C

10

30

7

7.5

A

10

60

8

7.5

B

15

30

9

7.5

C

5

45

Vertical dip angle (°)

Submerged depth (mm)

5.3 Numerical Modeling A numerical model of the EAF with submerged gas injection was developed using the volume of fluid (VOF) approach [19].

5.3.1 Assumptions 1. The gas phase was ideal gas and the liquids were incompressible Newtonian fluids. 2. The flow of liquid and gas (molten slag, molten steel and oxygen) were threedimensional, non-isothermal and transient.

98

5 Modeling and Arrangement of Submerged Nozzles

3. A non-slip condition was applied to all of walls and Viscosity and surface tension were assumed constant. 4. The effect of the chemical reactions and the electromagnetics forces was not taken into consideration in the numerical simulation.

5.3.2 Governing Equations For an element system consisting of n phases, the total sum of each phase volume fraction is 1, just as Eq. (5.5) shows. n ∑

αi = 1

(5.5)

i=1

where, α i is the i-th phase volume fraction in an element system. In addition, the effective specific heat cp and the effective density ρ in the element system are also obtained using the volume average method in VOF model. The effective density is calculated as: ρ=

n ∑

αi ρi

(5.6)

i=1

Accordingly, for a three-phase system which includes molten steel, molten slag and gas phase, its effective density of the fluid can be obtained using Eq. (5.7). The same computing method is used for calculating other physical properties, such as cp . ρ = αgas ρgas + αslag ρslag + αmetal ρmetal

(5.7)

The continuity equation:  n ∑   ∂αi 1 + ∇ · (αi νi ) = m ji − m i j Sαi + ∂t ρi i=1

(5.8)

where, mij and mji are, respectively, the mass flowing from the i-th phase to j-th phase and that from the j-th phase to i-th phase; vi is the velocity components in the direction i; S αi is the custom source and t is the time. The momentum equation: 

 ∂ (ρ v→) + ∇ · (ρ v→v→) = −∇ p + ∇ · μ ∇ v→ + ∇ v→T + ρ g→ + F ∂t

(5.9)

where, v is the fluid instantaneous velocity; μ is the dynamic viscosity; p is the static pressure; g is the gravity force; and F is the other outside volume force.

5.3 Numerical Modeling

99

In this study, the standard k-ε turbulence model was adopted. The turbulence kinetic energy and the dissipation rate can be calculated by Eqs. (5.10) and (5.11) [20, 21]. The turbulence kinetic energy k (m2 /s2 ) and dissipation rate ε (m2 /s3 ) are, respectively, determined by Eqs. (5.10) and (5.11) [20, 21].  

μt ∂k ∂(ρk) ∂(ρkvi ) ∂ μ+ · + G k + G b − ρε − Y M + = (5.10) ∂t ∂ xi ∂x j σk ∂ xi  

μt ∂ε ∂(ρε) ∂(ρεvi ) ε ∂ ε2 μ+ · + C1ε (G k + C3ε G b ) − C2ε ρ + = ∂t ∂ xi ∂x j σε ∂ xi k k (5.11) where, k is the turbulence kinetic energy; ε is the dissipation rate; Gk and Gb are, respectively, the turbulent kinetic energy generated by average velocity and buoyancy; μt is the turbulent viscosity; Y M is the turbulent dissipation rate generated by compressible turbulence pulsation. And the value of σ k , σ ε , C 1ε , C 2ε and C 3ε are, respectively, 0.9, 0.009, 1.44, 1.92 and 0.8. The energy equation: ∂(ρ E) + ∇ · [u(ρ E + p)] = ∇ · (keff ∇T ) + Sh ∂t

(5.12)

The energy E can be calculated by Eq. (5.13). ∑n i=1 αi ρi E i E= ∑ n i=1 αi ρi

(5.13)

5.3.3 Simulation Details An industrial 75 t EAF was used to study the effect of submerged gas injection on liquid flow in the EAF. In this study, the numerical model was developed based on the prototype of 75 t EAF. Figure 5.4 depicts the geometric model of the EAF with two submerged nozzles (A and B in Fig. 5.4) and the relative detailed grid arrangements. Tables 5.1 and 5.4 list the specific parameters of the EAF geometry model and the relative physical properties of the fluids. In the present study, three horizontal arrangement modes of two submerged nozzles, including Mode A, Mode B and Mode C, were calculated with the oxygen supplying rate being 600 Nm3 /h, the vertical dip angle being 10° and the submerged depth being 240 mm. The inlet of the submerged nozzles was set as mass-flow inlet. The outlet of the numerical model was set as pressure boundary condition and its value is 101,325 Pa. Using the commercial CFD software ANSYS Fluent version

100

5 Modeling and Arrangement of Submerged Nozzles

Fig. 5.4 Geometric model and detailed grid arrangements. a Geometric model; b detailed grid arrangements. In both (a) and (b), A and B are the two submerged nozzles Table 5.4 The relative fluid physical properties Items

Molten steel

Liquid slag

Water

Oxygen

Air 1.789 × 10–5

Viscosity (kg/(m s))

0.0065

0.35

0.001

1.92 ×

Density (kg/m3 )

7200

3000

1000

1.29

1.25

Thermal conductivity (W/ (m K))

15

1.2

0.6

0.0332

0.0242

Specific heat (J/(kg K))

670

1200

4182

2222

1006.43

Surface tension (N/m)

1.6

0.4

0.071

–

–

Temperature (K)

1873

1873

300

300

300

10–5

5.4 Results and Discussion

101

14.0.0, a pressure-based solver was applied to conduct the calculation using a transient solution model to simulate the liquid flow. The pressure–velocity coupling solved using a coupled scheme based on an explicit approach. The pressure values were calculated by the Body-force-weighted discretization approach. Other variables, including momentum and mass equations, were calculated using a second order upwind scheme. Solution convergence for the numerical simulation was determined by the residual of energy being less than 10–6 and other variables residuals being less than 10–3 . During the numerical simulation process, the time step of 10–5 s was used for computational stability.

5.4 Results and Discussion 5.4.1 Model Validation and Orthogonal Test Results The water model in this study was simulated by the numerical simulation method with the same conditions to confirm the validity of the numerical model developed. A LS130-A propeller type fluid flow velocimeter shown in Fig. 5.2 was used in water model experiments to measure the water velocity. Figure 5.5 shows the measured flow velocity in water model experiments and the calculated values obtained by numerical simulations. It can be concluded that the values obtained by the numerical simulation are in good agreement with the measured values and the error is no more than 0.018 m/ s, while the average error is 0.014 m/s. Therefore, the numerical simulation results for liquid flow in the EAF can be considered to be in agreement with the measured values. Table 5.5 lists the orthogonal test results and the Software SPSS (Statistical Product and Service Solutions) was used to analyze the orthogonal test result. The Kolmogorov–Smirnov value is 0.374 and the orthogonal test results fits the normal distribution [18]. Therefore, the test results were analyzed by analysis of variance (ANOVA) with an additional experiment Item 10 listed in Table 5.5, which could Fig. 5.5 Measured velocities from experiments and numerical simulations

102

5 Modeling and Arrangement of Submerged Nozzles

provide the degrees of freedom (df) of the error term. The results of variance analysis obtained by SPSS are shown in Table 5.6. According to the variance analysis, it can be found that the Sig. (Significance) value of vertical dip angle is more than 0.005, which indicates that the vertical dip angle has no significant impact on the liquid bath mixing time. However, the gas supply rate, the horizontal arrangement mode and the submerged depth have significant impact [18]. For an EAF with two submerged nozzles with certain gas flow rate and submerged depth, the horizontal arrangement mode of the two submerged nozzles would determine the molten bath flow characteristics. A detailed analysis of the data is presented in Sect. 5.4.2. Table 5.5 Orthogonal test results Item

Factors

Mixing time (s)

Gas flow rate (Nm3 /h)

Horizontal arrangement mode

Vertical dip angle (°)

Submerged depth (mm)

1

2.5

A

5

30

217.0

2

2.5

B

10

45

184.0

3

2.5

C

15

60

200.0

4

5.0

A

15

45

84.0

5

5.0

B

5

60

77.5

6

5.0

C

10

30

112.0

7

7.5

A

10

60

62.0

8

7.5

B

15

30

73.0

9

7.5

C

5

45

81.0

10

7.5

B

15

60

53.5

Table 5.6 The results of variance analysis obtained by SPSS Source Corrected model

Type III sum of squares

df

Mean square

F

Sig

34,273.01

8

4284.12

4819.64

0.011

140,075.85

1

140,075.84

157,585.32

0.002

30,040.29

2

15,020.14

16,897.66

0.005

608.67

2

304.33

342.37

0.038

Vertical dip angle

71.67

2

35.834

40.31

0.111

Submerged depth

780.27

2

390.13

438.91

0.034

Intercept Gas flow rate Horizontal arrangement mode

Error

0.89

1

0.89

–

–

Total

165,147.50

10

–

–

–

34,273.90

9

–

–

–

Total of corrected model

5.4 Results and Discussion

103

5.4.2 Analysis of Water Model Experiment Figure 5.6 shows the mixing time measured in the present experiment under different gas supplying rates. It can be observed that with the increase of gas supplying rate, the mixing time decreases. With lower gas supplying rate, the gas jet kinetic energy at the exit of the submerged nozzle is lower and less energy could be transferred to the liquid bath. Therefore, the fluid flow velocity is less and the liquid bath mixing time is longer. With increase in the gas flow rate, the liquid bath flowing is strengthened by the submerged gas jet with larger jet velocity. However, it can also be observed in Fig. 5.6 that with the increase of the gas supply rate, the rate of decrease of the mixing time becomes less. One reason is that the energy exchange loss between the gas phase and the liquid phase increases for larger gas jet velocity. Figure 5.7 shows the change of the mixing time under different vertical dip angles of submerged nozzle. The liquid bath mixing time obviously decreases when the vertical dip angle of the submerged nozzle changes from 5° to 10°. With increase in the vertical dip angle, the vertical penetration depth of the submerged gas jet increases and as a result, the length of the submerged gas jet trajectory increases, which promotes energy exchange in the molten bath. Therefore, the mixing time decreases. When the vertical dip angle of the submerged nozzle changes from 10° to 15°, it can be observed that there is less change in the liquid bath mixing time, which is in accordance with the findings of a previous study carried out by our research team [8]. As reported [8], the trend of the variation of submerged gas jet behavior is less obvious in the range of 10°–15°. Figure 5.8 shows the mixing time under different submerges depths of submerged nozzle. The liquid bath mixing time decreases with the increase of submerged depth. The main reason is that larger submerged depth favors the jet energy transformation in the liquid bath and raises the conversion utilization ratio of the gas jet kinetic energy. Meanwhile, as reported in a previous study [8], the submerged depth has a strong influence on the erosion of the refractory material and therefore, in industrial application, the submerged depth of the submerged nozzle should be designed by considering the liquid bath flow and the erosion rate of the refractory. Fig. 5.6 The mixing time of the liquid bath

104

5 Modeling and Arrangement of Submerged Nozzles

Fig. 5.7 The mixing time under different vertical dip angles of submerged nozzle

Fig. 5.8 The mixing time under different submerges depths of submerged nozzle

Figure 5.9 shows the measured mixing time under different horizontal arrangement modes of the submerged nozzle. Compared with Mode A and Mode C, the liquid bath mixing time in Mode B is the least. In case of Mode B, circulating flow is formed in the liquid bath, while in Mode A and Mode C, the fluid flow streams generated by the two submerged nozzles would cause a loss to the fluid kinetic energy by interfering and colliding with each other. Figures 5.10, 5.11 and 5.12 depict the liquid bath flow field under Mode A, Mode B and Mode C, respectively. Initially in Fig. 5.10a, the ink has been added into the liquid bath and clockwise diffusion of the ink occurred under the promotion of the submerged nozzle a (shown in Fig. 5.4) as shown in Fig. 5.10b. The fluid flow streams collide with each other in the central area under the agitation effect of the two submerged gas jet nozzles and the ink stays in the EBT area without continued clockwise diffusion as shown in Fig. 5.10c. In Fig. 5.10d, the ink spread directly from the EBT area to the furnace door area and meanwhile, the ink diffuses to the sides of the liquid bath just as shown in Fig. 5.10e. And finally, the ink mixes uniformly in the liquid bath as shown in Fig. 5.10f. The fluid flow characteristics of the liquid bath in Mode B are different from that in Mode A. Initially in Fig. 5.11a, the ink has been added into the liquid bath and

5.4 Results and Discussion

105

Fig. 5.9 The mixing time in different horizontal arrangement modes of submerged nozzle

subsequently, anti-clockwise diffusion has occurred from the submerged nozzle a to submerged nozzle b as shown in Fig. 5.11b. Then driven by the promoting action of submerged nozzle a, the ink continues to spread anti-clockwise to the EBT area in Fig. 5.11c and then continues to diffuse anti-clockwise through the EBT area to submerged nozzle a as shown in Fig. 5.11d, e. Based on the path of the ink, it can be observed that a circulating flow is formed in the liquid bath due to the mutual promotion of the two submerged nozzles instead of collision. Therefore, in Mode B, the ink can uniformly mix with the circulating flow in a short time and no “dead zone” is formed. Figure 5.12 shows the ink diffusion process in Mode C and it can be observed that its liquid bath flow field characteristics are similar to that in Mode A. The fluid flow streams collide with each other in the EBT area under the agitation effect of the two submerged gas jet nozzles and as a result, a circulating flow cannot be formed in the liquid bath, which weakens the fluid flow. Compared with Mode A, the fluid flow streams collide in an area, which is closer to the EBT area and the liquid bath mixing time is greater as shown in Fig. 5.10.

5.4.3 Analysis of Numerical Simulation Figure 5.13 shows the velocity distribution of the liquid at cross sections of different molten bath depths in different horizontal arrangement modes. Figure 5.14 shows the velocity distribution of the liquid at a longitudinal section in different horizontal arrangement modes. Different colors represent different liquid flow velocities. Red represents the maximum velocity and blue represents the minimum velocity. It can be observed that the horizontal arrangement modes of submerged nozzles have a significant effect on the flow field of liquid bath. In the central area and the furnace door area, the fluid flows slowly in Mode A. The low velocity region where the fluid flow velocity is less than 0.06 m/s is more obvious. The flow field distribution in Mode C is similar to that in Mode A. The fluid flow velocity in the center area of

106

5 Modeling and Arrangement of Submerged Nozzles

Fig. 5.10 The fluid flow field in horizontal arrangement mode A. a Ink initial addition; b ink clockwise diffusion; c ink diffusion in EBT area; d ink diffusion from EBT area to furnace door; e ink diffusion around; f final liquid bath

the molten bath in Mode C is found to be larger than that in mode A. According to both Figs. 5.13 and 5.14, the liquid flow velocity in the furnace door area in Mode A is less than that in Mode C. However, the liquid flow velocity in the EBT region in Mode A is larger than that in Mode C. This phenomenon is consistent with that observed in the water model experiment. In Mode A, the fluid flow streams generated by the two submerged nozzles collide with each other in the central area and then, the kinetic energy cancellation occurs. However, in Mode C, the collision between the fluid flow streams occurs in the EBT area. In both Figs. 5.13 and 5.14, it is can be observed that the uniformity of velocity distribution under Mode B is better than that under Mode A and Mode C. Figure 5.15

5.4 Results and Discussion

107

Fig. 5.11 The fluid flow field in horizontal arrangement mode B. a Ink initial addition; b ink anticlockwise diffusion; c ink anti-clockwise diffusion to EBT area; d ink anti-clockwise diffusion through EBT area; e ink anti-clockwise diffusion after EBT area; f final liquid bath

shows the average value of the liquid flow velocity on cross section under different experimental schemes. The average liquid flow velocities in Mode B are, respectively, 0.314, 0.277 and 0.196 m/s at 100, 400 and 700 mm below the molten steel level, which are larger than those in Mode A and Mode C. It is also observed that for these three cross sections, the average value of the flow velocity in Mode A is a marginally greater than that in Mode C, which concurs reasonably well with the measured data obtained from the water model experiments. The turbulence kinetic energy is also used a measure to characterize the degree of mixing. Large value of turbulence kinetic energy typically implies more intense

108

5 Modeling and Arrangement of Submerged Nozzles

Fig. 5.12 The fluid flow field in horizontal arrangement mode C. a Ink initial addition; b ink clockwise diffusion; c ink diffusion in EBT area; d ink diffusion from EBT area to furnace door; e ink diffusion around; f final liquid bath

mixing [9]. The liquid flow characteristics in the EAF has been represented by variations in the turbulence kinetic energy. Figure 5.16 lists the average value of the turbulent kinetic energy in the liquid bath on cross section of different liquid bath depths in different horizontal arrangement modes. It can be observed that the trend of variation and the distribution characteristics of the turbulent kinetic energy are in accordance with that of the liquid flow velocity. The turbulent kinetic energy in Mode B is the largest and that in Mode C is the least. According to the results from numerical simulations, it can be concluded that the kinetic energy of high-speed gas jets generated by the submerged nozzles can be absorbed better by the EAF molten bath and transformed into the kinetic energy of

5.4 Results and Discussion

109

Fig. 5.13 Flow velocity distribution of the liquid on cross sections in different horizontal arrangement modes

the liquid steel in Mode B as compared to Modes A and C. Hence, the flow velocity distribution and homogeneity in an EAF in Mode B are better than that in Mode A and Mode C.

5.4.4 Industrial Application Research A commercial 75 ton EAF was modified to investigate the influence of the arrangement mode of two submerged nozzles on the metallurgical effects by a series of industrial standard tests. Three arrangement modes of submerged nozzles were adopted and Fig. 5.17 shows the details of the firebrick masonry for the submerged nozzle installation. The submerged nozzle was inserted through a previously made hole in the refractory bricks. In order to increase the service life of the EAF refractory during submerged CO2 – O2 injection, 750 mm long high-carbon magnesia carbon bricks were used around

110

5 Modeling and Arrangement of Submerged Nozzles

Fig. 5.14 Flow velocity distribution of the liquid on longitudinal sections in different horizontal arrangement modes Fig. 5.15 Average fluid flow velocity on different cross sections

the submerged nozzle. Refractory bricks in other regions were 500 mm long and made of conventional refractory material. The temperature homogeneity and the composition homogeneity of the EAF molten bath represents the agitation state of the molten bath. Figure 5.18 depicts the measured data of molten steel temperature deviation between the EBT area and furnace door area under different arrangement modes. The values of the temperature deviation are 19.3, 13.2 and 17.9 °C in Mode A, Mode B and Mode C, respectively.

5.4 Results and Discussion

111

Fig. 5.16 Average turbulent kinetic energy on different cross sections

Fig. 5.17 Firebrick masonry for submerged nozzle installation

It can be found that the temperature deviation in Mode B is minimum and that in Mode C is less than Mode A, which is in reasonable accordance with the results of numerical simulation and water model experiment. Figure 5.19 shows the measured data of the final molten steel with respect to carbon content difference between EBT area and furnace door area under different arrangement modes. It can be observed that the tendency of change in the carbon content deviation is consistent with that of the molten steel temperature deviation. The measured value in Mode B is minimum, which indicates that the molten bath stirring in Mode B is the best among the three modes. In addition to these, the average erosion rate of the firebrick around the submerged nozzle was calculated according to the industrial application data. The furnace life in Mode A, Mode B and Mode C is 242, 267 and 231 heats, respectively, with the remaining firebrick thickness being 98, 104 and 107 mm. Therefore, as shown in Fig. 5.20, the averaged erosion rate of the firebrick is about 2.69 mm/heat, 2.42 mm/ heat and 2.78 mm/heat in Mode A, Mode B and Mode C, respectively. It can be concluded that Mode B can reduce the erosion of the firebrick due to better molten

112

5 Modeling and Arrangement of Submerged Nozzles

Fig. 5.18 Temperature deviation between EBT area and furnace door area in different modes

Mode A Mode B Mode C

40

Carbon content/×10-5

Fig. 5.19 Carbon content difference between EBT area and furnace door area in different modes

30

29.0 25.6

20

18.2

10 0

10

20

30

Heats

bath agitation effect. It is believed that the fluid flow pattern under Mode B reduces interaction of gas plume with the refractory whereas collision between the liquid streams in Mode A and Mode C increase the impact. This appears to be the prime reason causing the difference in refractory wear behavior.

5.5 Conclusions A computational fluid dynamics model and a water model were developed to study and analyze the liquid flow characteristics in an EAF during submerged gas injection. The water model experiment has been conducted first to verify the reliability of numerical simulation, and then to determine the reasonable design scheme of EAF with submerged gas injection by orthogonal test. The flow field distribution of the EAF liquid bath with different horizontal arrangement modes of two submerged

5.5 Conclusions

113

Fig. 5.20 Average erosion rate of the firebrick around submerged nozzle in different modes

nozzles has also been monitored and analyzed. The following conclusions can be drawn: 1. The reliability of the computational fluid dynamics model is validated by a reasonably good agreement between the measurement data of water model experiment and the results of numerical simulation. 2. Water model experiments with an orthogonal test scheme have been conducted to study and analyze the influence of horizontal arrangement mode, gas supplying rate, vertical dip angle and submerged nozzle depth on the flow characteristics of the liquid bath. It can be found that the gas supplying rate, the horizontal arrangement mode and the vertical dip angle have significant impact on the mixing time of the liquid bath. 3. For an EAF with two submerged nozzles with a certain gas flow rate and submerged depth, the horizontal arrangement mode of the two submerged nozzles determines the flow conditions in the molten bath. The collision among the fluid flow streams in the liquid bath generated by the promotion of submerged nozzles would leads to cancellation of kinetic energy in Mode A and Mode C, which weakens the molten bath flowing. In Mode B, a circulating flow is formed and as a result, the fluid flows at a faster velocity with no “dead zone” in the EAF molten bath. 4. On comparing with Mode A and Mode C, the EAF in Mode B can obtain better molten bath stirring with better temperature and composition homogeneity. The erosion rate of the fire bricks in the periphery of the submerged nozzle in Mode B is 2.42 mm/heat, with is less than that in Mode A (2.69 mm/heat) and Mode C (2.78 mm/heat). This is because the circulating fluid flow formed in the molten bath in Mode B is able to reduce the erosive effect of the submerged gas jet on the refractory.

114

5 Modeling and Arrangement of Submerged Nozzles

References 1. Lee B, Sohn I (2014) Review of innovative energy savings technology for the electric arc furnace. JOM 66(9):1581–1594 2. Wang C, Brämming M, Larsson M (2013) Numerical model of scrap blending in BOF with simultaneous consideration of steel quality, production cost, and energy use. Steel Res Int 84(4):387–394 3. Li L, Jiang M, Duan Z (1996) Simulation research on the rising height of the bath in bottom stirring EAF. Steelmaking 2(3):49–52 4. Wei G, Zhu R, Han B, Yang S, Dong K, Wu X (2020) Simulation and application of submerged CO2 -O2 injection in electric arc furnace steelmaking: modeling and arrangement of submerged nozzles. Metall Mater Trans B 51(6):1101–1112 5. Alam M, Naser J, Brooks G et al (2010) Computational fluid dynamics modeling of supersonic coherent jets for electric arc furnace steelmaking process. Metall Mater Trans B 41(6):1354– 1367 6. Alam M, Naser J, Brooks G et al (2012) A computational fluid dynamics model of shrouded supersonic jet impingement on a water surface. ISIJ Int 52:1026–1035 7. Wei G, Zhu R, Wu X, Dong K, Yang L, Liu R (2018) Technological innovations of carbon dioxide injection in EAF-LF steelmaking. JOM 70(3):969–976 8. Wei G, Zhu R, Tang T, Dong K, Wu X (2019) Study on the impact characteristics of submerged CO2 and O2 mixed injection (S-COMI) in EAF steelmaking. Metall Mater Trans B 50(3):1077– 1090 9. Liu F, Zhu R, Dong K et al (2015) Simulation and application of bottom blowing in electrical Arc furnace steelmaking process. ISIJ Int 55(3):2365–2373 10. Ramirez M, Alexis J, Trapaga G, Jonsson P, Mckelliget J (2001) Modeling of a DC electric arc furnace-mixing in the bath. ISIJ Int 41(3):1146–1155 11. Alam M, Irons G, Brooks G et al (2011) Inclined jetting and splashing in electric arc furnace steelmaking. ISIJ Int 51(3):1439–1447 12. Ma J, Zhou P, Cheng W (2016) Dimensional analysis and experimental study of gas penetration depth model for submerged side-blown equipment. Exp Thermal Fluid Sci 75(3):220–227 13. Zhan S, Lai C, Hsiao T (2003) CFD analysis of gas stirring behavior in side-blown metallic bath. Cent South Univ Technol 32(2):148–151 14. Cafeery G, Warnica D, Molloy N, Lee M (1977) International conference on CFD in mineral & metal processing and power generation. CSIRO, pp 87–100 15. Li B (2000) Fluid flow and mixing process in a bottom stirring electrical arc furnace with multi-plug. ISIJ Int 40(3):863–869 16. Singh V, Kumar J, Bhanu C et al (2007) Optimisation of the bottom Tuyeres configuration for the BOF vessel using physical and mathematical modelling. Trans Iron Steel Inst Jpn 47(3):1605–1603 17. Zhou X, Ersson M, Zhong L et al (2013) Mathematical and physical simulation of a top blown converter. Steel Res Int 85(3):273–281 18. Norusis M (1993) SPSS for windows: base system user’s guide release 5.0. 437–483 19. Hirt CW, Nichols BD (1981) Volume of fluid (VOF) method for the dynamics of free boundaries. J Comput Phys 39(1):201–225 20. Wei G, Zhu R, Dong K et al (2016) Research and analysis on the physical and chemical properties of molten bath with bottom-blowing in EAF steelmaking process. Metall Mater Trans B 47(3):3066–3079 21. Launder B, Spalding D (1972) Lectures in mathematical model of turbulence. Academic Press, London, pp 124–129

Chapter 6

Combined Blowing and Industrial Application

6.1 Introduction In Chap. 5, water model experiments, numerical simulations, and industrial tests were performed to determine the arrangement of the submerged gas jet, and Mode B was considered the optimized arrangement scheme (see Sect. 5.2.1 for a schematic of this mode) [1–4]. As the oven wall oxygen lance is a significant element in EAF steelmaking owing to its superiority in controlling the metal bath stirring, chemical reaction kinetics, and metal bath recirculation, it is of great importance to establish a new oxygen supplying system by combining oven wall oxygen lances and submerged nozzles optimally. Various studies have been conducted to optimize the combination of bottomblowing and top oxygen-supplying in converter steelmaking [5–7], while only a few studies on EAF combined blowing have been reported. Zhou et al. [5] analyzed the flow field of a top–bottom-side blown converter metal bath by a physical model and a mathematical model. Chu et al. [6] analyzed how the flow rate distribution of bottom-blown gas affects the metal bath mixing efficiency in converter steelmaking using a 3D multiphase flow numerical model. Logar et al. [8–13] developed numerical models to analyze the physicochemical phenomena of EAF steelmaking by considering mass transfer, heat transfer, and thermo-chemical reactions. Ma et al. [14] optimized the arrangement of bottom-blowing nozzles and oven wall oxygen lances in a 50 t EAF and analyzed the corresponding metallurgical effect. Dong et al. [15] studied the industrial applicability of bottom-blowing and oven wall oxygensupplying in a Consteel EAF and found that the effective combination of the two methods could realize energy saving and ferrous charges consumption reduction. However, it is still necessary to optimize the arrangement of the submerged nozzles and oven wall oxygen lances in combined blowing EAFs for improving the industrial applicability of this novel technology. This chapter focused on the determination and analysis of the metallurgical reaction characteristics of the EAF metal bath with submerged gas injection and oven wall oxygen supply. To this end, a physical model and a numerical model of an EAF © Metallurgical Industry Press 2024 G. Wei and R. Zhu, Electric Arc Furnace Steelmaking with Submerged Mixed Injection, https://doi.org/10.1007/978-981-99-4602-0_6

115

116

6 Combined Blowing and Industrial Application

with two oven wall oxygen lances and two submerged nozzles were developed to study the flow field of metal bath under combined blowing. The arrangement of two oven wall oxygen lances was optimized so that the flow field produced by the lances matched that produced by the two submerged nozzles in Mode B. Finally, combined blowing was implemented in a 75 t commercial EAF to investigate the variation of some key smelting technical indexes with industrial production data.

6.2 Water Model Experiment 6.2.1 Model Instruments and Orthogonal Test Scheme Based on the water model instruments, the experimental setup was optimized by adopting two oven wall oxygen lances and their relative control lines to investigate the effect of combined blowing, which included submerged gas injection and oven wall oxygen supply, on the EAF metal bath stirring. The optimized water model experimental setup is shown in Fig. 6.1. The geometric and injection parameters were selected according to geometric and dynamic similarity criteria; details are provided in a previous study [4]. Tables 6.1 and 6.2 list the geometric parameters and gas flow rates of the oven wall oxygen lances used in this study, respectively. In Table 6.1, d throat and d exit are the diameters of the throat and exit of the oven wall oxygen lance, respectively, while Ma is the Mach number of the oven wall oxygen lances, which was equal to 2.0 in both the EAF prototype and water model. According to a previous study [4], Mode B (red lines in Fig. 6.2) is the optimal mode of the submerged nozzle horizontal arrangement because it presents the shortest

Fig. 6.1 Water model experiment platform

6.2 Water Model Experiment

117

Table 6.1 Geometric parameters of oven wall oxygen lances used in this study Item

Ma

Parameters of the oven wall oxygen lances (mm) d throat

d exit

Length of expansion section

Length of contraction section

EAF prototype

2.0

23

30

49

28

Water model

2.0

1.8

2.4

3.88

2

Table 6.2 Gas flow rates of a single oven wall oxygen lance

Item

Gas flow rate (Nm3 /h)

EAF prototype

1000

1500

2000

Water model

12.4

18.6

24.8

mixing time and the largest fluid flow velocity. In this study, the influence of the oven wall oxygen lance horizontal arrangement modes on the metal bath stirring was investigated with the horizontal arrangement mode of the submerged nozzle being Mode B. Figure 6.2 presents different horizontal arrangement modes—Modes D, E, and F—for two oven wall oxygen lances in an EAF. It should be noted that the arrangement scheme of the oven wall oxygen lances was designed according to the initial design and structure of a commercial 75 t EAF. In this study, an orthogonal test scheme with three factors and three levels was employed to study the influence of horizontal arrangement mode, gas flow rate, and vertical dip angle of oven wall oxygen lances on the characteristics of the EAF metal bath; Table 6.3 provides details about this scheme, including the corresponding measured data.

20°

20° 20° 38°

38°

20° 20°

10°

10°

20° 20° 20°

D

E

F

Fig. 6.2 Horizontal arrangement modes in an EAF with two oven wall oxygen lances. The red and blue lines represent submerged nozzles and oven wall oxygen lances, respectively

118 Table 6.3 Orthogonal test scheme and the corresponding measured data

6 Combined Blowing and Industrial Application

Factors

Mixing time (s)

Gas flow rate (Nm3 /h)

Vertical dip angle (°)

Horizontal arrangement mode

12.4

40

D

52.5

12.4

42

E

47.0

12.4

44

F

41.5

18.6

40

E

38.0

18.6

42

F

34.5

18.6

44

D

38.0

24.8

40

F

29.5

24.8

42

D

36.5

24.8

44

E

32.0

6.2.2 Analysis of the Water Model Experiment Results Based on the results listed in Table 6.3, the Kolmogorov–Smirnov value of the data, which was obtained using SPSS, was 0.811, which indicates that the orthogonal test results present a normal distribution. Table 6.4 shows the relative results of the variance analysis obtained by SPSS. The measured significance (Sig.) results show that the vertical dip angle was not a significant influencing factor for bath stirring; the opposite is observed for the gas flow rate and horizontal arrangement mode. The change of liquid bath mixing time at different oven wall oxygen lance gas flow rates is shown in Fig. 6.3. The mixing time clearly decreases with increasing the gas flow rate of the oven wall oxygen lance. The main cause was due to the energy transformation between gas jet and liquid bath becoming more intense at larger gas flow rates; i.e., the fluid flow velocity increases and the mixing time decreases. In addition, it was found that the amplitude of the mixing time decreased with the gas flow rate increases. The main possible reason is that the liquid splashes more Table 6.4 Results of the variance analysis Source

df

Mean square

F

Sig

416.50

6

69.42

39.67

0.025

13,572.25

1

13,572.25

7755.57

0.000

326.16

2

163.08

93.19

0.011

Vertical dip angle

13.16

2

6.58

3.76

0.210

Horizontal arrangement mode

77.16

2

38.58

22.05

0.043

Error

3.50

2

1.75

–

–

Total

13,992.25

9

–

–

–

420.00

8

–

–

–

Corrected model Intercept Gas flow rate

Total for the corrected model

Type III sum of squares

6.2 Water Model Experiment

119

Fig. 6.3 Mixing time at different oven wall oxygen lance gas flow rates

vigorously under the impact of the gas jet at larger gas flow rates, which may cause partial energy loss. In Fig. 6.4, it can be seen that as the vertical dip angle of oven wall oxygen lance increases, the mixing time decreases. For small vertical dip angles, the force exerted on the liquid is determined by the gas–liquid interface boundary layer. However, the jet’s vertical velocity increases with the vertical dip angle increasing, and the presence of a boundary layer becomes a nonrestrictive condition. As a result, the impact pits can transform the kinetic energy of the jet into kinetic energy of the fluid. However, for a large vertical dip angle, the horizontal force exerted on the liquid would be significantly small, which would weaken the fluid flow in the liquid bath. Figure 6.5 shows the mixing times of the different horizontal arrangement modes of the oven wall oxygen lances. It can be observed that with the oxygen lance’s deflection angle increasing, the mixing time decreases. Compared with Modes D and E, the mixing time of Mode F is the shortest because this mode can accelerate the circulating flow in the liquid bath by reducing flow stream intersection and collision. Fig. 6.4 Mixing time at different oven wall oxygen lance vertical dip angles

120

6 Combined Blowing and Industrial Application

Fig. 6.5 Mixing time under different horizontal arrangement modes of oven wall oxygen lances

According to the experimental results, the scheme in Mode F and with a vertical dip angle of 44° is the optimized scheme for the EAF with combined blowing.

6.3 Numerical Simulation 6.3.1 Numerical Modeling The promotion of combined blowing on the EAF metal bath flow field was investigated by a numerical simulation that considered two submerged nozzles and two oven wall oxygen lances. The volume of fluid (VOF) method [16] was used, and the modeling method presented in a previous study [2, 4], which has been validated using measured data, was employed. And in previous studies [2, 4], the corresponding governing equations have also been listed. As shown in Fig. 6.6, a commercial 75 t EAF with two submerged nozzles and two oven wall oxygen lances was employed to study the EAF metal bath flow field with combined blowing. The model was built using a scale ratio of 1:1; and Table 6.1 lists the relative parameters of EAF geometry model. The two submerged nozzles were in Mode B and their vertical dip angle was 10°; the two oven wall oxygen lances were in Mode F and their vertical dip angle was 44°. The gas flow rate of the oven wall oxygen lances was 2000 Nm3 /h, and those of the submerged nozzles were 200, 400, and 600 Nm3 /h. In this study, a new method for modeling the impingement of EAF oven wall oxygen jet was adopted. Three different computational domains are used: two domains for a single-phase jet simulation and one domain for a three-phase simulation. For two oven wall oxygen lances, the high-speed oxygen jets were calculated in the gas-phase domains, while the gas-steel-slag interactions were calculated in the three-phase domain. Figure 6.7 presents the grid arrangements of the computational domains used in this study.

6.3 Numerical Simulation

121 Oven wall oxygen lances

Submerged jet nozzles

Fig. 6.6 Geometric model of an EAF combining submerged nozzles and oven wall oxygen lances

Fig. 6.7 Connection between the three computational domains and detailed grid arrangements

A velocity boundary condition at the inlet of the oven wall oxygen lances and a mass-flow condition at the inlet of the submerged nozzles were adopted in the three-phase domain. For the single-phase domains, a separate file was formed by the calculated data of the interface which includes the vector velocity data and the total pressure data. And in the three-phase numerical simulation, the derived separate file was applied as the boundary condition of oven wall oxygen lance inlet. Furthermore, the boundary condition of the outlet of EAF was defined as pressure outlet with its value being 101,325 Pa. The calculations were performed in a transient solution mode and the pressure–velocity coupling scheme was achieved with the pressure-implicit with splitting of operators (PISO) algorithm. The momentum and mass equations were solved using second-order upwind schemes. In addition, for all cases, the numerical simulation was considered to be convergent when the residuals of the energy and other dependent variables were less than 10–6 and 10–3 , respectively.

122

6 Combined Blowing and Industrial Application

6.3.2 Analysis of Numerical Simulation Results The flow fields of EAF metal bath on a cross section with various submerged gas flow rates are depicted in Fig. 6.8, where red and blue areas denote maximum and minimum velocities, respectively. Figures 6.8a, b present the fluid flow velocity distribution across the slag-steel interface and on the cross section 100 mm below the molten steel level, respectively. The flow field distributions in these figures are highly similar; therefore, the gas flow rate of submerged nozzle has no significant influence on the fluid flow velocity in this region, which is located in the metal bath’s upper part. Thus, in this region, the liquid is mainly stirred by the oven wall oxygen lance. Figures 6.8c, d depict the fluid flow velocity distribution on the cross section 400 mm and 700 mm below the slag-steel interface, respectively. Clear differences can be found among these figures, which indicates that the gas flow rate of the submerged nozzle has a stronger effect on the bath stirring in these regions; as the submerged flow rate increases, the flow velocity of metal bath increases. Within the metal bath’s middle and lower parts, the agitation produced by the submerged gas jet is the factor that increases the metal bath stirring and accelerates the fluid flow. In this region, the agitation effect of the oven wall oxygen lance is limited owing to the velocity attenuation and limited impact penetration capacity of the high-speed oxygen jet. The flow field distributions across different longitudinal sections of the metal bath at different gas flow rates of the submerged nozzle are depicted in Fig. 6.9. The fluid flow field variations are in reasonable agreement with the phenomena shown in Fig. 6.8. The low-speed flow zone is mainly located in the metal bath’s middle and lower parts, and as the gas flow rate of submerged nozzle increases, the fluid flow velocity increases significantly and the span of the low-speed zone decreases sharply. Figure 6.10 lists the flow velocity data of the EAF metal bath for the two cases and as expected, the flow velocity is increased when two oven wall oxygen lances are employed, and the velocity increase proportion varies among cross sections. The average fluid flow velocity at the slag-steel interface is increased by 47.9% and that of the cross section 100 mm, 400 mm, and 700 mm below the slag-steel interface are increased by 21.7%, 6.1%, and 5.6%, respectively. Based on the above measured data, it can also be confirmed that the oven wall oxygen lance performs an important function in increasing the stirring of the EAF metal bath’s upper part; thus, the kinetics of the metallurgical reaction in the slag-steel region can be improved. Figure 6.11 depicts the flow field in the EAF metal bath with combined blowing. The results obtained by the numerical simulation are consistent with those from the water model experiment (Sect. 6.2.1). The directions of the fluid flow streams produced by the two oven wall oxygen lances in Mode F and the two submerged nozzles in Mode B match each other, thereby realizing a circulating flow in the metal bath and reducing the kinetic energy cancellation caused by the collision of fluid flow streams. In this case, the agitation effect of the oven wall oxygen lances and submerged nozzles improves the metal bath stirring significantly. The impact cavity formed by the high-speed oven wall oxygen jet can promote the flow of molten steel

6.3 Numerical Simulation

123

Fig. 6.8 Velocity distribution on different cross sections of the metal bath. a slag-steel interface; b 100 mm below slag-steel interface; c 400 mm below slag-steel interface; d 700 mm below slag-steel interface

from the cavity center to the surroundings. In addition, the radial velocity induced by the oven wall oxygen jet horizontal inclination angle promotes the molten steel flow in a certain direction. For the submerged nozzles, the molten steel flows in the direction of the submerged gas jet. Figure 6.12 depicts the cavity-containing profile caused by combined blowing impingement in EAF steelmaking. It can be seen that the surface fluctuations around the cavities are more significant under the intense operation of the submerged nozzles and oven wall oxygen lances, which can benefit the metallurgical reactions. The

124

6 Combined Blowing and Industrial Application

Fig. 6.9 Velocity distribution on different longitudinal sections of the metal bath Fig. 6.10 Average flow velocity on different cross sections

molten slag is pushed away by the high-speed oxygen jet generated by both the submerged nozzles and the oven wall oxygen lances. The shape of the impinging cavity is determined by the interaction between the oxygen jet and the metal bath, which simultaneously affects the flow state of the metal bath. Table 6.5 lists the exposed areas of molten steel at different gas flow rates of the submerged nozzles. As the gas flow rate increases, the interaction between gas jet and molten steel is strengthened; however, part of the jet kinetic energy could not be transformed into the energy of metal bath. As a result, the amount of gas that escapes together with the

6.3 Numerical Simulation

125

Fig. 6.11 Fluid flow at 100 mm below slag-steel interface

surrounding refractory increases, and the exposed molten steel area expands. This gas can strengthen the mixing of the molten slag and steel, thereby enlarging the slag-steel reaction interface and increasing the metallurgical reaction rate. Fig. 6.12 Cavity-containing profile caused by combined blowing impingement in EAF steelmaking

126 Table 6.5 Exposed areas of molten steel at different submerged nozzle gas flow rates

6 Combined Blowing and Industrial Application Gas flow rate (Nm3 /h)

Exposed area of molten steel (Nm2 )

200

0.110

400

0.168

600

0.273

Fig. 6.13 Arrangement of the oxygen injectors in a commercial EAF. a Conventional smelting; b combined blowing

6.4 Industrial Application Research 6.4.1 Industrial Application Scheme According to the above analysis, a commercial 75 t eccentric bottom taphole (EBT) EAF was reformed by installing two submerged nozzles and two oven wall oxygen lances, as shown in Fig. 6.13, to study the metallurgical effects of submerged CO2 – O2 injection in EAF steelmaking. In a previous study [4], the submerged nozzles were inserted through previously-made holes in the refractory bricks; details about the firebrick masonry used for submerged nozzle installation can be found in that work. Table 6.6 lists the installation parameters of the submerged nozzles and oven wall oxygen lances.

6.4.2 Analysis of the Industrial Application Results The furnace profile of EAF is flat, and the stirring of EAF metal bath affects the temperature and composition homogeneity; the weaker the bath stirring, the larger the temperature and composition deviations. In EAF steelmaking, the temperature

6.4 Industrial Application Research Table 6.6 Installation parameters of the submerged nozzles and oven wall oxygen lances

127

Parameter

Submerged nozzle

Oven wall oxygen lance

Arrangement mode

Mode B

Mode F

Vertical dip angle

10°

44°

Submerged depth

240 mm

–

Gas jet composition (volume fraction)

CO2 = 10% O2 = 90%

O2 = 100%

homogeneity and composition homogeneity of the final molten steel are key technical parameters. Figures 6.14 and 6.15 show the temperature deviation and carbon content deviation of the final molten steel between the EBT area and furnace door area, respectively. The average temperature deviation for conventional smelting is 20.2 K, while that for combined blowing is 11.9 K; i.e., combined blowing reduces the temperature deviation by 41.1%. The average carbon content deviation for conventional smelting is 3.3 × 10–4 , while that for combined blowing is 1.6 × 10–4 ; i.e., combined blowing reduces the carbon content deviation by 51.5%. Figure 6.16 shows the FeO content distribution in the molten slag for conventional smelting and combined blowing. Compared with EAF conventional smelting, the FeO content distribution in the molten slag shifts to the left side of the graph for combined blowing, which indicates that the content of FeO in the molten slag decreases with combined blowing. Table 6.7 lists the average CaO, SiO2 , MgO, TFe, FeO, and P2 O5 contents of the molten slag for conventional smelting and combined blowing. The contents of FeO and TFe are 25.65% and 23.56% for conventional smelting, respectively, and 20.37% and 18.93% for combined blowing, respectively; i.e., the slag FeO content and the TFe loss are reduced by 5.27% and 4.63%, respectively, when the submerged injection technology is adopted. The lower FeO content indicates an improvement in the smelting efficiency, and the reduction of the TFe Fig. 6.14 Temperature deviation between the EBT area and the furnace door area

128

6 Combined Blowing and Industrial Application

Fig. 6.15 Carbon content deviation in the molten steel between the EBT area and the furnace door area

loss increases the metal yield rate. The main reason for this is that the EAF metal bath stirring was improved under combined blowing, which enlarged the molten slag-steel reaction interface and increased the mass transport velocity. Hence, the slag-steel metallurgical reaction rate was improved, and the iron loss in the molten slag was reduced.

Fig. 6.16 FeO content distribution in the molten slag

Table 6.7 Average composition of the molten slag in EAF steelmaking Item

Composition of molten slag (%) CaO

SiO2

MgO

TFe

FeO

P2 O5

Conventional smelting

40.33

12.12

3.36

23.56

25.65

0.82

Combined blowing

41.47

15.13

2.77

18.93

20.37

1.27

6.4 Industrial Application Research

129

In EAF steelmaking, the phosphorus content of the molten steel is a fundamental parameter of steel quality, and phosphorus removal is a crucial task during smelting. Equation (6.1) presents the dephosphorization reaction, and Eq. (6.2) shows the expression for calculating the equilibrium phosphorus content. As reported previously [17], both a molten slag of high basicity (2.0–4.0) and a suitable smelting temperature (1823–1853 K) can increase the removal of phosphorus. The oxidizing property of the molten slag is another factor to consider because increasing this parameter can enable the removal of phosphorus. 2[P] + 5(FeO) + 4(CaO) = 4CaO · P2 O5 (s) + 5[Fe] lg

(6.1)

  (mass% P) = 0.072 (mass%CaO) + 0.3(mass%MgO) [mass% P] 11570 + 2.5 lg(mass% T.Fe) + − 10.52 T

(6.2)

On the other hand, it should be noted that the kinetic conditions of the metallurgical reactions are of great significance for phosphorus removal and for strengthening the metal bath stirring because they can intensify the phosphorus transfer from steel to slag, which can improve the dephosphorization process. Figure 6.17 shows the final phosphorus content of the molten steel for conventional smelting and combined blowing. The final phosphorus content is decreased by 3.8 × 10–3 % in EAF steelmaking with combined blowing, which reflects a marked improvement in dephosphorization. Although Fig. 6.16 and Table 6.7 show that combined blowing reduces the FeO content and oxidizing properties of the molten slag, which negatively affects phosphorus removal, most of the phosphorus can be effectively removed by combined blowing in the actual industrial application. Therefore, it is demonstrated that the improvement in phosphorus removal during EAF smelting was mainly caused by the enhanced metal bath stirring and improved kinetic conditions realized by combined blowing. 25

Phosphorus content /10 -5

Fig. 6.17 Phosphorus content distribution in the final molten steel

Conventional smelting Combined blowing

20

15

15.1

11.3

10

5 0

10

20

30

Heats

40

50

60

130

6 Combined Blowing and Industrial Application

Fig. 6.18 Endpoint carbon–oxygen equilibrium of molten steel

The final carbon–oxygen equilibrium, [%C][%O], is another fundamental parameter in EAF steelmaking because it can represent the peroxidation degree and quality of the molten steel [18]. In industrial applications, to improve the molten steel quality, significant efforts are exerted to decrease the oxygen content of molten steels with a certain carbon content. The more intense the metal bath stirring, the more complete the reaction of carbon and oxygen in the metal bath; moreover, the lower the peroxidation degree of molten steel, the lower the endpoint carbon–oxygen equilibrium value. Figure 6.18 lists the final molten steel’s carbon content and oxygen content for the studied industrial application. The endpoint carbon–oxygen equilibrium for EAF conventional smelting is 0.00318, while that for combined blowing is 0.00252; i.e., on average, combined blowing reduces the endpoint carbon–oxygen equilibrium by 0.00066. Therefore, it is certain that EAF steelmaking with submerged injection reduces the final molten steel’s oxygen content, allows the reaction of carbon and oxygen to reach the equilibrium state more easily, and improves the molten steel quality owing to the superior metal bath stirring. The endpoint molten steel’s temperature was depicted in Fig. 6.19. For conventional EAF smelting, the molten steel temperature range is wider and the temperature deviation among different heats is larger than those for combined blowing. Moreover, for EAF combined blowing, the proportion of temperature data points in the range of 1893–1903 K (1620–1630 °C) is higher than 90%. As mentioned previously, the temperature homogeneity in the EAF metal bath with combined blowing is improved by the enhanced metal bath stirring; hence, the control of both the final temperature hit rate and final molten steel temperature stability is improved, which can benefit the subsequent refining process.

6.5 Conclusions

131

Fig. 6.19 Distribution of the final temperature of the molten steel

6.5 Conclusions In this chapter, water model experiments and numerical simulations were performed to investigate the flow field characteristics in the EAF metal bath under combined blowing, considering two submerged nozzles and two oven wall oxygen lances. In addition, the metallurgical effects and technical indicators of EAF steelmaking with combined blowing were studied by performing industrial application experiments. The main results and conclusions of this study are summarized as follows. 1. Based on Mode B of the submerged nozzles, the optimized EAF combined blowing scheme was found to be the scheme presenting the two oven wall oxygen lances in Mode F and a vertical dip angle of 44°. 2. The submerged gas jets mainly agitate the middle and lower parts of the metal bath, while the oven wall oxygen lances mainly agitate the upper part of the metal bath. The directions of the flow streams produced by two oven wall oxygen lances in Mode F and the two submerged nozzles in Mode B match each other, forming a circulating flow in the metal bath and reducing the kinetic energy cancellation caused by the collision of fluid flow streams. As a result, the metal bath stirring can be strengthened significantly by EAF combined blowing. 3. EAF combined blowing with submerged injection demonstrates good metallurgical advantages in industrial application; compared with conventional EAF smelting, it improved the molten steel’s temperature and composition homogeneity, reduced the FeO content in the molten slag by 5.28%, and decreased the average phosphorus content in the final molten steel by 0.0038%. In addition, the endpoint carbon–oxygen equilibrium with combined blowing was 0.00252, which was improved by 0.00066 owing to the enhanced metal bath stirring.

132

6 Combined Blowing and Industrial Application

References 1. Wei G, Zhu R, Wu X, Dong K, Yang L, Liu R (2018) Technological innovations of carbon dioxide injection in EAF-LF steelmaking. JOM 70(3):969–976 2. Wei G, Zhu R, Tang T, Dong K, Wu X (2019) Study on the impact characteristics of submerged CO2 and O2 mixed injection (S-COMI) in EAF steelmaking. Metall Mater Trans B 50(3):1077– 1090 3. Wei G, Zhu R, Cheng T, Dong K, Yang L, Tang T, Wu X (2018) Effect of main gas composition on flow field characteristics of supersonic coherent jets with CO2 and O2 mixed injection (COMI) at steelmaking temperature. ISIJ Int 58(5):842–851 4. Wei G, Zhu R, Han B, Yang S, Dong K, Wu X (2020) Simulation and application of submerged CO2 -O2 injection in electric arc furnace steelmaking: modeling and arrangement of submerged nozzles. Metall Mater Trans B 51(6):1101–1112 5. Zhou X, Ersson M, Zhong L, Jönsson P (2015) Numerical and physical simulations of a combined top-bottom-side blown converter. Steel Res Int 86(11):1328–1338 6. Chu K, Chen H, Lai P, Wu H, Liu Y, Lin C, Lu M (2016) The effects of bottom blowing gas flow rate distribution during the steelmaking converter process on mixing efficiency. Metall Mater Trans B 47(3):948–962 7. Li Y, Lou W, Zhu M (2013) Numerical simulation of gas and liquid flow in steelmaking converter with top and bottom combined blowing. Ironmaking Steelmaking 40(7):505–514 8. Logar V, Dovzan D, Skrjanc I (2011) Mathematical modeling and experimental validation of an electric arc furnace. ISIJ Int 51(3):382–391 9. Logar V, Dovzan D, Skrjanc I (2012) Modeling and validation of an electric arc furnace: part 1, heat and mass transfer. ISIJ Int 52(3):402–412 10. Logar V, Dovzan D, Skrjanc I (2012) Modeling and validation of an electric arc furnace: part 2, thermo-chemistry. ISIJ Int 52(3):413–423 11. Logar V, Skrjanc I (2012) Modeling and validation of the radiative heat transfer in an electric arc furnace. ISIJ Int 52(7):1225–1232 12. Fathi A, Saboohi Y, Skrjanc I, Logar V (2015) Low computational-complexity model of EAF arc-heat distribution. ISIJ Int 55(7):1353–1360 13. Logar V, Fathi A, Skrjanc I (2016) A Computational model for heat transfer coefficient estimation in electric arc furnace. Steel Res Int 87(3):330–338 14. Ma G, Zhu R, Dong K, Li Z, Liu R, Yang L, Wei G (2016) Development and application of electric arc furnace combined blowing technology. Ironmaking Steelmaking 43(3):594–599 15. Dong K, Zhu R, Liu W (2012) Bottom-blown stirring technology practiced in consteel EAF. Adv Mater Res 363(3):639–643 16. Hirt C, Nichols B (1981) Volume of fluid (VOF) method for the dynamics of free boundaries. J Comput Phys 39(1):201–225 17. Manning CP, Fruehan RJ (2013) The rate of the phosphorous reaction between liquid iron and slag. Metall Mater Trans B 44(4):37–44 18. Wei G, Zhu R, Dong K et al (2016) Research and analysis on the physical and chemical properties of molten bath with bottom-blowing in EAF steelmaking process. Metall Mater Trans B 47(5):3066–3079

Chapter 7

Innovations of Injection Metallurgy in EAF Steelmaking

7.1 Introduction In the above chapters, simulation and application of EAF steelmaking with submerged mixed injection were studied systematically. As we known, EAF steelmaking process demonstrates greater advantages than converter steelmaking process, with carbon dioxide (CO2 ) emissions of EAF steelmaking process being about 600– 800 kg/t and that of converter steelmaking process being about 2000–3000 kg per ton [1, 2]. According to the World Steel Association statistical data on steel production in 2016, more than 410 million tons of steel were produced by EAF steelmaking process, which was about 25.3 pct of the world’s steel supply in 2016 [3]. In China, the amount of newly-built EAFs has surpassed more than 100 in 2017 with the issuance of new policies about structural transition of iron and steel industry. It can be predicted that EAF steelmaking process will play an increasingly important role in steel production for its green environmental protection and flexibility in materials utilization and production. However, there are also some limitations for high quality steel production and some smelting operations to be optimized in EAF steelmaking process. The first problem is that a great deal of dust is generated and the iron loss is heavy [4, 5]. Restricted by the furnace structure of eccentric bottom tapping (EBT) EAF, the flow velocity of molten steel is slow and the metallurgical reaction kinetics is weak, which is bad for dephosphorization [6, 7]. During EAF steelmaking, nitrogen can be absorbed by molten steel with arc ionization and the amount of CO bubbles generated among molten bath is less with lower melting carbon content, which makes nitrogen removal difficult [8, 9]. In recent years, researchers have carried out series of investigations to solve these problems by injecting CO2 into EAF [10, 11]. What’s more, CO2 also can be injected into the ladle furnace (LF) by bottom blowing nozzles to cut down the argon consumption, which is helpful to improve molten steel quality and reduce smelting cost [12, 13]. As a greenhouse gas, the emission of CO2 from steel industry is around 1.84 billion tons per year [14, 15]. As a weak oxidizer, CO2 could react with elements in molten steel and demonstrates some advantages in © Metallurgical Industry Press 2024 G. Wei and R. Zhu, Electric Arc Furnace Steelmaking with Submerged Mixed Injection, https://doi.org/10.1007/978-981-99-4602-0_7

133

134

7 Innovations of Injection Metallurgy in EAF Steelmaking

metallurgical reactions [16]. Hence, as green steelmaking technologies, CO2 injection technologies in EAF steelmaking not only improve metallurgical effects and steel quality but also benefit the emission reduction and resource utilization of CO2 . In this chapter, the recent advances and improvements of injection metallurgy with CO2 utilization in EAF Steelmaking have been reviewed. Metallurgical reaction principles and effects of relative innovative technologies, such as coherent jets with CO2 and O2 mixed injection (COMI), submerged O2 and powder injection with CO2 , and bottom blowing CO2 in EAF and LF are illustrated. Based on these innovations, a possible future process for cyclic utilization of CO2 in EAF-LF steelmaking process is proposed and discussed.

7.2 Coherent Jets with COMI Comparing with conventional coherent jets, the coherent jet with COMI is a new method that CO2 is mixed into the main O2 and then the mixed gas is injected out by the central Laval nozzle, which is helpful to reduce the amounts of iron loss and dust during EAF steelmaking [17]. The fluid flow characteristics, the reaction mechanism between CO2 and elements in molten steel and the thermodynamic calculation of COMI are analyzed as follows [18]. Figure 7.1 shows the total temperature and CO2 distributions of coherent jets with COMI at steelmaking temperature (1700 K) on longitudinal section. Compared with pure O2 , CO2 addition in main gas has little influence on the fluid flow characteristics and the impact agitation capacity of coherent jet, which can be negligible in EAF industrial production. Table 7.1 shows the reactions between CO2 and [C], [Si], [Mn] and Fe in molten steel at steelmaking temperature. Based on the thermodynamics data of these reactions, it can be found that CO2 can react with bath elements at steelmaking temperature. At the earlier stage of blowing process in EAF steelmaking, especially in EAF steelmaking with hot metal charging, the reactions between CO2 and [Si] or [Mn] primarily occur, which generates only about 30 pct of heat output of the reactions between O2 and [Si] or [Mn] in molten steel. At the middle and later smelting stages, CO2 would react with [C] or Fe and these reactions between CO2 and [C] or Fe are endothermic. On these grounds, it can be concluded that CO2 injected into the molten bath can take place of partial O2 to decrease the consumption of O2 and lessen heat release generated in the high-temperature reaction zone during EAF steelmaking. Hence, coherent jets with COMI can reduce the fire-spot temperature and as a result, the evaporation of molten steel and the dust generation can be weakened. According to the measurement in practice by Takamasa et al. [19], the temperature of fire-spot reaction zone with COMI can be obtained according to the calculation of material and heat balance. Figure 7.2 shows the calculation temperature of fire-spot with different reaction proportion of CO2 . The calculated temperature of fire-spot reaction zone decreases with the reaction proportion of CO2 increasing. When the reaction proportion of CO2 is more than 5.0 pct, the temperature of fire-spot will

7.2 Coherent Jets with COMI

135

Fig. 7.1 Total temperature and CO2 distributions of coherent jets with COMI at steelmaking temperature (1700 K) on longitudinal section. a The main gas contains 100 volume pct O2 ; b The main gas contains 75 volume pct O2 and 25 volume pct CO2

Table 7.1 Thermodynamics data of reactions between CO2 /O2 and molten steel Gas CO2

O2

Chemical reaction ΔGº (J/mol)

T ≥ 1523 K

ΔH 1773K (kJ/kg)

2CO2(g) + [Si] = (SiO2 ) + 2CO(g)

− 3,577,967 + 357.27 T

ΔGº