Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies (The Minerals, Metals & Materials Series) 3030368297, 9783030368296

Citation preview

Energy Technology 2020 Recycling, Carbon Dioxide Management, and Other Technologies

EDITED BY

Xiaobo Chen Yulin Zhong Lei Zhang John A. Howarter Alafara Abdullahi Baba Cong Wang Ziqi Sun

Mingming Zhang Elsa Olivetti Alan Luo Adam Powell

The Minerals, Metals & Materials Series

Xiaobo Chen Yulin Zhong Lei Zhang John A. Howarter Alafara Abdullahi Baba Cong Wang Ziqi Sun Mingming Zhang Elsa Olivetti Alan Luo Adam Powell •

•

•

•

•

•

•

•

•

•

Editors

Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies

123

Editors Xiaobo Chen Royal Melbourne Institute of Technology Melbourne, VIC, Australia

Yulin Zhong Griffith University Queensland, Australia

Lei Zhang University of Alaska Fairbanks, AK, USA

John A. Howarter Purdue University West Lafayette, IN, USA

Alafara Abdullahi Baba University of Ilorin Ilorin, Nigeria

Cong Wang Northeastern University Shenyang, China

Ziqi Sun Queensland University of Technology Brisbane, QLD, Australia

Mingming Zhang ArcelorMittal Global R&D Schererville, IN, USA

Elsa Olivetti Massachusetts Institute of Technology Cambridge, MA, USA

Alan Luo The Ohio State University Columbus, OH, USA

Adam Powell Worcester Polytechnic Institute Worcester, MA, USA

ISSN 2367-1181 ISSN 2367-1696 (electronic) The Minerals, Metals & Materials Series ISBN 978-3-030-36829-6 ISBN 978-3-030-36830-2 (eBook) https://doi.org/10.1007/978-3-030-36830-2 © The Minerals, Metals & Materials Society 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, 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 publisher, 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 publisher 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 publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

This volume contains selected papers presented at two symposia organized in conjunction with the TMS 2020 Annual Meeting & Exhibition in San Diego, California, USA: • Energy Technologies and CO2 Management (sponsored by the TMS Energy Committee) • Recycling of Secondary, Byproduct Materials and Energy (sponsored by the TMS Recycling and Environmental Technologies Committee) The papers in this volume intend to address the issues, intricacies, and the challenges relating to energy and environmental sciences. The Energy Technologies and CO2 Management Symposium was open to participants from both industry and academia and focused on energy efficient technologies including innovative ore beneficiation, smelting technologies, recycling, and waste heat recovery. Topics cover various technological aspects of sustainable energy ecosystems and processes that improve energy efficiency, reduce thermal emissions, and reduce carbon dioxide and other greenhouse emissions. Papers addressing renewable energy resources for metals and materials production, waste heat recovery and other industrial energy efficient technologies, new concepts or devices for energy generation and conversion, energy efficiency improvement in process engineering, sustainability and life cycle assessment of energy systems, as well as the thermodynamics and modeling for sustainable metallurgical processes are included. This volume also includes topics on CO2 sequestration and reduction in greenhouse gas emissions from process engineering, sustainable technologies in extractive metallurgy, as well as the materials processing and manufacturing industries with reduced energy consumption and CO2 emission. Contributions from all areas of non-nuclear and non-traditional energy sources, such as solar, wind, and biomass, are also included in this volume. The Recycling of Secondary, Byproduct Materials and Energy Symposium provided a forum for papers exploring the valorization of materials and their embodied energy including byproducts or coproducts from ferrous and non-ferrous industries, batteries, electronics, and other complex secondary materials. Although v

vi

Preface

most recycling only involves mechanical and physical manipulation of matter, which typically require one to two orders of magnitude less energy than the chemical manipulation involved in primary production, recycling processes still require energy to operate, and they do not run emission-free. Recycling processes must be designed to deal with materials that are potentially quite different from the original base material. However, there has been a significant mismatch between the technical needs for responsible treatment of secondary, byproduct materials, and embodied energy of materials and the ability to achieve economically feasible and sustainable operations. These materials and their embodied energy are generally low value and can be quite complex due to the significant variation in properties leading to potential mismatch among complexity, regulations, and available resources. The papers included in this volume can provide readers a broad perspective on both the technical as well as policy-based challenges. We hope this volume will serve as a reference to materials scientists and engineers as well as metallurgists for exploring innovative energy technologies and novel energy materials processing. We would like to acknowledge the contributions from the authors of the papers in this volume, the effort of the reviewers involved with the manuscripts review process, and the help received from the publisher. We also acknowledge the organizers of both symposia contributing the papers to this volume.

Energy Technologies and CO2 Management Symposium Organizers Xiaobo Chen, RMIT University Yulin Zhong, Griffith University Lei Zhang, University of Alaska Fairbanks John A. Howarter, Purdue University Alafara Abdullahi Baba, University of Ilorin Cong Wang, Northeastern University Ziqi Sun, Queensland University of Technology

Recycling of Secondary, Byproduct Materials and Energy Symposium Organizers Mingming Zhang, ArcelorMittal Global R&D Elsa Olivetti, Massachusetts Institute of Technology Alan Luo, The Ohio State University Adam Powell, Worcester Polytechnic Institute

Contents

Part I

Energy Technologies and CO2 Management Symposium

The Impact of Solar Resource Characteristics on Solar Thermal Pre-heating of Manganese Ores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lina Hockaday, Tristan McKechnie, Martina Neises von Puttkamer and Matti Lubkoll

3

The Compatibility of Metallic Thermal Storage Materials and Housing Materials: A Computational Survey and Accelerated Reaction Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anthony Joseph Rawson, Tina Gläsel, Benedikt Nowak, David Boon, Veronika Stahl and Florian Kargl

15

Effect of Fly Ash from Coal-fired Boiler on Heat Transfer Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jiapeng Liang, Haibin Zuo, Yingli Liu and Shenhui Liu

31

Optimization and Management of On-Site Power Plants Under Time-of-Use Power Price: A Case Study in Steel Mill . . . . . . . . . . . . . . Xiancong Zhao, Huanmei Yuan, Zefei Zhang and Hao Bai

39

Economic Metals Rescue from Spent Zinc–Carbon Batteries for Industrial Value Additions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Alafara A. Baba, Folahan A. Adekola, Rafiu B. Bale, Abdul G. F. Alabi and Mustapha A. Raji

49

Characterization of the Hot-Pressed Coal Briquettes Prepared with the HyperCoal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yajie Wang, Haibin Zuo, Kaikai Bai, Jun Zhao and Jiansheng Chen

57

The Co-extraction of Low-Rank Coal and Biomass by Polar Solvent at Mild Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jun Zhao and Haibin Zuo

69

vii

viii

Contents

Discussion on the Application of Rooftop Photovoltaic Power Plant in the Steel Enterprise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiancong Zhao, Huanmei Yuan, Yuzhao Han, Zefei Zhang and Hao Bai

79

Performance of Anodes with Proper Active Metal Elements Added to the Al–0.16wt%In in Alkaline Electrolyte for Al-Air Batteries . . . . . Huimin Lu, Neale Neelameggham, Leng Jing and Jianxue Liu

89

Theoretical and Experimental Research on the Mass Changes of Elements in Molten Steel with CO2 Used as RH Lifting Gas . . . . . . . Baochen Han, Rong Zhu, Guangsheng Wei, Chao Feng and Jianfeng Dong

99

Hydrogen as a Fuel and Ramifications . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Ashok Khandkar and Neale R. Neelameggham Applying Biochar Composite Briquette for Energy Saving in Blast Furnace Ironmaking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Kai Fan, Zi Yu and Huiqing Tang Properties and Microstructure of Copper and/or Nickel Supported on GO, rGO, and NGO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Xiangyong Lv, Guangfen Liang, Yandong Li, Huamei Duan, Dengfu Chen and Mujun Long Investigation of H2 Addition Effects on CO/CO2/H2-Air Flames by a Combustion Diagnostic System Based on TDLAS . . . . . . . . . . . . . 137 Yu Liu, Jingsong Wang, Qingguo Xue, Haibin Zuo and Xuefeng She MnOx-Decorated Fe–Zr-Based Nano-Catalysts for Low-Temperature NH3-SCR: Improvement of Catalytic Activity . . . . . . . . . . . . . . . . . . . . 147 Chen Yang, Jian Yang, Qingrui Jiao, Yuanmeng Tian, Qingcai Liu, Shan Ren and Jiangling Li Part II

Recycling of Secondary, Byproduct Materials and Energy

Toward 100% Recycling of Steelmaking Offgas Solid Wastes by Reallocating Zinc-bearing Materials . . . . . . . . . . . . . . . . . . . . . . . . . 159 Naiyang Ma Granulation and Carbonization Process of Titanium-Bearing Blast Furnace Slag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 Mingrui Yang, Gangqiang Fan, Feifei Pan, Jie Dang, Xuewei Lv and Chenguang Bai Study on Cracking Control of Cold Bonded Pellets Containing Converter Dust Based on Nonhydraulic Hardening Principle . . . . . . . . 179 Xiang Li, Ping Tang, Xueqin Zhu, Pengpeng Qin and Guanghua Wen

Contents

ix

Experimentation, Modeling, and Optimum Conditions of Pyro-Hydrometallurgical-Precipitation Reaction Technology for Recovery of Copper as Oxide of Nanoparticles from a Copper Dust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 A. A. Adeleke, A. P. I. Popoola, O. M. Popoola and D. O. Okanigbe Recovery Nickel-Ferrous Compound from Nickel-Bearing Secondary Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Qiuju Li, Songyan Liu, Caixiang Yu, Fanxi Yang and Ziyang Wang Extraction and Processing of Crystalline Metallurgical-Grade Silicon Prepared from Rice Husk Byproduct . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 C. Iyen, B. O. Ayomanor and V. Mbah Separation and Recovery of Copper from Copper-Bearing Pyrite Cinder via an Acid Leaching Process . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Yikang Tu, Zijian Su, Manman Lu, Yuanbo Zhang and Tao Jiang A Recycling System for Sustainable Management of Waste Solar Photovoltaic Panels in Taiwan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 Esher Hsu and Chen-Ming Kuo Controllable Synthesis of Battery-Grade Iron Oxalate with Waste Ferrous Sulfate from Titanium Dioxide Production . . . . . . . . . . . . . . . . 249 Keyu Zhang, Yin Li, Runhong Wei, Yunke Wang, Yongnian Dai and Yaochun Yao Mechanical Beneficiation of End-of-Life Lithium-Ion Battery Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259 Haruka Pinegar and York R. Smith Assessing the Techno-Economic Feasibility of Solvent-Based, Critical Material Recovery from Uncertain, End-of-Life Battery Feedstock . . . . 269 Chukwunwike O. Iloeje, Yusra Khalid, Joe Cresko and Diane J. Graziano Thermodynamic Analysis and Reduction of Anosovite with Methane at Low Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 Run Zhang, Gangqiang Fan, Mingbo Song, Chaowen Tan and Jie Dang Recovering Plastics from Electronics Waste . . . . . . . . . . . . . . . . . . . . . . 295 Brian Riise Additive Manufacturing via the Direct Ink Writing Technique of Kaolinite-Based Clay with Electric Arc Furnace Steel Dust (EAF Dust) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 Edisson Ordoñez and Henry A. Colorado

x

Contents

Characterization of Wasted LEDs from Tubular Lamps Focused on Recycling Process by Hydrometallurgy . . . . . . . . . . . . . . . . . . . . . . . 317 Rafael Piumatti Oliveira, Amilton Barbosa Botelho Junior and Denise Crocce Romano Espinosa Comprehensive Utilization of Vanadium Extraction Tailings: A Brief Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 Xin Wang, Junyi Xiang, Jiawei Ling, Qingyun Huang and Xuewei Lv Crystallization and Carbonization of TiO2–CaO–SiO2 Ternary Slag . . . 335 Gangqiang Fan, Jundan Tan, Run Zhang, Jie Dang, Chenguang Bai, Huxu Lei and Chaowen Tan Gravity Separation of Zinc Mine Tailing Using Wilfley Shaking Table to Concentrate Hematite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347 Jonathan Tenório Vinhal, Raquel Húngaro Costa, Amilton Barbosa Botelho Junior, Denise Crocce Romano Espinosa and Jorge Alberto Soares Tenório Minimization of Copper Contamination in Steel Scrap . . . . . . . . . . . . . 357 Hyunsoo Jin and Brajendra Mishra Recycling of Blast Furnace Flue Dust with In-flight Reduction Technology: Reduction Behavior and Kinetic Analysis . . . . . . . . . . . . . 365 Jin Xu, Nan Wang, Min Chen and Haiyang Yu Recycling Technologies of Zn–C Batteries: Review and Challenges for a Circular Economy in Colombia . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 Natalia Cardona-Vivas, Mauricio A. Correa and Henry A. Colorado Selective Recovery of Lithium from Ternary Spent Lithium-Ion Batteries Using Sulfate Roasting-Water Leaching Process . . . . . . . . . . . 387 Chang Di, Chen Yongming, Xi Yan, Chang Cong, Jie Yafei and Hu Fang Separation and Recovery of Zinc and Cobalt from Zinc Plant Residue by Alkali Leaching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397 Yangbo Geng, Guihong Han, Yukun Huang, Zuoqi Ma, Yanfang Huang and Weijun Peng Study on Mineral Phase Composition and Viscosity of Hot Metal Pretreatment Desulfurization Slag Based on FactSage . . . . . . . . . . . . . . 405 Tengfei Ma, Wufeng Jiang, Suju Hao and Yuzhu Zhang Study on Oxidation Roasting Process of Cathode Sheets from Spent Lithium Ion Batteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413 Xi Yan, Chen Yongming, Chang Di, Jin Wei and Jie Yafei Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 427 Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 431

About the Editors

Xiaobo Chen was awarded a Ph.D. in Materials Science and Engineering from Deakin University (2010), worked as a postdoctoral research fellow and ARC DECRA senior research fellow at Monash University (2009–2017), and now is a vice chancellor senior research fellow in the School of Engineering at RMIT University, Melbourne, Australia. His research is multidisciplinary and spans from chemistry and materials science to corrosion, electrochemistry, and biomaterials. His research aims to provide functional characteristics of the surface of light metals to satisfy a large range of engineering applications in automotive, 3C, and biomedical industries. Yulin Zhong completed his Ph.D. in Chemistry (2010) at the National University of Singapore (NUS) and did his postdoctoral training at Princeton University, Massachusetts Institute of Technology, and Monash University. He was awarded an ARC DECRA fellowship in 2014 and joined Griffith University as a senior lecturer in 2016. His research group interests include electrochemical production of 2D nano-materials, 3D printing, smart windows, and wearable devices.

xi

xii

About the Editors

Lei Zhang is an associate professor in the Department of Mechanical Engineering at the University of Alaska Fairbanks (UAF). Prior to joining the UAF, She worked as a postdoctoral associate in the Department of Chemical and Biomolecular Engineering at the University of Pennsylvania. She obtained her Ph.D. degree in Materials Science and Engineering from Michigan Technological University in 2011, and her M.S. and B.E. degree in Materials Science and Engineering from China University of Mining and Technology, Beijing, China, in 2008 and 2005, respectively. Her current research mainly focuses on the synthesis of metal-organic frameworks (MOFs) and MOF-based nano-composites, and the manipulation of their properties and applications in gas storage, separation, and water treatment. She is also working on the development and characterization of anti-corrosion coatings on metallic alloys for aerospace and biomedical applications. She has served on the TMS Energy Committee since 2014, including the vice-chair role in 2018–2019, and served on a Best Paper Award Subcommittee of the committee. She has served as a frequent organizer and session chair of TMS Annual Meeting symposia (2015– present). She was the recipient of the 2015 TMS Young Leaders Professional Development Award. John A. Howarter is an associate professor in Materials Engineering at Purdue University with a joint appointment in Environmental & Ecological Engineering. His research interests are centered on synthesis, processing, characterization, and end-of-life fate of sustainable polymers and nano-composites. His research impacts water treatment, thermal management in electronic devices, and material design for recycling and value recovery. He has been involved in the Public and Governmental Affairs (P&GA) Committee of TMS, serving as the chair from 2017 to 2020. Dr. Howarter earned a B.S. from The Ohio State University in 2003 and Ph.D. from Purdue University in 2008, both in Materials Engineering. From 2009 to 2011, he was a National Research Council postdoctoral scholar in the Polymers Division of the National Institute of Standards and Technology in Gaithersburg, Maryland.

About the Editors

xiii

Alafara Abdullahi Baba is a professor of Analytical/Industrial and Materials Chemistry in the Faculty of Physical Sciences, University of Ilorin, Nigeria. He holds a Ph.D. degree in Chemistry from the University of Ilorin in 2008. His dissertation titled Recovery of Zinc and Lead from Sphalerite, Galena and Waste Materials by Hydrometallurgical Treatments was judged the best in the area of Physical Sciences at University of Ilorin, Nigeria, in 2010. Until his current appointment as a head, Department of Industrial Chemistry in 2017, he was a deputy director—Central Research Laboratories, University of Ilorin (2014– 2017). He is a fellow of the Chemical Society of Nigeria (FCSN) and Materials Science & Technology Society of Nigeria (FMSN), is currently the secretary of the Hydrometallurgy and Electrometallurgy Committee of the Extraction and Processing Division (EPD) of The Minerals, Metals & Materials Society (TMS), is a co-organizer of the Rare Metal Extraction & Processing Symposium and Energy Technologies & Carbon Dioxide Management Symposium at the TMS Annual Meeting and Exhibition, and is on the TMS Materials Characterization, Education, and EPD Awards committees. He has keen interest in teaching, community services, and research covering solid minerals and materials processing through hydrometallurgical routes; reactions in solution and dissolution kinetic studies; and preparation of phyllosilicates, porous, and bio-ceramic materials for industrial value additions. He has 113 publications in nationally and internationally acclaimed journals of high impact, and has attended many national and international workshops, conferences, and research exhibitions to present his research breakthroughs. He is the recipient of several awards and honors including 2015 MISRA AWARD of the Indian Institute of Mineral Engineers (IIME) for the best paper on Electro-/Hydro-Bio-Processing at the IIME International Seminar on Mineral Processing Technology—2014 held at Andhra University, Visakhapatnam, India; 2015 MTN Season of Surprise Prize as Best Lecturer in the University of Ilorin— Nigeria category; Award of Meritorious Service in recognition of immense contributions to the

xiv

About the Editors

Development of the Central Research Laboratories, University of Ilorin, Nigeria (2014–2017); and 2018 Presidential Merit Award in Recognition of Passion, Outstanding and Selfless Service to the Materials Science and Technology Society of Nigeria. Cong Wang is a professor in the School of Metallurgy, Northeastern University, China. Prior to joining the faculty of his alma mater, he worked in Northwestern University, Saint-Gobain, and Alcoa, all in the USA. He obtained his Ph.D. from Carnegie Mellon University, M.S. from Institute of Metal Research, Chinese Academy of Sciences, and B.S. (with honors) from Northeastern University. He is now leading a group dedicated to oxide metallurgy. He is an active member and a prolific scholar in the global metallurgy community. He has been recognized with distinctions such as TMS Early Career Faculty Fellow Award, CSM Youth Metallurgy S&T Prize, Newton Advanced Fellowship, JSPS Invitational Fellowship, TÜBİTAK Fellowship, SME Outstanding Young Manufacturing Engineer Award, and ASM Silver Medal. He serves as a key reader and vice-chair for the Board of Review for Metallurgical and Materials Transactions B; review editor for Journal of Materials Science and Technology; editorial board member of International Journal of Refractory Metals and Hard Materials and Journal of Iron and Steel Research, International; and corresponding expert for Engineering. He chaired the TMS Energy Committee from 2016 to 2017. He is the inaugural chair for the ASM Shenyang Chapter, and faculty advisor for Material Advantage Northeastern University. He initiated the International Metallurgical Processes Workshop for Young Scholars (IMPROWYS) and organized major conferences/ symposia of technical significance.

About the Editors

xv

Ziqi Sun is currently an associate professor and ARC future fellow in the Queensland University of Technology (QUT), Australia. His research interest includes developing bio-inspired smart nano-materials and 2D metal oxide nano-materials for sustainable energy and environmental applications, such as rechargeable batteries, oil-water separations, and catalysis. He received his Ph.D. degree in Materials Science and Engineering from the Institute of Metal Research, Chinese Academy of Sciences in 2009. After one year of experience as the NIMS postdoctoral fellow (Japan) on solid oxide fuel cells, he joined the University of Wollongong (UOW), Australia, in 2010 and moved to QUT as a faculty member in 2015. He has served in some prestigious leadership roles in both the academic and professional communities, such as the chair of the TMS Energy Committee; Editor-in-Chief of Sustainable Materials and Technologies (CiteScore = 8.43, Elsevier); handling editor of Physics Open (Elsevier); principal editor of Journal of Materials Research (MRS); associate editor of Surface Innovations (ICE); and editorial board member of Scientific Reports, Journal of Materials Science and Technology, and Nano Materials Science. Mingming Zhang is a lead research engineer at ArcelorMittal Global R&D in East Chicago, Indiana. He has more than 15 years of research experience in the field of mineral processing, and metallurgical and materials engineering. He obtained his Ph.D. degree in Metallurgical Engineering from The University of Alabama and his master’s degree in Mineral Processing from General Research Institute for Non-ferrous Metals in China. Prior to joining ArcelorMittal, he worked with Nucor Steel in Tuscaloosa, Alabama, where he was a metallurgical engineer leading the development of models for simulating slab solidification and secondary cooling process. He has conducted a number of research projects involving mineral beneficiation, thermodynamics and kinetics of metallurgical reactions, electrochemical processing of light metals, metal recycling, and energy efficient and environmentally cleaner technologies. He has published more than 50 peer-reviewed research papers, and he is the recipient of

xvi

About the Editors

several US patents. He also serves as editor or reviewer for a number of prestigious journals including Metallurgical and Materials Transactions A and B, JOM, Journal of Phase Equilibria and Diffusion, and Mineral Processing and Extractive Metallurgy Review. He has made more than 30 research presentations at national and international conferences including more than ten keynote presentations. He was the recipient of the 2015 TMS Young Leaders Professional Development Award. He has served as a conference/ symposium organizer and technical committee chair in several international professional organizations including The Minerals, Metals & Materials Society (TMS), Association for Iron & Steel Technology (AIST), and the Society for Mining, Metallurgy & Exploration (SME). Elsa Olivetti is the Atlantic Richfield associate professor of Energy Studies in the Department of Materials Science and Engineering at the Massachusetts Institute of Technology (MIT). Her research focuses on improving the environmental and economic sustainability of materials using methods informed by materials economics, machine learning, and techno-economic analysis. She has received the NSF Career award for her experimental research focused on beneficial use of industrial waste materials. She received her B.S. degree in Engineering Science from the University of Virginia. Her Ph.D. in Materials Science and Engineering from MIT was focused on the development of cathode materials for lithium ion batteries. Alan Luo is a professor of Materials Science and Engineering and Integrated Systems Engineering (Manufacturing) at The Ohio State University (OSU) in Columbus, OH, USA. He is also the director of OSU Light Metals and Manufacturing Research Laboratory (LMMRL) and on the steering board of OSU Center for Simulation Innovation and Modeling (SIMCenter). He is an elected fellow of American Society of Metals (ASM) International and Society for Automotive Engineers (SAE) International. He served on the Board of Directors of TMS (The Minerals, Metals & Materials Society) and as chair of its Light Metals Division. He is also a director of the board of

About the Editors

xvii

International Magnesium Association (IMA) and serves as chair of its annual conference program chair. He is a past chair of SAE Materials Engineering Activities. He has 20 patents and more than 240 technical publications on advanced materials and manufacturing, specializing in lightweight materials and applications. Prior to joining OSU in July 2013, He was a GM technical fellow at General Motors Global Research and Development Center (Warren, MI, USA) with 20 years of industrial experience. He has 20 patents and more than 240 technical publications in advanced materials, manufacturing, and applications. He won two John M. Campbell Awards for his fundamental research, and three Charles L. McCuen Awards for research applications at GM. Over the years, he has received the TMS Brimacombe Medalist Award, SAE Forest R. McFarland Award, United States Council for Automotive Research (USCAR) Special Recognition Award, ASM Materials Science Research Silver Medal, and International Magnesium Association (IMA) Award of Excellence in cast projects. His research is also recognized by several Best Paper awards from TMS, SAE, and AFS (American Foundry Society). His research in sustainable energy and sustainable resources includes (1) lightweight materials (aluminum, magnesium, titanium and high-entropy alloys, bio-metals, super-wood, and metal matrix nano-composites); (2) advanced manufacturing processes (casting, forming, additive, and multi-material manufacturing); and (3) lightweight design and integrated computational materials engineering (ICME). Adam Powell is an associate professor in the Mechanical Engineering department who joined the WPI faculty in August 2018. His field is materials processing, and research focuses on validated mathematical modeling of metal process development for clean energy and energy efficiency. His research group is developing new projects whose goals are to reduce vehicle body weight, lower solar cell manufacturing cost with improved safety, reduce or eliminate environmental impact of aerospace emissions, and improve grid stability with up to 100% clean energy.

xviii

About the Editors

His research has resulted in 68 publications across materials classes: metal extraction/refining and product development, thin films, ceramic coatings, polymer membranes, batteries, and electromagnetic propulsion. He is the author of nine open-source computational tools in materials processing, microstructure, and thermodynamics modeling.

Part I

Energy Technologies and CO2 Management Symposium

The Impact of Solar Resource Characteristics on Solar Thermal Pre-heating of Manganese Ores Lina Hockaday, Tristan McKechnie, Martina Neises von Puttkamer and Matti Lubkoll

Abstract The proposed paper evaluates an alternative ferromanganese production flowsheet seeking to pre-heat manganese ores with concentrating solar thermal energy to 600 ◦ C. The benefits of solar thermal pre-heating will be evaluated based on a cost discounted economic model taking into account the variability of the solar resource, capital costs, and operating costs of a solar thermal plant over the lifetime of the project. Solar variability will be discussed based on possible implementation sites for such technologies, and the cost and benefits of thermal storage in the flowsheet will also be evaluated. This work is part of the PreMa project, aiming to advance novel energy systems in the drying and pre-heating of furnace materials. The PreMa project has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No 820561. Keywords Concentrating solar thermal · Pre-heating · Ferromanganese production

L. Hockaday (B) Mintek, 200 Malibongwe Drive, Praegville 2094, Randburg, South Africa e-mail: [email protected] T. McKechnie Solar Thermal Energy Research Group (STERG), Stellenbosch 7600, South Africa e-mail: [email protected] M. N. von Puttkamer · M. Lubkoll Institute of Solar Research, German Aerospace Center (DLR), Pfaffenwaldring 38-40, 70569 Stuttgart, Germany e-mail: [email protected] M. Lubkoll e-mail: [email protected]

© The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_1

3

4

L. Hockaday et al.

Introduction Manganese is an important additive to steel. Manganese content in steel improves toughness and wear resistance of steel, and on average about 0.8% manganese is added to steel. 90% of manganese is used as steel additive in the form of ferromanganese alloys. Ferromanganese alloys are produced in either blast furnaces or electric arc furnaces with carbon as a reductant. Detailed description of manganese ferroalloy production, both for high carbon ferromanganese and low carbon silicomanganese alloys can be found in literature [11, 19]. Global manganese ore mine production is summarized in Table 1 as adapted from [21]. The Republic of South Africa (RSA) is the leading producer of manganese ores and has the largest land based manganese ore reserves. Global manganese ferroalloy production which includes different grades of ferromanganese and silicomanganese is given in Table 2 as adapted from [10]. The PreMa project [17] aims to investigate the optimal pre-heating option for a high carbon ferromanganese furnace in order to reduce electricity consumption and greenhouse gas emission from manganese ferroalloy production [8]. Although

Table 1 Global mine production and reserves of manganese ores by country, manganese content Year

Unit RSA

2017

kt/a

5400

735

1160

2820

2190

1700

1278

2018

kt/a

5500

740

1200

3100

2300

1800

1342

18,000

230,000

140,000

110,000

99,000

65,000 54,000

62,000

760,000

Reserves kt

Ukraine

Brazil

Australia Gabon

China

Other

World total 17,300

Table 2 Manganese ferroalloy production by country, based on manganese content. China was the largest manganese ferroalloy producer, with production being four times more than India and ten times more than South Africa. Norway and Spain were the largest European producers of manganese ferroalloys Country Production (000 mt) China India RSA Ukraine South Korea Norway Japan Russia Australia Spain Other World total

10,349 2372 741 713 686 608 483 352 254 243 1447 18,249

The Impact of Solar Resource Characteristics …

5

the project also investigates pre-heating with furnace off-gas, bio-carbon, and fossil carbon, this paper focuses on the novel use of concentrating solar thermal energy as the energy source for pre-heating. The cost of using concentrating solar thermal process heat is dependent on the available solar resource at the location it is captured, as well as the technology choices selected. This paper studies three possible locations for concentrating solar thermal plants in proximity to current manganese ferroalloy smelters, as well as one location near manganese ore mines. It was attempted to select locations with existing smelters and good solar radiation in Europe, Africa, and China. The locations selected for evaluation are listed below. These locations were not selected as ideal sites, for example China has locations with better solar resources in the Inner Mongolia Province, but is evaluated to provide insight into the factors involved in the application of solar thermal process energy to a hightemperature industrial process. • • • •

Jiayuguan, Gansu Province, China Huesca, Spain Hotazel, Northern Cape Province, South Africa (RSA 1) Emalahleni, Mpumalanga Province, South Africa (RSA 2).

Manganese Ferroalloy Production Process Modeling To investigate the energy demand for pre-heating of manganese ores, a HSC model [14, 15], Version 9.9.2.3, was constructed for the PReMA project. The HSC model is based on the possible reactions that can take place during pre-heating and smelting and the extent they progress towards completion. Traditional pre-heating systems rely on fossil fuel combustion, and a reducing atmosphere with low partial pressures of oxygen is generally practiced [20]. The novel solar thermal pre-heating unit relies on heated air, maintaining an oxidative atmosphere in the unit, and therefore, the reactions differ from those expected in a reducing atmosphere and are given in Eqs. 1–8. Equation 7 is the Boudouard reaction where carbon dioxide reacts with carbon to form carbon monoxide. This reaction is likely to start taking place at temperatures above 500 ◦ C and to proceed fully only at temperatures above 800 ◦ C. Similarly, this preliminary investigation has been guided by calculated equilibrium reactions for the thermal decomposition of MnO2 to Mn2 O3

Table 3 Illustrative modeling assumptions and resulting energy demand for the pre-heater Reaction 1 2 3 4 5 6 7 8 Completion (%) 100 80 100 100 100 100 0 100 Pre-heating target, ◦ C Pre-heater energy demand, kWh/t feed Process CO2 emission factor, t/t alloy

600 339.8 2.31

6

L. Hockaday et al.

as published by [19, p. 74]. Future work will involve the determination of kinetics for these reactions. The completion of these reactions will influence the final energy demand of the pre-heater, and values in Table 3 are for illustrative purposes only. The results from process modeling for pre-heating to 600 ◦ C are shown in Table 3. The process CO2 emission factor of 2.31 is a reduction in 7% on the emissions factor for a process not employing pre-heating. The energy demand for a pre-heater feeding a 30MW high carbon ferromanganese furnace that requires a manganese ore feed of approximately 40t/h will therefore have an energy demand of 13.6MW to achieve a product temperature of 600 ◦ C. Due to the variable nature of the solar resource, a solar thermal plant will only be able to meet this demand in part. The following section describes the methodology to size a solar thermal plant, with thermal storage to improve availability and electrical heating as back-up for the four different locations identified as possible sites. Electrical heating was chosen as back-up technology due to the increase in zero emission electricity options available to industry [16]. Using electricity as back-up rather than a fossil fuel also prevents pre-heating cycling between an oxidizing and a reducing environment, which may lead to problems with control of the carbon balance in the submerged arc furnace (SAF). MgCO3 −→ MgO + CO2 (g) CaMg(CO3 )2 −→MgO + CaCO3 + CO2 (g) H2 O(l) −→ H2 O(g)

(1) (2) (3)

2FeO · OH −→ Fe2 O3 + H2 O(g) MnCO3 −→ MnO + CO2 (g)

(4) (5)

2AlO(OH) −→ Al2 O3 + H2 O(g) C + CO2 (g) −→ CO2 (g)

(6) (7)

4MnO2 −→ 2Mn2 O3 + O2 (g)

(8)

Solar Thermal Plant Modeling Methodology In recent years, solar thermal technology has advanced through the development of solid particle receivers [5, 6]. Solid particle receivers operate with the solid particles directly exposed to the concentrating solar flux. The layer of solid particles in the Centrec® receiver shields the rotating structure of the receiver and makes possible particle temperatures in excess of 900 ◦ C [2]. Figure 1 shows a schematic of a CST plant that would provide high-temperature process heat to an industrial process as envisaged in the PreMa project [17]. The purpose of this section is to compare the effect of solar resource variability on the potential for incorporating concentrating solar thermal (CST) technologies in manganese ore pre-heating. The integration of CST technologies is envisioned to lead to lower energy costs and significant reductions in carbon emissions, as already presented in Section Manganese Ferroalloy Production Process Modeling. The CST

The Impact of Solar Resource Characteristics …

7

Fig. 1 Concentrating solar thermal technologies

plant model assumes the German Aerospace Center’s (DLR) particle receiver technology [2], CentRec® , for receiver and thermal energy storage. For the purpose of this assessment, their receiver sizing and performance characteristics are based on [1]. The model follows [12] with the thermal receiver size fixed at 1 m2 and 2.5 MWt output. The solar field is then sized to provide a system output of 2.5 MWt at solar equinox. Each such CST tower system can then at best provide 2.5 MWt peak to a consumer. Multiple CST towers are foreseen to be deployed when the heat demand exceeds the supply of one tower. The lowest levelized cost of heat, LCOH, of a CST system is usually found with the solar components being significantly over-sized compared to the thermal demand. This over-sizing permits thermal storage and is expressed through the solar multiple, defined as follows: SM =

Q rec , Q process

(9)

where Q rec is the thermal output of the receiver at the solar field design point, and Q process is the thermal output to process. The solar plant annual performance assessment is conducted by modeling at hourly steady state conditions. The solar resource data is obtained as typical meteorological year (TMY) from Meteonorm [13], Version 7.3. Details regarding the plant and economic modeling and model inputs are explained further in the paper of [12]. The solar plant operation was simulated to determine the energy produced, from which the LCOH was determined. A parametric study was then performed to determine the most suitable solar plant configuration to obtain the lowest LCOH for each site.

Field Layout The positioning of the tower within the heliostat field was investigated, resulting in improved optical efficiency for the solar field. This is an improvement on the

8

L. Hockaday et al.

(a) Tower at center

(b) Tower at 30%

(c) Tower at 60%

(d) Tower at 90%

Fig. 2 Tower position optimization

Fig. 3 Tower position within field and resulting field optical efficiency

methodology described by [12]. The receiver is modeled angle downward 45◦ from the horizontal. This allows heliostats placed behind the tower to have line of sight to the receiver opening. Heliostats placed near and behind the tower have improved optical efficiency compared to heliostats in front of the tower but further away. The optimal placement of the tower within the field was determined as presented in Fig. 2. Figure 3 shows the resulting field optical efficiency. It can be seen that the tower placed at 0.6 × rfield from the center of the field resulted in the maximum field optical efficiency. These results agree with those of [7]. All fields sized in this paper will therefore have a layout similar to Fig. 2c. The specification of the heliostat field now allows the capacity factor (CF) to be calculated. The capacity factor is defined as the average annual energy production divided by the process heat demand.

Operating Strategy The configuration of a CST plant producing heat at the lowest levelized cost of heat, LCOH, does not typically have a 100% capacity factor; for this reason back-up electric heaters are included as auxiliary heating for when solar heat is insufficient to meet demand. LCOH is determined by dividing the total costs over the project lifetime

The Impact of Solar Resource Characteristics … Table 4 Plant locations and resulting solar fields Location Units RSA 1 Site data

Solar plant specifications

Latitude Longitude DNI h tower

– 27.240 S – 22.902 E kWh/(m2 a) 2795 m 40

ar ec α ηrec

m2 ◦

%

9

RSA 2

Spain

China

25.886 S 29.123 E 2117

41.926 N 0.183 E 1929

39.897 N 98.318 E 1520

1 45 90

by the total amount of energy supplied over the lifetime of the project. A combined LCOH of solar electric heating was calculated to determine the configuration of the solar plant that results in the lowest produced combined solar electric heat. The cost of electrically generated heat is simplified as the cost per MWh of electricity. The combined solar electric LCOH was calculated as follows: LCOHcomb =

LCOHCST QCST + LCOHel Qel , Qtot

(10)

where QCST is the total annual solar generated heat, Qel is the total annual electrical generated heat, Qtot is the total annual generated heat, LCOHCST is the cost of solar generated heat and LCOHel is the cost of electrically generated heat. The operating strategy for the plant is to deliver the thermal demand whenever the receiver and/or TES has sufficient energy available. At any point when the solar plant does not output the rated thermal demand, then electric heating is supplemented.

Locations Table 4 shows the locations that were assessed and the common solar plant parameters with DNI data from [18] and solar plant specifications as modeled by [12]. The two South African locations have similar latitudes, but differ in the available solar resource. Likewise for the Chinese and Spanish locations, the solar plant specifications listed are h tower tower height, ar ec receiver aperture area, α the receiver tilt angle from the horizontal, and ηrec the receiver solar to particle efficiency.

10

L. Hockaday et al.

Results and Discussion Table 5 summarizes the results from the solar plant modeling and parametric studies. The CST parameters are configured for the lowest annual combined LCOH, incorporating solar with electric back-up for constant heat production. LCOHCST,pot represents an optimized CST only plant configuration and provides reference of the lowest possible solar LCOH. Relative to this, the configuration represented by LCOHCST has significantly higher solar capacity factor to reduce the LCOHcomb by suppressing electricity usage. The economic benefit of increasing the solar capacity factor is more than the added cost for a larger CST systems. This is because for all locations, CST heat is more affordable than electrical heat. Locations with similar latitudes (RSA 1 and RSA 2, and Spain and China) experience similar sun angles throughout the year. As the solar fields are sized for all locations with a common design point DNI, the locations with similar latitudes therefore have similar sized fields. Further, the annual solar field efficiency can be seen to have increased compared to the results from [12], as the tower has been located at an improved location. The Spanish and Chinese locations can be seen to have significantly more expensive electricity, and therefore, favor systems with higher SM and TES size, thereby off-setting the amount of electricity required. Even then, the Chinese location shows relatively moderate CF due to the poor direct normal irradiance (DNI). The South African locations feature less storage, nonetheless show high CF due to high solar resources. RSA 1 is able to achieve a high capacity factor as the solar resource for the location is excellent. RSA 2’s solar plant configuration is not further over-sized as the electricity cost is relatively low. All locations benefit from incorporating CST technologies. The higher the electricity tariff for a location, the larger the solar system will be to suppress electricity use. The benefit of the solar with electric back-up compared to total electrification is shown in the final row of Table 5. Electricity cost data for South Africa was obtained from [4], for Spain from [9], and for China from [3]. The benefit of solar thermal heating as compared to electrification is of course larger for countries with higher electricity costs such as Spain. It should be noted that the cost savings will increase with electrical tariff increases; whereas, the solar heat cost will remain steady over the life of the system. The model in this paper does not include electricity price escalation.

Conclusion This paper evaluated the energy demand for a pre-heater driven by a solar thermal plant providing hot air and backed up by electric heating elements. The aim of the study was to investigate the feasibility of high-temperature solar thermal process heat for pre-heating as a cost effective alternative to electrification of the process as a way of limiting green house gas emissions. The results confirmed that the combined solar

The Impact of Solar Resource Characteristics … Table 5 Solar plant configuration Type Parameters Site data: CST potential: Combined system per tower:

For pre-heater integration:

Benefit versus total electrification a LCOH

11

Units

RSA 1

RSA 2

Spain

China

DNI LCOHel LCOHCST,pot TES

kWh/(m2 a) $/MWh $/MWh h

2795 47.29 35.48 14

2117 47.29 43.43 14

1929 115.51 46.28 22

1520 74.65 56.02 16

Asf ηsf,a SM Q˙ process C FCST LCOHCST LCOHcomb Number of towers

m2 % – MW % $/MWh $/MWh –

3563 65 3.2 0.79 79 36.25 38.55 18

3563 66 3.0 0.82 63 43.85 45.13 17

3616 62 4.5 0.55 76 54.02 68.98 28

3616 62 4.2 0.60 63 57.46 63.75 23

Total heliostat field area LCOHa

ha

6.4

6.1

9.1

8.3

%

19

5

40

17

= (LCOHel − LCOHcomb )/LCOHel

thermal and electric heating produced lower energy costs over a project lifetime of 25 years compared to heating through electrification only for all locations evaluated. Locations with a high annual DNI had lower levelized energy costs than locations with lower annual DNI levels, but the cost of electricity at each location also had an influence on the solar thermal plant design. High electricity costs increased the amount of thermal energy storage and the solar multiple to ensure that the most cost effective solution has a high solar share. Countries with high annual DNI and low electricity costs may in future have a global competitive advantage for low emission, high-temperature process energy applications. The methodology was optimised for heliostat field size, tower position, solar multiple (SM), and thermal energy storage (TES) at each location. The optimization resulted in higher capacity factors than previously published [12] for systems that were not optimized to achieve the lowest LCOH. In conclusion, although combustion heating with fossil fuels such as metallurgical coke and coal remains the least cost alternative at the time of writing, solar thermal process energy can compete favorably with process heating by electrification for projects with a lifetime of 25 years. With industry targets of lowering greenhouse gas emissions becoming more urgent [16], evaluating where solar thermal process energy can be a cost effective alternative is of relevance to industry. Acknowledgements The PreMa project has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No 820561. This paper is published with the permission of Mintek.

12

L. Hockaday et al.

References 1. Amsbeck L, Behrendt B, Prosin T, Buck R (2014) Particle tower system with direct absorption centrifugal receiver for high temperature process heat. Energy Procedia 00 (2015). Elsevier, Beijing, China, pp 000–000 2. Amsbeck L, Buck R, Ebert M, Gobereit B, Hertel J, Jensch A, Rheinländer J, Trebing D, Uhlig R (2018) First tests of a centrifugal particle receiver with a 1m2 aperture. In: AIP Conference Proceedings, AIP Publishing, Santiago, Chile, vol 2033. https://doi.org/10.1063/1.5067040, URL http://aip.scitation.org/doi/abs/10.1063/1.5067040 3. China Briefing News (2019) China electricity prices for industrial consumers: a guide for investors. https://www.china-briefing.com/news/china-electricity-prices-industrialconsumers/ 4. ESKOM (2019) 2019/20 Tariffs and charges. http://www.eskom.co.za/CustomerCare/ TariffsAndCharges/Pages/Tariffs_And_Charges.aspx 5. Gallo A, Roldán MI, Alonso E, Fuentealba E (2016) Considerations for using solar rotary kilns for high temperature industrial processes with and without thermal storage. In: Proceedings of EuroSun2016, International solar energy society, Palma de Mallorca, Spain, pp 1–10. https://doi.org/10.18086/eurosun.2016.02.04, http://proceedings.ises.org/citation? doi=eurosun.2016.02.04 6. Gobereit B, Amsbeck L, Buck R, Pitz-Paal R, Röger M, Müller-Steinhagen H (2015) Assessment of a falling solid particle receiver with numerical simulation. Solar Energy 115:505–517. https://doi.org/10.1016/j.solener.2015.03.013, https://linkinghub.elsevier.com/ retrieve/pii/S0038092X15001310 7. Hallberg M, Hallme E (2018) Introducing a central receiver system for industrial hightemperature process heat applications. PhD thesis, Kth Royal Institute of Technology, Stockholm, Sweden. http://www.diva-portal.org/smash/get/diva2:1303784/FULLTEXT01.pdf 8. Haque N, Norgate T (2013) Estimation of greenhouse gas emissions from ferroalloy production using life cycle assessment with particular reference to Australia. J Clean Prod 39:220– 230. https://doi.org/10.1016/j.jclepro.2012.08.010, https://linkinghub.elsevier.com/retrieve/ pii/S0959652612004179 9. International Energy Agency (2018) Key world energy statistics. https://www.iea.org/statistics/ kwes/prices/ 10. International Manganese Institute (2013) 2013 IMnI Public Report. http://www.manganese. org/files/publications/PUBLIC%20RESEARCH%20REPORTS/2013_IMnI_Public_Report. pdf 11. International Manganese Institute (2019) About Manganese. http://www.manganese.org/ about-manganese/ 12. Lubkoll M, Hockaday SAC, Harms TM (2018) Integrating solar process heat into manganese ore pre-heating. Durban, p 8. https://www.sasec.org.za/full_papers/57.pdf 13. Meteonorm (2019) Intro - Meteonorm (de). https://meteonorm.com/en/ 14. Outotec (2019a) HSC Chemistry. https://www.outotec.com/products/digital-solutions/hscchemistry/ 15. Outotec (2019b) HSC Sim – Process Simulation Module. https://www.outotec.com/products/ digital-solutions/hsc-chemistry/hsc-sim-process-simulation-module/ 16. Philibert C (2017) Renewable Energy for Industry. Tech rep. https://www.iea.org/publications/ insights/insightpublications/Renewable_Energy_for_Industry.pdf 17. Ringdalen E (2019) What is PREMA? URL https://www.spire2030.eu/prema 18. SolarGIS (2019) iMaps. https://solargis.com/maps-and-gis-data/overview/ 19. Sverre E Olsen MT, Lindstad T (2007) Production of manganese ferroalloys. Tapir academic press

The Impact of Solar Resource Characteristics …

13

20. Tangstad M, Ichihara K, Ringdalen E (2015) Pretreatment unit in ferromanganese production. In: The fourteenth international ferroalloys congress, INFACON, Kiev, Ukraine, pp 99–106. https://www.pyrometallurgy.co.za/InfaconXIV/099-Tangstad.pdf 21. US Geological Survey (2019) Mineral Commodity Summaries 2019: U.S. Geological Survey. https://doi.org/10.3133/70202434

The Compatibility of Metallic Thermal Storage Materials and Housing Materials: A Computational Survey and Accelerated Reaction Experiments Anthony Joseph Rawson, Tina Gläsel, Benedikt Nowak, David Boon, Veronika Stahl and Florian Kargl Abstract Metals can provide an energy-dense, high-conductivity solution to the problem of storing heat latently in electric vehicles for space heating. However, many molten metals will react with container materials (e.g. stainless steel) when held for long periods at high temperatures. In this work, computational and experimental methods are introduced and results are presented for the compatibility of the eutectic alloy Al-12.7 wt.% Si with a number of potential container materials. Several promising new container materials are identified from a survey of two CALPHAD databases. Sodium silicide and vanadium silicide were identified as compatible at equilibrium and both viable options as they have been applied as coatings on steel in past work. Experimental results for static pellet compatibility tests for periods of up to two weeks are given for several other materials and are shown to conform to the literature and computational predictions. Recent developments in an experimental apparatus for the simulation of thermal storage materials undergoing erosive-corrosive wear are briefly discussed, providing an outlook for future research at the German Aerospace Centre (DLR). Keywords Thermal energy storage · Metallic phase change material · Compatibility of thermal storage materials · CALPHAD

Introduction Storage of energy as heat is a simple solution to various challenges in intermittency for many technologies, e.g. power generation, air conditioning and temperature sensitive components. Many materials are available for thermal energy storage (TES); however A. J. Rawson (B) · T. Gläsel · B. Nowak · D. Boon · F. Kargl Institute of Material Physics in Space, German Aerospace Centre (DLR), Linder Höhe, 51170 Cologne, Germany e-mail: [email protected] V. Stahl Institute of Vehicle Concepts, German Aerospace Centre (DLR), Pfaffenwaldring 38-40, 70569 Stuttgart, Germany © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_2

15

16

A. J. Rawson et al.

those with low cost and volume per quantum of energy stored are most attractive. One method of greatly increasing energy density is to melt a TES material and store energy in its latent heat of fusion. Materials with a higher melting point tend to have a higher heat of fusion. That is, melting materials at high temperature are ideal from a cost and density perspectives. Passenger vehicles in cold climates require space heating to keep passengers comfortable and for correct operation of auxiliary systems. Increasing battery capacity to accommodate space heating without compromising driving range is an expensive solution. An alternative is to charge a TES system (whilst charging the battery) and release heat in a controlled manner, keeping the passenger space comfortable. Metals are an appropriate TES material for this application due to the small volume available and rapid charging rate requirement [1]. When metals are melted and used for TES they may be called metallic phase change materials (mPCM)s. The metal or alloy selected should ideally have a single melting point such that heat is stored and delivered at a known temperature. This enables optimisation of the heat exchanger and limits spatial separation of the alloy constituents. Metal alloys generally melt over a temperature range, unless they are entirely a pure compound (Al3 Mg2 or Mg2 Si for example) or at a eutectic composition. Eutectic alloys tend to have higher thermal conductivity and be more malleable than metallic compounds, and hence, they are generally preferred for the vehicle application. TES in a vehicle introduces additional challenges (relative to static applications) due to external loads during driving directly straining the container material and inducing relative motion between liquid metal and container. External loads due to rough-road driving lead to vibration in sprung masses, the case for vehicles with suspension. The load differs between passenger and commercial vehicle applications. Standardised tests for vehicle components suggest random vibration with frequencies from 10 to 2000 Hz with mean acceleration values of around 58 m/s2 which are typical [2, 3]. These vibrations and any extraordinary acceleration or deceleration lead to sloshing and vibration of the molten mPCM relative to the container. It is known that erosive relative motion accelerates corrosion processes which, in this application, may lead to the loss of usable storage capacity and the failure of the container [2, 3]. Materials show a greater propensity to react with or dissolve into molten metal when at a high temperature and in relative motion. For the electric vehicle application, finding a compatible container is a materials science challenge. Ensuring that the container can also withstand pressures due to the weight of metal, volume change on phase transition, and external loads from vehicle operation, increases the challenge. The container ideally should also be cheap, workable, or machinable into the required container shape, resistant to corrosion from the driving environment and safe for humans to handle.

The Compatibility of Metallic Thermal Storage …

17

Table 1 Comparison of thermal properties between Al-12.7 wt.% Si and several pure metals. These values are taken or calculated from the TCAL5 database of Thermo-Calc [4]. Cost estimates are calculated from recent commodity prices Composition wt.%

Melting temperature (°C)

Heat of fusion kJ/kg

Energy density of fusion kWh/L

Constituent cost estimate USD/kWhlatent

Al-12.7 Si

577

505

0.34

18

Al

660

397

0.26

22

Mg

650

349

0.21

26

Si

1414

1788

1.14

5

Thermal Energy Storage in Al-12.7 wt.% Si There exists one eutectic of aluminium and silicon at atmospheric pressure. It occurs at 12.7 wt.% silicon (or 12 mol%) and at a temperature of 577 °C (850 K). The heat of fusion of this eutectic is around 505 kJ/kg, and the energy density of latent storage is approximately 0.34 kWh/L. Aluminium and silicon are common metals in industry and are relatively inexpensive; an estimate of the current cost of the alloy is 1740 USD/T or 18 USD/kWhlatent . These thermal properties are summarised in Table 1. The heat transport properties of the alloy are also exceptional; recent measurements of the thermal diffusivity are provided in the Appendix. Both solid and molten aluminium form a thin passivating layer of Al2 O3 when exposed to air. This layer prevents further oxidation and inhibits reaction with other solids even though aluminium has a very high activity. The same is true for the eutectic composition of interest. The integrity of the oxide layer after long periods above the melting point, during repeated freezing and melting and when subjected to shear stress due to inertial loads, is of great relevance to the alloy’s compatibility under operation. The thermal and transport properties of Al-12.7 wt.% Si make it an attractive alloy for the storage of heat. However, the compatibility of the alloy with a cost-effective, strong, and manufacturable container must be established. Progress towards this goal is the subject of this work. If such a container cannot be identified, TES in the alloy cannot be practically realised.

Compatibility Compatibility can be defined in several ways, each with advantages and limitations. Compatibility at equilibrium refers to materials that show no thermodynamic drive to react. This can be inferred by analysis of phase diagrams, if they are available and accurate. For this definition, there is no thermodynamic drive for a TES material and a container material to react with or dissolve into each other. Compatibility under

18

A. J. Rawson et al.

operation may be shown through simulation of a specific thermal storage application in an experiment. These potentially long-term experiments should replicate the time, temperature and erosive environment of the application. Compatibility under operation requires a definition specific to the application scenario. For the electric vehicle application, there is a strong dependency of the TES operation time on the typical distance profile of the vehicle and on the climatic conditions. The highest temperature of storage would be specified somewhat higher than the mPCM melting point; 650 °C was selected for Al-12.7 wt.% Si. As the application of TES in battery electrical vehicles is considered, the life time of batteries can provide a reasonable approximation of the lifetime requirements for the TES. Batteries for application in electric vehicles typically operate for a maximum of 4500 cycles [5]. Other sources mention a battery life of 150,000 km or 5 years to be reasonable [6]. This is assumed to correspond to around 600 h operation time of the electric vehicle. Considering that cold climatic conditions that necessitate the use of TES occur only for a fraction of the trips taken, the required lifetime of a TES container could be considerably shorter. The numbers above can be considered a worst-case scenario for a passenger electric vehicle application. A material pair compatible under operation does not need to be compatible at equilibrium. For example, graphite and the Al-12.7 wt.% Si eutectic have been shown to be compatible under most operating conditions but the phase diagram shows that the Al8 C7 Si and Al4 C4 Si carbides are more stable at equilibrium. Consideration of only equilibrium compatible phases may preclude more conventional materials that are only compatible under operation.

Relevant Literature In 1981, Birchenall et al. published a major report on heat storage in eutectic alloy transformation, measuring thermal properties, presenting heat transfer theory and compatibility experiments. The reaction layer growth of several molten metals was recorded as a function of time up to three weeks when in contact with SS304 and SiC. This led to the conclusion that silicon carbide was a preferred container material for aluminium and copper rich alloys [7], though still not inert. They also found that aluminium-rich alloys were prone to attacking SS304 and SS316L containers and magnesium-rich alloys would attack graphite. In 2010, Salman et al. published a work presenting results on the performance of thermally sprayed titanium-based coatings on H13 tool steel, concluding that a Ti (Al,O) metallic binder is more resistant to attack by molten aluminium than the intermetallic TiAl [8]. Xu et al. investigated the compatibility of several slags, recycled from the casting industry, with molten aluminium alloys. They found that steel slag showed little chemical composition change after 100 h at 1000 °C immersed in pure aluminium and aluminium-silicon alloys [9]. These experiments were all undertaken in a static environment. Discussion of corrosive-erosive behaviour, where relative motion occurs between the melt and the container, is less common in the literature than static tests. A wide

The Compatibility of Metallic Thermal Storage …

19

reaching review paper was published by Yan and Fan in 2001 discussing the compatibility of aluminium alloys with various metallic and ceramic materials [10]. They discussed, amongst others, their own recent work from 2000 where H21 tool steel pins were rotated in A380 aluminium alloy and the corrosion rate determined [11]. Laboratory erosive-corrosive wear experiments were also published by Miller in 2006 [12]. The rate at which 4140 steel pins immersed and rotated in liquid aluminium A356 reacted was quantified, agreeing with observations in the die casting industry. All the literature reviewed suggests that relative motion significantly increases the reaction rate as compared to static tests and the rate increases either asymptotically or with diminishing returns as relative velocity increases. This brief review of the literature is far from exhaustive but presents the articles of most relevance to this work. Many reviews exist summarising the thermophysical properties of TES materials and occasionally their compatibility with containers. Kenisarin published one such article in 2010 discussing mPCMs and other TES materials [13]. Several comprehensive reviews on erosive-corrosive wear of molten metals can be found, from which one of the most informative is that by Yan and Fan [10].

CALPHAD Survey The calculation of phase diagrams (CALPHAD) method aims to collect all relevant information on phase equilibria and thermochemical properties for a system and describe them in a computer database. Gibbs free energy parameters for individual phases with dependence on temperature, composition, and pressure are stored. If all phases are known for a system, the phase diagram and various thermodynamic properties can be recalculated [14]. Of particular relevance for this work is the calculation of stable phases and their proportions as a function of temperature.

Methodology Container compatibility was determined by identifying a miscibility gap at equilibrium between Al-12.7 wt.% Si and a container phase. This was achieved by calculating a pseudo-binary phase diagram between the alloy and a terminal container phase. The terminal container phases selected were every element for which an assessed ternary phase diagram was available in the relevant database and five oxides; alumina, lime, magnesia, silica, and zirconia. The databases used in this work were the TCAL5 and TCOX8 databases offered by Thermo-Calc [4, 15]. The mass fractions of all stable phases were output as a function of temperature for a system with 1% mass of the terminal container phase. This modelled the reaction products formed at equilibrium within a mPCM rich volume. The temperature was

20

A. J. Rawson et al.

Fig. 1 From left to right. i a plane corresponding to a fixed eutectic composition of a and b was extracted from a ternary phase diagram of components a, b, and c. ii The resulting pseudo-binary phase diagram was drawn. iii The mass fraction as a function of temperature for the stable phases was extracted for a small contribution (1 wt.%) of c. In this demonstrative case in panel iii: α and β (the solid phases of the eutectic), marked with dotted lines, and the liquid L, a dashed line, were compatible with an intermediate phase k, the solid line, but not with the c-rich γ phase

varied from room temperature to 100 °C above the melting point of the eutectic. This process is demonstrated in the schematic of Fig. 1. A phase compatible at equilibrium was defined as a phase that coexisted with the solid and liquid mPCM without dissolving or being dissolved into it over the entire temperature range. Three possibilities were observed: (1) the terminal container phase was compatible, (2) a new intermediate phase was compatible, (3) no compatible phase existed.

Identified Compatible Phases of Al-12.7 wt.% Si The survey of databases identified some previously unknown compatible phases, revealed some containers to be unstable at equilibrium (not necessarily under operation) and highlighted the stability of alumina for Al-12.7 wt.% Si. The results are summarised in Tables 2 and 3. Several silicide phases were identified as compatible at equilibrium, with varying potentials for application. The boron and carbon systems demonstrated behaviour suggestive of operational compatibility and are, based on experience, worthy of further investigation. However, the majority of the systems surveyed either showed solubility or showed intermediate phases that would melt below the maximum operating temperature.

The Compatibility of Metallic Thermal Storage …

21

Table 2 Phases showing equilibrium compatibility with Al-12.7 wt.% Si identified in the survey of the TCAL5 thermodynamic database Elements Terminal phase

Chemical symbol

Compatible phase at equilibrium identified

Notes

Boron

B

AlB12 transforms to AlB2 at 210 °C

Carbon

C

Al8 C7 Si transforms to Al4 C4 Si at 420 °C

Calcium

Ca

CaSi2 melts just above the eutectic melting temperature

Cerium

Ce

Al4 Ce3 Si6 shows solubility in the solids

Chromium

Cr

Al13 Cr4 Si4 shows slight solubility in the solids

Copper

Cu

Al2 Cu dissolves into solids

Iron

Fe

Al9 Fe2 Si2 melts congruently with the solid phases

Germanium

Ge

Complete solubility with diamond Si

Lithium

Li

Solubility and formation of several compounds

Magnesium

Mg

Mg2 Si melts congruently with the solid phases

Manganese

Mn

Al15 Si2 Mn4 melts congruently with the solid phases

Sodium

Na

Nickel

Ni

Al3 Ni melts just above the eutectic temperature

Tin

Sn

Nearly immiscible but melts before eutectic

Strontium

Sr

Titanium

Ti

Vanadium

V

Zinc

Zn

Zirconium

Zr

NaSi

Si2 Sr Al3 Ti switches to AlSi3 Ti at 330 °C Si2 V Soluble in face-centred cubic Al-rich phase Si2 Zr

22

A. J. Rawson et al.

Table 3 Phases showing equilibrium compatibility with Al-12.7 wt.% Si identified in the survey of the TCOX8 thermodynamic database Oxides Terminal phase

Chemical symbol

Compatible phase at equilibrium identified

Alumina

Al2 O3

Al2 O3

Lime

CaO

Magnesia

MgO

Silica

SiO2

Al2 O3

Zirconia

ZrO2

Al2 O3

Notes

Formation of calcium aluminates Formation of spinel Stability of ZrO2 is uncertain

Silicide-coated steel shows great potential as a viable container for molten Al12.7 wt.% Si. Sodium silicide, NaSi, is a water-insoluble silicon source. Thin coatings (up to 10 μm) have been prepared on stainless steel and galvanised steel by dipping and electrophoretic deposition from a sol-gel basic sol [16, 17]. Vanadium silicide, Si2 V, was also identified as a compatible phase at equilibrium. Si2 V has been investigated as a stable coating for components submerged in molten sodium [18]. Vanadium silicide coatings can be generated using pack-cementation methods [16–18]. Neither Si2 Sr nor Si2 Zr shows great viability as container materials. Strontium silicide, Si2Sr, has not been commercially produced as a bulk product or a coating but has been synthesised to explore its capacity as an infrared ray detector and a thermoelectric [19]. Zirconium-silicide coatings have been applied to zirconiumalloys using magnetron sputter deposition, slurry methods, cold spray and thermal spray methods [20]. Zirconium alloys are, however, rarely used outside of the nuclear industry due to their cost. The compatibility of oxides with Al-12.7 wt.% Si can be partially established by analysis of an Ellingham diagram. The most stable oxides (in order of decreasing stability for the relevant temperature range) are CaO, MgO, ZrO2 , Al2 O3 , and SiO2 . Thus quartz could be expected to be the only analysed oxide to show instability. However, appealing to the Ellingham diagram does not yield any information on solubility, nor allotropes, nor about intermediate oxides (e.g. the calcium aluminates common in cement 3CaO Al2 O3 ). Surveying the stable phases formed between the eutectic and various oxides revealed that only alumina is stable at equilibrium (from the five investigated). Silica was expected to reduce to silicon, yielding oxygen to the aluminium metal forming alumina. This was found to be the case suggesting that quartz is not compatible at equilibrium. This was not expected for zirconia; however, the Gibbs free energy of oxidation between aluminium and zirconium are so close that zirconia could not be definitively identified as compatible. Lime and magnesia are more stable than alumina at equilibrium, but intermediate oxide compounds form. Three different calcium

The Compatibility of Metallic Thermal Storage …

23

aluminates are stable for different temperature ranges (C1A1 up to around 350 °C, C1A2 to 450 °C then C1A6 above), though no single phase is stable at equilibrium over the entire temperature. Magnesia was found to form a spinel (MgO-Al2 O3 also written as MgAl2 O4 ). This spinel is not stable for the entire temperature range. Alumina (Al2 O3 ) is a well-known inert crucible and refractory material in laboratories and industry. Limitations to its use stem from its brittle behaviour, difficult machinability and relatively high cost. A flexible alumina coating on another alloy would be an ideal solution for many mPCM applications.

Reaction Experiments To explore the operating compatibility of Al-12.7 wt.% Si with various common container materials, an experimental apparatus and method were created. A modified version of the method employed by Fukahori and colleagues was undertaken [21]. The experiments involved placing a pellet of the eutectic alloy within a crucible of the potential container material. The pellet crucible samples were then placed in a specially designed furnace where the atmosphere, temperature, and hold time could be controlled. The extent of the reaction was investigated post-experiment through observation and microscopy.

Apparatus The apparatus consisted of a tube furnace and its controller, a high strength quartz tube, a sample holder for four samples and thermocouples, four outer alumina crucibles, and a frame on which all these pieces could slide (Fig. 2). The quartz tube could be slid over the sample holder and attached through gas tight fittings to a watercooled housing. This housing contained inert gas and vacuum ports through which the gas environment within the tube could be evacuated and refilled. The thermocouple and pressure sensor feed throughs were also located here. The furnace could be slid over the quartz tube to heat the samples radiatively. The temperature of each sample and the gas pressure within the tube were recorded at one-second intervals during the experiments on a laptop computer.

Methodology Samples consisted of a pellet of Al-12.7 wt.% Si and a crucible of the container material of interest. The eutectic alloy was purchased from a local supplier (Oetinger Aluminium NU GmbH) with a measured composition of 87.33 wt.% Al, 12.3 wt.% Si, and 0.37 wt.% other impurities (the majority of which was iron). Pellets utilised

24

A. J. Rawson et al. Tube Furnace Samples Sliding rails

Quartz tube

Furnace Control

Vacuum and gas ports

Fig. 2 Static pellet test apparatus schematic

were roughly 12 mm in diameter and 3 mm high. Crucibles were nominally 12.7 mm in inner diameter, 15 mm high, and had a wall thickness of 1.65 mm. Considerable tolerance was allowed between crucible materials. The pellets and crucibles were weighed before being placed into the outer alumina crucibles. The alumina crucible and samples were then placed onto the sample holders; the quartz tube slid over them and attached to the housing with screws and an elastomer gasket. The air in the tube was pumped down to 1 × 10−6 mbar before being refilled to around 400 mbar with argon (99.993 mol% purity). The furnace was slid over the samples within the quartz tube and set to 650 °C. The heating rate was not controlled, and the maximum rate measured was approximately 60 K/min. After the specified time period, the furnace was switched off and allowed to cool uncontrolled. The cooling rate was at most 5 K/min. The samples were weighed after the experiment to establish whether any mass loss due to evaporation occurred. Afterwards they were observed, sectioned and analysed with scanning electron microscopy (SEM) and energy-dispersive X-ray (EDX) spectroscopy. The extent of the reaction was qualitatively recorded and phases identified were appropriate.

Results Five different crucibles containing Al-12.7 wt.% Si pellets were held at 650 °C with hold times up to two weeks. The reaction extent was investigated qualitatively, first

The Compatibility of Metallic Thermal Storage …

25

Table 4 Summary of static pellet and static cast crucible experiments Eutectic name and composition (wt.%)

Al-12.7 wt.% Si

Al-12.7 wt.% Si

Al-12.7 wt.% Si

Al-12.7 wt.% Si

Compatibility test method

Static pellet|crucible

Static pellet|crucible

Static pellet|crucible

Static pellet|crucible

Temperature (°C)

650

650

650

650

Time

0.5 h

8h

72 h

2 weeks

Alumina >99.5 wt.%

No reaction observed

No reaction observed

No reaction observed

No reaction observed

Graphite >99 wt.%

No reaction observed

No reaction observed

No reaction observed

Iron > 99 wt.%

Potential reaction

Reaction observed

Quartz Glass >99.9 wt.% Stainless steel (SS304)

Reaction observed Reaction observed

Potential reaction

Reaction observed

Reaction observed

by observation and where possible by SEM and EDX mapping analysis. The results conformed to expectations from the CALPHAD survey, and the microscope images reveal the reaction behaviour (Table 4). Both alumina and graphite showed no reaction with the Al-12.7 wt.% Si pellet for periods of up to two weeks. Alumina is compatible at equilibrium, and no reaction was expected. Graphite, however, has two stable carbides over the temperature range analysed, neither of which were observed. This conforms to expectations from literature [10, 22]. Iron, quartz, and stainless steel (304 grade) all reacted with the pellet as expected. The reaction was not immediately observable after half an hour at temperature suggesting that longer hold times are necessary to confirm reaction with the static pellet crucible method. To demonstrate the analysis method, the results of SEM and EDX analysis are shown below for iron and graphite crucibles [10, 22]. Clear reaction layers were visible between the Al-12.7 wt.% Si pellet and the pure iron crucible (Fig. 3). An EDX line analysis revealed that from the virgin iron, a layer of Al13 Fe forms followed by Al9 Fe2 Si2 with significant porosity. The original aluminium-silicon eutectic was next albeit with platelets of Al5 FeSi phase distributed throughout. The reaction characteristics are consistent with those in the literature [10–12]. Clearly, pure iron cannot serve as an inert container for Al-12.7 wt.% Si [10–12]. No reaction was observed between the Al-12.7 wt.% Si pellet and the graphite crucible after 72 h. The pellet was easily removed from the crucible by simply inverting it. No significant colour changes occurred, and as can be seen from the SEM micrograph and the EDX line scan, no reaction is observable (Fig. 4). This is consistent with findings in the literature [22] and industry experience [23].

26

A. J. Rawson et al. 100

Fe

I

II

III

IV

Atom Percent [%]

80

Al

60

I

II

III

IV

40

Si

20

0

0

0.5

1

2

1.5

2.5

Distance [mm]

Fig. 3 Cross section of Al-Si in Fe static pellet crucible 72 h experiment. I bcc iron, II Al13 Fe4 , III Al9 Fe2 Si2 , IV Al-12.7 wt.% Si with Al5 FeSi platelets 100 C 80

II

III

Atom Percent [%]

I

Al

60

40

Si 20

0

0

0.5

1

1.5

2

Distance [mm]

Fig. 4 Cross section of Al-Si in C static pellet crucible 72 h experiment. I graphite, II void, III Al12.7 wt.% Si. The zero readings between 0.3 and 0.45 mm are due to a void between the reinserted sample and the crucible

Compatibility Experiment Outlook A TES system for space heating in electric vehicles should be guaranteed for a long working lifetime, 600 h for a typical worst-case scenario in a passenger vehicle. A crucial contributor to TES system life is compatibility between the mPCM and its container. Demonstrating material compatibility over such long time periods requires a means of accelerating the reaction. Reaction between molten metal alloys and containers may be accelerated through increased temperature and relative motion which also acts to remove inhibiting layers (oxides or stable compounds).

The Compatibility of Metallic Thermal Storage …

27

A high-temperature stirring furnace is being developed at the Institute of Material Physics in Space at the DLR Cologne. This apparatus will provide precise relative motion between potential container materials and mPCMs with a controlled temperature and atmosphere. This will enable quantification of the activation energy and reaction rates relevant to the storage of high-temperature molten metals and other liquids in various containers.

Conclusion Computational and experimental methods for the identification and demonstration of compatible container materials for Al-12.7 wt.% Si, an mPCM relevant to vehicular space heating applications, have been introduced. Several potential container materials have been identified for the eutectic and their viability discussed. Experimental results for long-term static pellet crucible reaction experiments with iron, steel, alumina, quartz, and graphite have been presented. SEM micrographs with EDX analysis of reaction layers were shown for iron and graphite crucibles in contact with Al-12.7 wt.% Si confirming predictions from the CALPHAD survey and expectations from the literature. Finally, the outlook for future accelerated reaction experiments at the DLR was briefly discussed. Acknowledgements The authors would like to acknowledge the financial support of a post-doctoral scholarship provided by the German Academic Exchange Service (DAAD). The static pellet test apparatus was developed by Patrick Lehmann and Christof Dreißiacker at the Institute of Material Physics in Space. Advice in the selection of initial experiments was provided by Prof. Dr. Jürgen Brillo, and expertise in SEM and EDX analysis was given by Dr. Mathias Kolbe and Dr. Mareike Wegener all of the same institute. The authors gratefully acknowledge their contribution to this work.

Appendix The thermal transport properties of Al-12.7 wt.% Si are relatively high as compared to other metals and extremely high relative to salt, wax, or oil TES materials. The thermal diffusivity for the eutectic as a function of temperature was measured using the Netzsch—LFA 467 HT HyperFlash® at the Institute of Material Physics in Space at the DLR. The alloy was purchased from Oetinger Aluminium NU Gmbh with mass percentages of 87.33% Al, 12.30% Si and 0.37% impurities (Fe being the main impurity). The results are shown in Fig. 5.

28

A. J. Rawson et al. 90 80

Thermal Diffusivity [mm2/s]

70 60 50 40 30 20 10 0

0

100

200

300

400

500

600

700

800

Temperature [°C] Fig. 5 Measured thermal diffusivity of Al-12.7 wt.% Si as a function of temperature. The solid line is a simple linear fit to guide the eye

References 1. Kraft W et al (2018) Thermal high performance storages for use in vehicle applications. Paper presented at the 2nd ETA Conference, Berlin, Germany, 22–23 November 2018 2. ISO/FDIS (2012) 16750-3 Road vehicles—environmental conditions and testing of electrical and electronic equipment—Part 3: mechanical loads. International Standards Organization. https://www.iso.org/about-us.html. Accessed 6 June 2019 3. IEC (1993) 60068-2-64 Environmental testing part 2: Test methods—Test Fh: vibration, broadband random (digital control) and guidance 4. Thermo-Calc Software (2018) TCAL5 Al-based Alloy Database. https://www.thermocalc.com/ media/19849/tcal5_extended_info.pdf. Accessed 7 July 2019 5. Hannan MA, Hoque MM, Mohamed A, Ayob A (2017) Review of energy storage systems for electric vehicle applications: Issues and challenges. Renew Sustainable Energy Rev 69:771–789 6. Jussani AC, Wright JTC, Ibusuki U (2017) Battery global value chain and its technological challenges for electric vehicle mobility. RAI Revista de Administração e Inovação 14:333–338 7. Birchenall EC, Güceri SI, Farkas D, Labdon MB, Nagaswami N and Pregger B (1981) Heat storage in alloy transformations. NASA. https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/ 19810019065.pdf. Accessed 19 June 2019 8. Salman A, Gabbitas BL, Cao P, Zhang DL (2011) The performance of thermally sprayed titanium based composite coatings in molten aluminium. Surf Coat Technol 205:5000–5008 9. Xu H, Sadiki N, Dal Magro F, Py X, Mancaux JM, Romagnoli A (2017) Compatibility tests between molten aluminium alloys and recycled ceramics from inorganic industrial wastes. Energy Procedia 142:3689–3696 10. Yan M, Fan Z (2001) Review: Durability of materials in molten aluminum alloys. J Mater Sci 36:285–295

The Compatibility of Metallic Thermal Storage …

29

11. Yan M, Fan Z (2000) The erosion of H21 tool steel in molten A380 alloy. J Mater Sci 35:1661– 1667 12. Miller AE, Maijer DM (2006) Investigation of erosive-corrosive wear in the low pressure die casting of aluminum A356. Mater Sci Eng, A 435–436:100–111 13. Kenisarin MM (2010) High-temperature phase change materials for thermal energy storage. Renew Sustain Energy Rev 14:955–970 14. Spencer PJ (2008) A brief history of CALPHAD. CALPHAD 32(1):1–8 15. Thermo-Calc Software (2018) TCOX8 metal oxide solutions database. https://www. thermocalc.com/media/6001/tcox7_extended_info.pdf. Accessed 9 July 2019 16. Castro Y, Duran A, Damborenea JJ, Conde A (2008) Electrochemical behaviour of silica basic hybrid coatings on stainless steel by dipping and EPD. Electrochim Acta 53:6008–6017 17. Conde A, De Damborenea J, Duran A, Menning M (2006) Protective properties of a sol-gel coating on zinc coated steel. J Sol-Gel Sci Technol 37:79–85 18. Chaia N, Mathieu S, Cozzika T, Rouillard F, Desgranges C, Courouau J, Petitjeam C, David N, Vilasi M (2013) An overview of the oxidation performance of silicide diffusion coatings for vanadium-based alloys for generation IV reactors. Corros Sci 66:285–291 19. Hashimoto K, Kurosaki K, Imamura Y, Muta H, Yamanaka S (2007) Thermoelectric properties of BaSi2, SrSi2 and LaSi. J Appl Phys 102(6):063703. https://doi.org/10.1063/1.2778747 20. Sridharan K, Mariani R, Bai X, Xu P, Lahoda E (2012) Development of self-healing zirconium silicide coatings for improved performance zirconium-alloy fuel cladding. U.S. Department of Energy Office of Scientific and Technical Information. https://www.osti.gov/servlets/purl/ 1430630. Accessed 26 June 2019 21. Fukahori R, Nomura T, Zhu C, Sheng N, Okinaka N, Akiyama T (2016) Thermal analysis of Al-Si alloys as high temperature phase change material and their corrosion properties with ceramic materials. Appl Energy 163:1–8 22. Reed S, Sugo H, Kisi E, Richardson P (2019) Extended thermal cycling of miscibility gap alloy high temperature thermal storage materials. Sol Energy 185:333–340 23. Luxel Corporation (2017) Crucible selection guide. https://luxel.com/wp-content/uploads/ 2017/08/Crucible-Selection-Guide-Rev.08-2017.pdf. Accessed 2 May 2019

Effect of Fly Ash from Coal-fired Boiler on Heat Transfer Efficiency Jiapeng Liang, Haibin Zuo, Yingli Liu and Shenhui Liu

Abstract Energy conservation and emission reduction are the development goals of future industry, and heat exchangers play a vital role in the efficient use of energy. But the fly ash formed by the insufficiently burnt coal powder and other impurities has many negative effects in the waste heat recovery. Tube bundle wear and ash deposition in heating surface are common problems in engineering. To explore the effect of fly ash on heat transfer, a small heat transfer testor was established in laboratory, and the characteristics of fly ash, such as particle concentration, velocity, and particle size were studied in this testor. Based on experimental work, it is helpful to improve the service life of equipment. The summed empirical formula has certain guiding significance for actual engineering production. Keywords Heat transfer coefficient · Dusty airflow · Fly ash characteristics · Ash deposit · Experimental study

Introduction Energy conservation and emission reduction are one of the key objectives in industrial production, and heat exchangers play an important role in energy recovery. However, the high-temperature exhaust gas contains a large amount of coal ash particles and corrosive impurities, which makes them easy to deposit scale on the surface of the metal heat exchanger to block the pipeline, reduce heat exchange efficiency [1, 2], J. Liang · H. Zuo (B) · Y. Liu · S. Liu State Key Laboratory of Advanced Metallurgy, University of Science and Technology Beijing, Beijing 100083, China e-mail: [email protected] J. Liang e-mail: [email protected] Y. Liu e-mail: [email protected] S. Liu e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_3

31

32

J. Liang et al.

and in addition, the fly ash particles in the flue gas to the heat exchange tube bundle continuously scouring and causing heat exchanger failure and maintenance, causing major safety accidents in severe cases. Therefore, by studying the deposition and wear of fly ash on the heat transfer surface, and their impact on the heat transfer process is necessary, which provides a reference for the efficient utilization of waste heat resources and heat exchanger design. The fine particles generated after the combustion of the pulverized coal gradually deposit on the heat transfer surface as the flue gas flows, causing the tube bundle to clog and the thermal efficiency to decrease. Due to the complexity of fly ash deposition, in recent years, researchers have used numerical simulation of fluent to calculate the deposition process of fly ash, using simulation methods to study the effect of dirt on the heat exchanger [3–5]. In this process, a series of theories about particle deposition and collision are proposed [6–8]. However, it is necessary to start from the actual experiment. This paper mainly based on the experimental aspect, and calculates the heat transfer effect under the fly ash and the influence of concentration, temperature and flow on the heat transfer coefficient by setting up the test bench and formula.

Experimental Experimental Device Schematic diagram of the heat exchange system is shown in Fig. 1. The experimental system consists of a supply air system, a feeding system, a test section, a circulating water system, and a data acquisition system. The total length of the rectangular flue is 2.5 m, and the heat transfer test section is made up of 9 metal tubes with a length of 320 mm and an outer diameter of 28 mm arranged in a row of 3 × 3. In order to reduce the error caused by heat loss, the entire flue is covered with a layer of quartz fiber insulation. The circulating water flows into the heat exchange section through the water pump to take away the heat in the flue gas, and the heat is lost in the water tank and then continues to circulate. The fly ash used in the experiment was taken from the actual power plant dust ash, and the tail bag dust collector was used to recover fly ash to prevent pollution.

Experimental Procedure The air is blown into the electric heating pipe by the air compressor to be heated, and the screw feeder is added to the fly ash to enter the mixing section for mixing, so that the particles with the particle diameter less than 0.50 mm are accelerated to more than 95% of the air velocity, and the gas-solid two-phase flow is gradually stabilized. The

Effect of Fly Ash from Coal-fired …

33

Fig. 1 Heat transfer test bench. 1—fan, 2—heating pipe, 3—temperature control device, 4—flow meter, 5—screw feeder, 6—test section, 7—bag dust collector, 8—water pump, 9—water tank

dust-containing gas stream enters the heat exchange test section, and the cold liquid in the tube is exchanged for heat, and is discharged after being collected by the bag. In the experiment, the blower frequency is adjusted to control the flue gas flow rate, and the temperature control box can set the required temperature and change the screw feeder frequency to control the fly ash drop concentration. Thermocouples were installed before and after the heat exchange test section to measure the temperature of the flue gas inlet and outlet, and the thermocouple was connected to the inlet and outlet of the circulating water. The flow meter is installed at each of the flue gas inlet and the circulating water inlet, and the average flow rate is calculated before and after the experiment to calculate the average flow rate. Waiting 2 h after each change of working condition, and record the data after the meter value is stable. Flue gas inlet flow range: 30–55 (m3 /h), temperature setting range: 110–180 (°C), and fly ash concentration range: 3.5–27 (g/m3 ).

34

J. Liang et al.

Results and Discussion Heat Transfer Calculation Nomenclature Φ

Heat flow

k

Heat transfer coefficient

c

Specific heat capacity

Q

Flue gas flow

q

Mass flow

T

Temperature (°C)

ΔT

Fluid temperature difference

n

Ash concentration

ΔT m

Average temperature difference

The outer diameter of the heat exchange tube is d = 32 mm, the number of tubes is n = 9, and the total heat exchange area ‘A = nlπ d’ is calculated to be 0.2532 m2 . Φ = k Atm Φ1 = c1 q1 T1 Φ2 = c2 q2 T2 Φ = Φ1 + Φ2 tm =

tmax − tmin t ln tmax min

When checking the manual, the temperature selected by cp1 , cp2 is the qualitative temperature of the fluid t f = (t 1 + t 2 )/2.

Effect of Temperature on Heat Transfer Figure 2 shows the effect of temperature on the heat transfer coefficient for ash concentration of 13.7 (g/m3 ) and no fly ash. As shown in the figure, under the condition of no ash, the heat transfer coefficient decreases with the increase of temperature and gradually decreases to 150 °C and then stabilizes to 72.3 W/m2 k. Under a certain concentration of fly ash, the heat transfer coefficient first increases with the increase of temperature, and reaches a maximum at 160 °C, and then gradually stabilizes 71.2 W/m2 k. For the heat exchange tube wall under no ash conditions, the heat transfer surface is not obstructed by dirt. As the temperature rises, the heat transfer surface temperature increases, and the heat is transferred to the cooling water in

Heat transfer coefficient (W/m2·k)

Effect of Fly Ash from Coal-fired …

35

91 No fly ash Fly ash

84

77

70

63 105

120

135

150

165

180

Temperature (䉝㻕 Fig. 2 Effect of temperature on heat transfer

the tube through the wall surface heat conduction. As the temperature of the heat exchange surface increases, the heat conduction of the wall surface gradually stabilizes, so the whole system tends to be steady state, and the heat transfer coefficient tends to be constant. In the ash condition, the heat is first transferred to the surface ash due to the deposition of fly ash on the surface of the tube bundle, and then the heat is transferred to the wall by surface deposition ash. Finally, the heat conduction to the wall is stabilized. In addition, because of the presence of heated surface area ash, the heat transfer coefficient after stabilization is 1.1 lower than that of the clean condition

Effect of Flow on Heat Transfer Figure 3 shows the effect of flow on the heat transfer coefficient for a ash concentration of 13.7 (g/m3 ) and no fly ash. As shown in the figure, the effect of flow on the heat transfer coefficient, whether there is ash or no ash, has this similar law. As the flow rate increases, the heat transfer coefficient gradually increases. Compared with the heat transfer coefficient under the ash, the heat transfer coefficient under clean conditions is relatively high. The flow rate of the flue gas increases, the heat passing through the unit cross-section increases, and the amount of heat contacted by the heated surface per unit time increases, so the heat transfer coefficient increases. However, after the fly ash is deposited on the heated surface, the surface heat transfer resistance increases and the heat transfer coefficient is greatly reduced. As the flow rate increases, the amount of heat flowing through the unit area of the ash deposit increases, so similar

J. Liang et al.

Heat transfer coefficient (W/m2·k)

36

76

No fly ash Fly ash

72 68 64 60 30

40

Gas flow (m3/h)

50

60

Fig. 3 Effect of flow rate on heat transfer coefficient

55

72

1 2

69

50 45

66 63

40

60

35

57 0

5

10

15

20

Ash concentration (g/m3)

Fig. 4 Effect of ash concentration on heat transfer

25

30 30

Gas flow (m3/h)

2 Heat transfer coefficient (W/m ·k)

to the clean working condition, the heat transfer coefficient also increases as the flow rate increases. The heat transfer efficiency of the clean heated surface at steady state is 5.4 higher than that of the ash. The curve 1 shown in Fig. 4 is the heat transfer law at different concentrations, and the curve 2 is the heat transfer law at different concentrations under different flow rates. Curve 1 shows that other conditions are unchanged, the ash concentration is increased from 3.5 to 27 g/m3 , and the heat transfer coefficient is only reduced

Effect of Fly Ash from Coal-fired …

37

by about 4. The thickness of the ash in the heated area does not increase with the increase of the ash concentration, and the ash concentration has little effect on the heat transfer. The ash concentration of curve 2 increased from 11.4 to 20.6 g/m3 , while the heat transfer coefficient decreased significantly. It is caused by the decrease of flow rate, which indicates that the flow rate plays a decisive role in the heat exchange process. After the heat per unit area increases, the heat will be transmitted to the heat receiving surface, and then the heat is transferred to the cold fluid through the tube wall for heat. Exchange, although the presence of ash will reduce heat transfer, but from the results, the effect of flow changes on the heat transfer coefficient is far greater than the impact of concentration.

Conclusions (1) There is a critical value for the heat transfer coefficient affected by temperature. After 150 °C, the temperature no longer affects the change of heat transfer coefficient. And the heat transfer coefficient of the clean working condition is higher than the value under the ashing condition by 1.1. (2) When other conditions remain unchanged, the heat transfer coefficient increases with increasing flow rate regardless of the presence of ash on the heated surface. And the heat transfer coefficient of the clean working condition is 5.4 higher than the average value when the dust is accumulated. (3) The flow rate plays a decisive role in the influencing factors of the heat transfer coefficient, and the effect of the flow rate is far greater than the effect of the concentration. Acknowledgements The author expresses gratitude to the national key research and development program, the research on the integrated control technology of coal-fired boiler pollutants (SO2 , NOx , PM) and engineering demonstration (2016YFB0600701).

References 1. Lee BE et al (2002) Computational study of fouling deposit due to surface-coated particles in coal-fired power utility boilers. Fuel 81(15):2001–2008 2. Pan Y et al (2011) An integrated theoretical fouling model for convective heating surfaces in coal-fired boilers. Powder Technol 210(2):150–156 3. Han H et al (2014) A parameter study of tube bundle heat exchangers for fouling rate reduction. Int J Heat Mass Transf 72:210–221 4. Sato N et al (2015) Growth and gravity shedding of ash deposition layer in pulverized coal combustors. Fuel Process Technol 134:1–10 5. Naganuma H et al (2013) Control of ash deposition in solid fuel fired boiler. Fuel Process Technol 105:77–81

38

J. Liang et al.

6. Nagarajan R et al (2009) Development of predictive model for fly-ash erosion phenomena in coal-burning boilers. Wear 267(1-4):122–128 7. Dong M et al (2013) A dynamic model for the normal impact of fly ash particle with a planar surface. Energies 6(8):4288–4307 8. Bazzo Edson (2012) Characterization and growth modeling of ash deposits in coal fired boilers. Powder Technol 217:61–68

Optimization and Management of On-Site Power Plants Under Time-of-Use Power Price: A Case Study in Steel Mill Xiancong Zhao, Huanmei Yuan, Zefei Zhang and Hao Bai

Abstract The implementation of time-of-use (TOU) power tariff in Chinese steel industry provides an opportunity for steel mills to reduce electricity bills through an optimal collaboration between the on-site power plant (OSPP) and energy storage equipment (gasholders). In this paper, a mixed-integer linear programming (MILP) based scheduling model was proposed to achieve the optimal operation of OSPP and gasholders in a steel mill under TOU tariff. Compared with previous models, we considered the influence of TOU power tariff on the optimal scheduling of OSPP. The results of a case study demonstrate that the optimization model can achieve better peak-valley shifting of the electricity generation and decrease the electricity purchasing cost by 7.5% with improved gasholder stability. In addition, the overall power generation efficiency can be increased by 2.13% using the proposed model, which indicates that the byproduct gases can be effectively and efficiently used. Keywords Steel making industry · Byproduct gases · Optimal scheduling · Combined cycle power plants · Time-of-use (TOU) power price

Introduction As an energy-intensive industry (EII), the steel making industry accounts for approximately 15% of total energy consumption in China [1]. In integrated steel works, the primary energy consumed are cleaned coal and blind coal, which accounts for approximately 70% of the total energy consumption [2], and more than 33% of the coal energy is converted into byproduct gases, namely coke oven gas (COG), blast X. Zhao (B) Department of Industrial Engineering and Management, Peking University, Beijing 100871, China e-mail: [email protected] H. Yuan · Z. Zhang · H. Bai State Key Laboratory of Advanced Metallurgy, University of Science and Technology Beijing, 30# Xueyuan Road, Beijing 100083, China School of Metallurgical and Ecological Engineering, University of Science and Technology Beijing, 30# Xueyuan Road, Beijing 100083, China © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_4

39

40

X. Zhao et al.

furnace gas (BFG), and Linz-Donawitz process gas (LDG) [3]. The majority of these gases are used for heating in different furnaces throughout the plant, and the rest is consumed by the on-site power plants (OSPP) to produce electricity and steam for internal or external use [4, 5]. Because the steel production process is unstable, both the generation and the consumption of byproduct gases are suffered from fluctuations [6]. As a result, the steel enterprises installed buffer units like gasholders and boilers to reduce the fluctuations, and thus to decrease the byproduct gas flaring and shortage [7]. Therefore, scheduling the operation loads of these buffer units is significant for reducing the fluctuations of the byproduct gas system. The importance of the byproduct gas scheduling had been noticed by some researchers decades ago. A linear programming (LP) based byproduct gas scheduling model was built by Robert in 1980 [8]. Results demonstrated that the utilization of byproduct gases was improved and the purchasing cost of external fuel was reduced. However, the optimal control of gasholders was not considered in his model. Akimoto considered the fluctuation of gasholders in the objective function and proposed a mixed integer linear programming based (MILP) scheduling model to realize the optimal allocation of byproduct gases between gasholders and boilers [9]. Based on Akimoto’s method, Kim considered the switching on/off penalty of the boiler burner and improved the MILP scheduling model [10]. In previous models, the OSPPs in steel enterprises mainly consist of steam boilers and turbines, and the average electricity generation power of them is only 20–35%. Recently, with the increasing need for energy saving and emission reduction in Chinese steel industry, the installation of combined cycle power plants (CCPP) was widely encouraged in steel enterprises. The CCPP is regarded as a better way to utilize the byproduct gases because its power generation efficiency is much higher. On the other hand, the installation of CCPP increased the complexity of the byproduct scheduling problem and the influenced of the time-of-use (TOU) power tariff on operation strategy should be considered in the scheduling model. In this paper, a MILP-based multi-objective scheduling model for byproduct gases in steel plant is studied and developed. Especially, the influence of the TOU power tariff on optimal operation of boilers and CCPPs is considered. The rest of the paper is organized as follows: Section “Problem statement” and “Formulation” described the byproduct gas system and the mathematical model. Section “Results and Discussion” discussed the optimization results in detail and the conclusion of this paper is drawn in Section “Conclusions”.

Problem Statement The byproduct gas system in steel mill is illustrated in Fig. 1. Generally, approximately 60% of byproduct gases are used for heating in the metallurgical process, and the flexibility of gas consumption is very limited because the regular steel production is stable [2–4]. The rest of the byproduct gases can be adjusted more flexibly either consumed by the OSPP or stored in the gasholders.

Optimization and Management of On-Site Power Plants…

41

Fig. 1 Schematic view of the byproduct gas system in a steel mill

The imbalance between the production and consumption of byproduct gases makes the overall system suffer from huge fluctuations, which may necessitate gas flaring or can negatively affect the gas supply of the OSPP. The holder level should be sustained around the middle level (1/2 of the maximum volume) to achieve the best anti-fluctuation function for the gasholder [5, 6]. However, situation can be quite different if TOU electricity tariff is considered. As described in Section “Introduction”, the gas storage function of the gasholders becomes vital for reducing the electricity cost under TOU electricity tariff. Table 1 shows a typical TOU electricity tariff in a steel plant in China, in which the electricity price during the peak period is 2.43 times higher than that in the valley period. This pricing allows the possibility of reducing the electricity cost by optimal collaboration between gasholders and OSPP, as shown in Fig. 2. The operation load of OSPP (boilers and CCPPs) needs to be adjusted dynamically in accordance with the TOU electricity tariff, and the influence of operation load on the thermal efficiency of OSPP should be considered. In short, the present work considers the following aspects: Table 1 Comparison of the manual operation and optimal calculation results for the total operation cost during the whole scheduling period

Item Gasholder penalty cost (CNY)

Manuel

Proposed

369,915

78,561

Electricity purchasing cost (CNY)

1,532,116

1,416,590

Total cost (CNY)

1,902,031

1,495,151

42

X. Zhao et al.

Fig. 2 Schematic view of the collaboration between the gasholder level and electricity generation of the OSPP under the TOU tariff

– The middle level of the gasholder has the best ability to keep the gas system stable. – The electricity purchasing cost can be minimised by collaborative scheduling between gasholders and OSPP under TOU tariff. – The thermal efficiency of OSPP changes with the operation load, which affects the optimisation result.

Formulation Objective Function The objective function is to minimise the operation costs of the byproduct gas system under the TOU tariff, which consists of the gasholder penalty cost (GPC) and the electricity purchasing cost (EPC), as shown in Eq. (1). Y = min{GPC + EPC}

(1)

Gasholder Penalty Cost (GPC) A calculation method for the GPC is formulated in Eq. (2). The penalty cost is zero when the holder level is kept at the middle level, whereas the penalty cost is high if there is a large deviation from the middle level of the gasholder, as illustrated in Fig. 2. The gasholder level should be kept above the lower level to avoid severe

Optimization and Management of On-Site Power Plants…

43

mechanical problems. On the other hand, gas flaring will occur when the holder level is above the higher level, which should also be avoided. GPC =

P   t=1

j WH S H,t

+

P   t=1

G

j WL SL ,t

+

G

P   t=1

G

j WD SD,t

+

P   t=1

j

C flar Sflar,t

G

(2) where WH , WL and WD are the penalty factors for high, low and normal deviations, j j j respectively; SH,t , SL,t and SD,t are the amounts of byproduct gases that deviate from the high, low, and middle levels of gasholder j during time period t, respectively; and j C flar and Sflar,t represent the unit price and volume for byproduct gas flaring during time period t, respectively.

Electricity Purchasing Cost (EPC) The EPC is described in Eq. (3), where Ctelec ,E t,dem , E t,pur and E t,gen are the unit price, overall electricity demand, electricity purchased from the main grid and electricity generated from the OSPP during time period t, respectively. EPC = Ctelec

P  t=1

E t,pur = Ctelec

P 

(E t,dem − E t,gen )

(3)

t=1

Constraints Constraints include mass balance constraints, energy balance constraints and operation constrains. For details please see Refs. [5, 6].

Results and Discussion A case study was carried out in a steel plant in Hebei province, China, with four blast furnaces (BF), six coke ovens (CO), five Linz-Donawitz converters (LDC), and two gasholders for each type of byproduct gas. The electricity generation, operation load and total cost comparison by manual operation and optimal calculation were discussed below.

44

X. Zhao et al.

Fig. 3 Electricity generation in each period before and after optimisation

Electricity Generation The total amount of power generation for each time period before and after optimization was shown in Fig. 3. The solid line represents manual operation result while the dashed line represents optimal calculation result. It can be noticed that the power output increased during the PPP while decreased during the VPP.

Operation Load Figure 4 depicts the operation load of CCPPs and boilers before and after optimization. Compared with manual operation, the operation load of 1# and 2# CCPP increased remarkably during the PPP while the 2# CCPP decreased significantly during the VPP. The operation load of boilers was decreased after optimization, especially for 1# boiler. According to the calculation results, the average operation load of 1# CCPP, 2# CCPP, 1# boiler, and 2# boiler was 94.7%, 84.8%, 42.6% and 55.3%, respectively. The proposed model tends to increase the operation load of higher efficiency units during the calculation process. The distribution of byproduct gases during the VPP was limited (some of byproduct gases were stored in holders for power generation during the PPP), which was difficult to sustain high operation load of power generation units. In other words, some boilers have to operate under very load operation load during the VPP. According to Fig. 4, the operation load of 1# boiler and 2# boiler during VPP were 44.33% and 50.18%, respectively.

Optimization and Management of On-Site Power Plants…

45

Fig. 4 Operation loads of boilers before and after optimisation (a 1# CCPP; b 2# CCPP; c 1# boiler; d 2# boiler 4#)

The scheduling model also tends to distribute more byproduct gases to CCPP/boiler with higher efficiency performance. Byproduct gases will also be preferentially distributed to boilers which were sensitive to the change of operation load, especially when the byproduct gases shortage happens. The aim was to improve the overall efficiency of CCPPs and boilers, which were increased by 2.13% after optimization.

Total Cost Comparison The total costs of manual operation and proposed method were shown in Table 1. The gasholder penalty cost and the electricity purchasing cost were reduced after optimization.

Sensitivity Analysis The sensitivity analysis was conducted on peak-valley price rate (PVR).

46

X. Zhao et al.

Fig. 5 Total electricity generation and PEG for each price period at various PVRs

The total amount of electricity generation and the percentage of electricity generation (PEG) under different PVR were shown in Fig. 5. The total amount of electricity generation was not sensitive to the increase of PRV. However, when the PVR increased to twice of its current value, the PEG during the PPP increased from 38.29% to 43.24% while the PEG during the MPP and VPP decreased from 33.13% to 30.81% and 28.58 to 25.95%, respectively. The increase of PVR contributes to the decrease of electricity purchasing cost.

Conclusions This paper proposed a scheduling model for the distribution of byproduct gases in steel enterprise under time-of-use power price. The influence of combined cycle power plant (CCPP) on the allocation of byproduct gases was considered in this work and the following conclusion can be drawn. (1) Peak load shifting under TOU tariff can be better realized through the optimization model, compared with manual operation results. (2) CCPP plays a dominate role compared with steam boiler in distributing byproduct gases because CCPP has higher efficiency performance. The operation load of 1#CCPP, 2#CCPP, 1#boiler, and 2#boiler was 94.7%, 84.8%, 42.6% and 55.3%, respectively. (3) The sum of standard deviation volume (SSDV) was sensitive to the peak-valley ratio (PVR). With the increase of PVR, the power generation during the peak price period (PPP) increases while during the medium price period (MPP) and valley price period (VPP) decreases.

Optimization and Management of On-Site Power Plants…

47

Acknowledgements This research was supported by Boya post-doctoral project of Peking University.

References 1. He K, Zhu H, Wang L (2015) A new coal gas utilization mode in China’s steel industry and its effect on power grid balancing and emission reduction. Appl Energ 154:644–650 2. Zhang XP, Zhao J, Wang W, Cong LQ, Feng WM (2011) An optimal method for prediction and adjustment on byproduct gas holder in steel industry. Expert Syst Appl 38(4):4588–4599 3. Li L, Li HJ (2015) Forecasting and optimal probabilistic scheduling of surplus gas systems in iron and steel industry. J Cent South Univ 22(4):1437–1447 4. Junior VB, Pena JG, Salles JL (2016) An improved plant-wide multiperiod optimization model of a byproduct gas supply system in the iron and steel-making process. Appl Energ 164:462–474 5. Zhao XC, Bai H, Lu X, Shi Q, Han JH (2015) A MILP model concerning the optimisation of penalty factors for the short-term distribution of byproduct gases produced in the iron and steel making process. Appl Energ 148(2):142–158 6. Zhao XC, Bai H, Shi Q, Lu X, Zhang ZH (2017) Optimal scheduling of a byproduct gas system in a steel plant considering time-of-use electricity pricing. Appl Energ 195:100–113 7. Zeng YJ, Sun YG (2015) Short-term scheduling of steam power system in iron and steel industry under time-of-use power price. J Iron Steel Res Int 22(9):795–803 8. Robert E (1980) Improving fuel utilization in steel mill operations using linear programming. J Oper Manage 2:95–102 9. Akimoto K, Sannomiya N, Nishikawa Y (1991) An optimal gas supply for a power plant using a mixed integer programming model. Automatica 3:513–518 10. Kim JH, Yi HS, Han C (2003) A novel MILP model for plant-wide multi-period optimization of byproduct gas supply system in the iron and steel making process. Chem Eng Res Des 8:1015–1025

Economic Metals Rescue from Spent Zinc–Carbon Batteries for Industrial Value Additions Alafara A. Baba, Folahan A. Adekola, Rafiu B. Bale, Abdul G. F. Alabi and Mustapha A. Raji

Abstract The increasing demands for metals with gradual depletion of unrenewable resources warrant the need for industrial metals recovery from secondary sources including zinc–carbon batteries. The recycling from wastes is important as cost of safe disposal of its hazardous components is quite high compared to the amount of waste produced-cum-limited storage capacity. For instance, sub-Sahara African countries’ including Nigeria has been found to have dominance of nonrechargeable spent batteries containing precious metals content. These metals which may be toxic are valuable industrial elements if re-processed. In this work, combinations of acid leaching, solvent extraction and precipitation techniques were utilized in processing spent Tiger Head Zinc–carbon batteries assaying majorly 41.30 wt% ZnO, 4.30 wt% Fe2 O3 and 2.69 wt% MnO2 . At optimal conditions, the leach liquor was selectively treated to achieve 96.7% zinc recovery efficiency by Cyanex® 272 extractant prior to its beneficiation as zinc oxide suitable as coating and industrial raw materials for some defined industries. Keywords Zinc–carbon batteries · Nigeria · Zinc oxide · Leaching · Solvent extraction

A. A. Baba (B) · F. A. Adekola · M. A. Raji Department of Industrial Chemistry, University of Ilorin, P.M.B. 1515, Ilorin 240003, Nigeria e-mail: [email protected] R. B. Bale Department of Geology and Mineral Sciences, University of Ilorin, P.M.B. 1515, Ilorin 240003, Nigeria A. G. F. Alabi Department of Material Science and Engineering, Kwara State University, P.M.B. 1530, Malete, Nigeria © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_5

49

50

A. A. Baba et al.

Introduction In recent times, the consumption of zinc–carbon batteries tremendously increases as the global population and economic explosion increase which thus accompanied by a larger spent battery wastes [1]. Thus, spent batteries contribute to the environmental menace due to its high proportion of heavy metals. Unlike secondary cells such as lead accumulators, primary cells are very unique with relative to chemical composition and behold 80–90% of all small batteries [2, 3]. In most sub-Sahara African countries, zinc batteries constitute by far the dominant portion and could therefore be considered as a valuable one of the secondary raw materials for the recovery of zinc-cum-assessment of energy properties of its beneficiated products [3]. For example, treatment of spent zinc–carbon (Zn–C) batteries requiring low energy during processing could find wide array of applications in calculators, remote controls, toys and several other objects where small amount of energy is required. The annual world production of small portable batteries characterized with distinct shapes and sizes is more than 100 billion. Out of the mass production, Zn–C batteries cover approximately 31% [4]. The spent Zn–C batteries powder is generally made up of ZnMn2 O4 , ZnO, Mn2 O3 , Mn3 O4 , Zn(NH3 )2 Cl2 coupled with other valuable compounds such as NH4 Cl, carbon and plastics [5]. Thus, the precious metals associated with spent Zn–C batteries discharged on farmland, for instance, may finds its way into the groundwater, while incineration discharge toxic gases cause environmental havoc and subsequently recycle precious metals for industrial utilities [6, 7]. Both pyro- and hydro-metallurgical routes have been utilized to extract and process precious metals of industrial utilities from spent batteries particularly Zn–C batteries [7, 8]. Pyrometallurgical method involves distillation of Zn and subsequent concentration of other associated valuable metals. These techniques are mostly employed [7], and though it may not be economically viable due to high energy consumption [9], hydro-metallurgical technique with leaching as important operational unit and due to its distinct properties of low cost, minimal energy and eco friendly is considered profitable for treating and processing of secondary resources for industrial applications [10]. Consequently, the amphoteric distinct indices of zinc make its leaching and separation carried out in acid or basic media, among others [11]. Many researchers having considered the energy differences have worked extensively on the utilization of hydro-metallurgical method for precious metals recovery from spent Zn–C batteries [7, 12, 13]. For example, Baba et al. [7] developed a combined pyro- and hydro-metallurgical route to treat spent zinc–carbon batteries by combinations of leaching and solvent extraction. It was reported that 90.3% dissolution efficiency was achieved at optimized conditions and the calculated activation energy of 22.78 kJ/mol supported the dissolution mechanism. About 94.23% of zinc metal ion was quantitatively extracted within 25 min at 25 ± 2 °C. Nogueira and Margarido [14] carried out zinc extraction from spent batteries by ammonium chloride solution, about 70% zinc was extracted at optimized conditions, and the calculated activation energy was 5.7 ± 0.7 kcal/mol.

Economic Metals Rescue from Spent Zinc–Carbon …

51

Therefore, in this study, process energy evaluation through combinations of acid leaching, solvent extraction and precipitation techniques was adopted in the treatment of spent zinc–carbon batteries followed by the extraction of precious metals that could serve as a valuable material for some defined indigenous industries [7].

Materials and Method Materials and Physical Treatments The spent zinc–carbon batteries (Tiger Head type: China product) found abundant in Nigeria market were used for this investigation. The batteries prior to chemical processing contain plastic films, paper pieces and metallic materials. With thorough washing followed by air drying to remove dust particulates, the dismantling was done followed by ashing of the metallic components at 550 °C.

Chemical Treatment (i) Leaching test Leaching test was performed in a 500 ml Pyrex glass reactor equipped with a mechanical stirrer. For each run, 10 g/L of ashed battery was added to the predetermined hydrochloric acid solution at 55 °C at various leaching time intervals (5– 120 min). The concentration that gave the highest dissolution efficiency was utilized for optimization of reaction temperature. The residual products obtained at optimal leaching conditions were accordingly examined by ICP-MS and XRD analyses, respectively [7, 15]. (ii) Zinc recovery from leachate and solution beneficiation Solvent extraction investigations were performed using the leachate at optimized conditions. The leachate was made up of 5594.33 mg/L Zn, 70.40 mg/L Pb and 198.74 mg/L Fe with other metal ions occurring from low to trace levels. Batch methods of extraction were employed by equilibrating equal volumes of aqueous (leachate) and organic (Cyanex 272 extractant) phases, and the concentration of zinc was spectrophotometrically determined [7, 16]. The stripping of organic-loaded phase was carried out with 0.1 mol/L HCl solution, and three series of tests were performed to define the optimum conditions for zinc and other metal ions recoveries, respectively. Purified zinc solution was beneficiated to obtain industrial zinc oxide, suitable as coating materials in refrigerating devices [7, 17].

52

A. A. Baba et al.

Results and Discussion Materials Composition The purity test of the spent zinc–carbon batteries analyzed by XRD was found to majorly consist of ZnO (36-1451), Fe2 O3 (39-1346), MnO2 (44-0142), SiO2 (461045) and C (41-1487), respectively. The major chemical assay of the Zn–C batteries under investigation by XRF gave 41.30 wt% ZnO, 4.30 wt% Fe2 O3 and 2.69 wt% MnO2 .

Leaching Studies In this study, the dissolution of ash zinc–arbon batteries in hydrochloric acid solution was systematically investigated as a function of the following variables: (i) Concentration of HCl: The molarity of the HCl dissolving agent was varied between 0.5 and 8.06 mol/L at 55 °C between 5 and 120 min. The other parameters being kept constant (S/L = 10 g/L, particle size = 0.050 mm, stirring speed = 720 r.p.m). The percentage of metal extracted was noted at various leaching times at different HCl molarities as defined. At 4.0 mol/L HCl, the extent of ash battery dissolution reached 56.7%, and there was no appreciable increase in the extraction when the concentration was furthered to 8.06 mol/L HCl solutions when the reaction decreases to 52.4%. The possible decrease could be due to nature of un-dissolved, higher chloro-complexes formation at higher molarities beyond 4.0 mol/L solution [7]. (ii) Reaction temperature: The leaching reaction temperature was varied between 28 and 80 °C, the other parameters were kept constant (S/L = 10 g/L, [HCl] = 4.0 mol/L, particle size = 0.050 mm, stirring speed = 720 r.p.m). Thus, increasing the reaction temperature from 28 to 80 °C increases the fraction of dissolution from 29.3% to 90.3%, respectively. However, temperature beyond 80 °C was not considered due to excess loss of hydrochloric acid solution at higher temperatures. The composition of the leachate at optimal conditions was determined and used for solvent extraction studies. Processing of some spent zinc batteries by HCl solution at different conditions with extent of zinc recovery (Table 1) affirmed considerable energy utilization as compared to other reported data.

Economic Metals Rescue from Spent Zinc–Carbon …

53

Table 1 Leaching conditions and recovery of zinc and from spent zinc batteries Battery type

Conditions

Recovery (%)

References

Zinc–carbon batteries

0.25–2.0 M H2 SO4 ; 10–60 min.; 40–70 °C

93.3

Avraamides et al. [19]

Zinc–manganese batteries

1–2 M HCl; 50 min.; 50 °C

60–75

Li and Xi [20]

Alkaline batteries

0.3–0.7% v/v H2 SO4 ; 120 min.; 25–40 °C

60–70

De Souza and Tenorio [21]

Spent zinc–carbon batteries

0.5–8.06 M HCl; 5–120 min.; 28–80 °C

96.7

This study

Solvent Extraction and Beneficiation Studies Prior to solvent extraction investigations, the iron and other gangues found deterrent to zinc species were initially processed and removed from the leach liquor through combinations of precipitation and cementation techniques [7]. The results of iron and other associated impurities removal yielded 99%. Extraction of Zn(II) from the resulting solution performed in triplicates quantitatively extract 96.7% Zn by 0.032 mol/L Cyanex 272 extractant within 25 min [18]. The zinc metal elaboration after stripping from the organic-loaded phase was purified through calcinations at a temperature of 550 °C to obtain high-grade zinc oxide. The well-characterized zinc oxide (ZnO: 16-2340, melting point = 1971 °C) could find applications as coating and industrial raw materials as depicted in Fig. 1. Thus, the flow-sheet in Fig. 1 defines operational route for the recovery of pure zinc solution that could further beneficiate to zinc oxide suitable as coating and industrial raw materials for some defined industries.

Conclusions In this investigation, leaching tests with low energy were performed to recover Zn and other precious metals from spent zinc–carbon batteries. These tests showed that Zn material was quantitatively dissolved within 120 min by 4.0 mol/L hydrochloric acid concentrations at 80 °C when reaction reached 90.3%. The pregnant solution was appreciably treated by precipitation of iron and other associated gangues, prior to solvent extraction investigations. Thus, pure zinc was successfully extracted from the resulting solution and subsequently beneficiated to obtain 98.1% pure zinc oxide suitable as coating and industrial raw materials for some defined indigenous industries.

54

A. A. Baba et al.

Fig. 1 Hydro-metallurgical flow-sheet for economic metal rescue from spent zinc–carbon batteries for defined industrial applications

Acknowledgements The authors wish to thank Dr. Oliver Rouher and Mrs. Christine Salomon of Cytec Industries, Rungis Cedex, France, for their benevolence in supplying the Cyanex® 272 extractant used for this investigation. One of the authors (Prof. Alafara Abdullahi Baba, FCSN, FMSN) sincerely thanks the Director Training, Mining Cadastral Office, Abuja–Nigeria; KAM Industries (Nig.) Ltd., Ilorin, Kwara State: One of the leading indigenous Steel Industries in Nigeria; Saolad Nig. Ltd. (Building Materials Merchants), Palm Avenue, Mushin, Lagos and S. S. Amao and Sons Limited (Importer and Exporters of Motor Spare parts and facilities), Ijora, Lagos, Nigeria for their immeasurable support to attend the TMS 2020 149th Annual Meeting and Exhibition in San Diego, California, USA, February 23–27, 2020.

References 1. Belardi G, Lavecchia R, Medici F, Piga L (2012) Thermal treatment for recovery of manganese and zinc from zinc-carbon and alkaline spent batteries. Waste Manag 32:1945–1951 2. Nougueira CA, Margarido F TMS (2012) Battery recycling by hydrometallurgy: evaluation of simultaneous treatment of several cell systems. Energy Technol 227–234 3. Bernardes AM, Espinosa DCR, Tenorio JAS (2003) Collection and recycling of portable batteries: a worldwide overview compared to the Brazilian situation. J Powder Sources 124(2):586–592 4. Buzatu T, Popescu G, Birloaga I, Saceanu S (2013) Study concerning the recovery of zinc and manganese from spent batteries by hydrometallurgical processes. Waste Manag 33:699–705

Economic Metals Rescue from Spent Zinc–Carbon …

55

5. Biswas RK, Karmakar AK, Kumar SL, Hossain MN (2015) Recovery of manganese and zinc from waste Zn-C cell powder: characterization and leaching. Waste Manag (In Press). https:// doi.org/10.1016/j.wasman.2015.09.008 6. Shin SM, Senanayake G, Sohn J, Kang J, Yang D, Kim T (2009) Separation of zinc from spent zinc-carbon batteries by selective leaching with sodium hydroxide. Hydrometallurgy 96:349–353 7. Baba AA, Adekola FA, Bale RB (2009) Development of a combined pyro- and hydrometallurgical route to treat spent zinc-carbon batteries. J Hazard Mater 171:838–844 8. Belardi G, Ballirano P, Ferrini M, Lavecchia R, Medici F, Piga L, Scoppettuolo A (2011) Characterization of spent zinc-carbon and alkaline batteries by SEM-EDS, TGA/DTA and XRPD analysis. Thermochim Acta 526:169–177 9. Espinosa DCR, Bernardes AM, Tenorio JAS (2004) An overview on the current processes for the recycling of batteries. J Powder Sources 311–319 10. Jha MK, Kumar V, Singh RJ (2001) Review of hydrometallurgical; recovery of zinc from industrial wastes. Conserve Recycl 33:1–22 11. Salgado AL, Veloso AMO, Pereira DD, Gontijo GS, Salum A, Mansur MB (2003) Recovery of zinc and manganese from spent alkaline batteries by liquid-liquid extraction with cyanex 272. J Powder Sources 115:367–373 12. Frohlic S, Sewing D (1995) The batenus process for recycling mixed battery waste. J Power Sources 57:27–30 13. Rabah MA, Barrakat MA, Mahrous YS (1999) Recycling metals values hydrometallurgically from spent dry battery cells. J Metals 41–43 14. Nogueira CA, Margarido F (2015) Selective process of zinc extraction from spent Zn-MnO2 batteries by ammonium chloride leaching. Hydrometallurgy 157:13–21 15. Baba AA, Raji MA, Muhammed OM, Abdulkareem AY, Olasinde FT, Ayinla IK, Adekola FA, Bale RB (2019) Potential of a nigerian biotite-rich kaolinite ore to industrial alumina by hydrometallurgical process. J Metall Res Technol 116:222. https://doi.org/10.1051/metal/ 2018076 16. Gomez E, Estele JM, Cerda V, Blanco M (1992) Simultaneous spectrophotometric determination of metal ions with 4-(pyridyl-2-azo) resorcinol (PAR). Frensenius J Anal Chem 342:318–321 17. Pattnaik S, Mukherjee P, Barik R, Mohapatra M (2019) Recovery of bi-metallic oxalates from low grade Mn ore for energy storage application. Hydrometallurgy 189:105139. https://doi. org/10.1016/j.hyfromet.2019105139 18. Baba AA (2008) Recovery of zinc and lead from sphalerite, galena and waste materials by hydrometallurgical treatments. PhD Thesis, Chemistry Department, University of Ilorin, Ilorin, Nigeria, 675pp 19. Avraamides J, Senansyake G, Clegg R (2006) Sulfur dioxide leaching of spent zinc-carbonbattery scrap. J Power Sources 159:1488–1493 20. Li Y, Xi G (2005) The dissolution mechanism of cathodic active materials of spent Zn-Mn batteries in HCl. J Hazard Mater B 127:244–248 21. De Souza CCBM, Tenorio JAS (2004) Simultaneous recovery of zinc and manganese dioxide from household alkaline batteries through hydrometallurgical processing. J Power Sources 136:191–196

Characterization of the Hot-Pressed Coal Briquettes Prepared with the HyperCoal Yajie Wang, Haibin Zuo, Kaikai Bai, Jun Zhao and Jiansheng Chen

Abstract Due to the high compressive strength and good thermal ability, the hotpressed coal briquette prepared with the HyperCoal is considered to be an excellent alternative fuel for the lump coal in the COREX. In this paper, state changes of GDcoal and KL-HyperCoal during the heating process were observed in situ through an ultra-high temperature laser confocal microscope. Compared with GD-coal, KLHyperCoal has an excellent thermoplasticity, which contributes to the preparation of the hot-pressed coal briquettes with high mechanical strength. The Raman spectra and infrared spectra analysis were used to investigate the material structure of the hot-pressed coal briquettes. It is proved that increasing the hot-pressing temperature can improve the compressive strength, while extending the holding time period can hardly change the compressive strength. The above results indicate that HyperCoal can be used as a binder to effectively improve the performance of the hot-pressed coal briquettes. Keywords HyperCoal · Coal briquette · Hot-pressed · Infrared spectra · Raman spectra

Y. Wang · H. Zuo (B) · K. Bai · J. Zhao · J. Chen State Key Laboratory of Advanced Metallurgy, University of Science and Technology Beijing, Beijing 100083, China e-mail: [email protected] Y. Wang e-mail: [email protected] K. Bai e-mail: [email protected] J. Zhao e-mail: [email protected] J. Chen e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_6

57

58

Y. Wang et al.

Introduction Coal briquette is a coal product with certain shape and mechanical strength, which is formed by processing one or several kinds of pulverized coal with a certain proportion of binder or sulfur-fixing agent under a certain pressure. It can be generally divided into the civilian-type coal briquette and the industrial-type coal briquette according to the different applications [1–4]. The civilian-type coal briquette is mainly consisted of honeycomb coal briquettes with various qualities, which are generally made of anthracite and used for diet and heating. The industrial-type coal briquette mainly includes the coking coal briquette, gasified coal briquette fuel coal briquette, etc. The coking coal briquette is mainly obtained by two methods, the one is forming the weakly cohesive pulverized coal first and then coking it in a continuous coke oven, the other is forming the weakly cohesive pulverized coal directly by hotpressing during which the plastic mass is generated to bond it due to the high-speed pyrolysis. Developing the coking coal briquette can expand the coking raw materials and improve the efficiency of the coke oven. The coking coal briquette is playing a more important role recently although it was first developed in the middle of the twentieth century [5]. Benk and Coban [6] used coke breeze and anthracite as raw materials, added resole, novalac and coal tar pitch blended binder to make formed coke, found that either anthracite of lower volatile matter should be used, or it should be blended with coke breeze to produce anthracite briquettes of higher tensile strength. Mollah et al. [7–9] repeatedly tried to produce the coke with Victoria brown coal and found that the reactivity and specific surface area of coking coal briquette after carbonization were too high to be applied in the blast furnace although the coking coal briquette with a certain compressive strength was obtained by adding the coal-derived additives. Zhong et al. [10] used the Xylene activation of coal tar pitch as binder, combined with coke breeze as raw material to produce metallurgical coal briquette, and found that the compressive strength of carbonized coal briquette was better than the ordinary coke, and it would be after being further improved after being fed into blast furnace. In addition, the gasified coal briquette and fuel coal briquette have a relatively mature preparation process. The gasified coal briquette is generally a coal ball having a diameter of 35–50 mm and formed by anthracite having a particle size of less than 3 mm or powdered lignite having a tar yield of more than 10%. It is mainly used for the production of chemical fertilizer, semi-coke, liquid or gaseous fuels, methanol and acetic acid, etc. [11]. The fuel coal briquette mainly includes the boiler coal briquette and power coal briquette. About 60% of soot emissions can be reduced by replacing the raw coal with fuel coal briquette and 15–27% of raw coal can be saved. The sulfurfixing rate can reach 40–60% if a sulfur-fixing agent is added to the coal briquette, and both the energy saving and emission reduction are implemented. The development of power coal briquette needs to solve the key problems in the combustion process, such as using the temperature control to master the thermal deformation characteristics, using the batching technology to improve the coal quality, and improving the melting

Characterization of the Hot-Pressed Coal Briquettes Prepared …

59

point of the ash. More than 8% of raw coal can be saved using the power coal briquette developed in China. In our precious study, a new metallurgical hot-pressed coal briquette with the weakly cohesive pulverized coal as raw material and the HyperCoal as binder was prepared, which was proved to be an excellent alternative fuel for the lump coal in the COREX due to the high compressive strength and good thermal ability [12]. In this paper, different analytical methods were used to characterize the hot-pressed coal briquettes and discuss the reason why they have those such high mechanical strength.

Experiments Materials and Samples Preparation Two kinds of coals were used to prepare the hot-pressed coal briquettes, which were KL-coal and GD-coal, respectively. The samples were dried in a drying oven at 378 K for 4 h to remove the external moisture, and they were then crushed and sieved to less than 0.074 mm and 1 mm, respectively. The proximate and ultimate analysis of the samples is shown in Table 1, according to Chinese standards GB/T 212-91 and GB/T 476-91, respectively. KL-HyperCoal was prepared from KL-coal via a high temperature solvent extraction method, as shown in Fig. 1. The liquid-to-solid ratio was 50 mL/g, where KL-coal was 8 g and N-methyl-2-pyrrolidone (NMP) was 400 mL. High-purity argon with a gas flow rate of 400 mL/min was introduced into the reactor to purge the air for 15 min, and the reactor was heated to 623 K and held for 1 h. Stirring was continued to keep the solid and liquid phases in contact with a stirring rate of 100 r/min. After the reactor was cooled to 323 K, the solid–liquid mixture was removed and separated with a suction filter apparatus. KL-HyperCoal was finally obtained by distilling the liquid phase through a rotary evaporator. Figure 2 is a schematic of the hot-pressing device. The hot-pressed coal briquettes were prepared as the following steps. Different proportions of HyperCoal and GDcoal were weighed and mixed firstly. Five-gram sample was taken into the hot-pressed Table 1 Proximate analysis and ultimate analysis of samples Samples

Proximate analysis (dry, wt%)

Ultimate analysis (daf, wt%)

V

A

FC

N

C

S

H

O

KL-coal

32.20

10.37

57.43

1.80

83.25

1.09

5.37

8.49

KL-HyperCoal

49.51

0.49

50.00

4.31

82.49

0.55

5.50

7.15

GD-coal

14.31

9.14

76.55

1.37

89.03

1.35

4.20

4.05

Coal briquette

16.58

8.27

75.16

1.74

88.70

0.93

4.15

4.48

60

Y. Wang et al.

Fig. 1 Schematic diagram of the extraction device. 1 Temperature control device, 2 Gas cylinder, 3 Heating jacket, 4 Thermocouple 5 Air outlet, 6 Reactor, 7 Pressure gauge, 8 Mechanical stirrer

Fig. 2 Schematic diagram of the hot-pressing device. 1 Press, 2 Indenter, 3 High-temperature furnace, 4 Mold, 5 Bracket, 6 Computer, 7 Temperature control device, 8 Table

Characterization of the Hot-Pressed Coal Briquettes Prepared …

61

mold (diameter: 20 mm) and then placed in the high-temperature furnace. The sample was heated for a period of time before molding, where the molding pressure was 30 MPa. The hot-pressed coal briquettes were finally obtained after the furnace had cooled to room temperature.

Analytical Methods The thermoplasticity of coals was observed in situ through an ultra-high temperature laser confocal microscope (VL2000DX-SVF17SP-SVF15FTC, Lasertec, Japan). A steel sample was selected as the observed substrate. The observation methods are described as follows: First, the steel sample was cut into a number of small cylinders (ϕ7 × 5 mm), and then, they were mounted by hot-stamping. After being polished, samples were taken out and washed with alcohol. Coal samples, including KLHyperCoal and GD-coal, were crushed and sieved to 180–200 mesh, and a small amount of coal sample was evenly sprinkled onto the surface of the steel sample. Afterwards, the sample was heated to the set temperature under a vacuum atmosphere with a heating rate of 100 K/min, whose state changes were observed through the laser confocal microscope during the experiment. In addition, a high-resolution Raman spectrometer (LabRAM HR Evolution, HORIBA, France) and a Fourier infrared spectrometer (NEXUS, Thermo Nicolet Corporation, America) were used to analyze the material structure of the hot-pressed coal briquettes, respectively.

Results and Discussion Thermoplasticity The thermoplasticity of raw materials is an important performance that affects the mechanical strength of the hot-pressed coal briquettes, normally, the better the thermoplasticity, the higher the mechanical strength. The state changes of KL-HyperCoal during the heating process were observed in situ through an ultra-high temperature laser confocal microscope, as shown in Fig. 3. As the temperature increased, the behaviors of shrinkage and expansion simultaneously occurred in the surface, which facilitated the full contact of the coal particles and bonded them together tightly, resulting in the high mechanical strength of the hot-pressed coal briquettes. As shown in Fig. 3, the area of the coal particles in the lower-left corner of the images expanded several times, while the area of the coal particles in the middle of the images shrunk by a certain percentage, when the temperature of KL-HyperCoal was heated from 623 to 723 K.

62

Y. Wang et al.

Fig. 3 Microscopic images of KL-HyperCoal at different temperatures

In addition, the state changes of GD-coal during the heating process are shown in Fig. 4. Different from Fig. 3, almost no significant change happened in Fig. 4, indicating that GD-coal has a poor thermoplasticity. Due to the poor thermoplasticity, GD-coal cannot be used to prepare the hot-pressed coal briquettes alone. However, the hot-pressed coal briquettes with high mechanical strength were prepared by adding the KL-HyperCoal. The results indicate that KL-HyperCoal has an excellent thermoplasticity, which contributes to the preparation of the hot-pressed coal briquettes with high mechanical strength.

Infrared Spectra Analysis The infrared spectra analysis is usually used to analyze and identify the molecular structure and chemical bond of a substance due to the characteristic absorption bands [13, 14]. Based on the reference books, there are three types of absorption bands in the infrared spectra of the coals, which are the absorption band of aliphatic structure (720, 1380, 1460, and 2800–3000 cm−1 ), the absorption band of aromatic structure (750, 810, 870, 1500, 1600, and 3035 cm−1 ), the absorption band of heteroatom groups (950, 1084, 1112, 1182, 1243, 1321, 1680, 1700, 1745, and 3400 cm−1 ) [15]. The valley value represents the absorption intensity of the molecular structure and chemical bond because the ordinate is the transmittance, and the ratio of T 1 (2800– 3000 cm−1 )/T 2 (3400 cm−1 ) is used to characterize the degree of aromatization in

Characterization of the Hot-Pressed Coal Briquettes Prepared …

63

Fig. 4 Microscopic images of GD-coal at different temperatures

this paper. Figures 5 and 6 show the infrared spectra of the hot-pressed coal briquettes and the relationship between the compressive strength and T 1 /T 2 of the hot-pressed coal briquettes, respectively. As shown in Fig. 6, the transmittance ratio of T 1 /T 2 has a similar trend to the compressive strength of the hot-pressed coal briquettes, that is, the degree of aromatization tends to increase first and then remain unchanged with the increase of holding time periods, while it tends to increase continuously with the increase in hot-pressing

(b)

(a) 90 min

798 K 773 K

60 min

723 K

T, %

T, %

748 K

30 min

698 K 673 K

15 min

648 K

7.5 min T1 T2

500

1000

1500

2000

2500

3000 -1

Wavenumber, cm

3500

T1

4000

500

1000

1500

2000

2500

T2

3000

3500

4000

-1

Wavenumber, cm

Fig. 5 Infrared spectra of the hot-pressed coal briquettes. a Prepared at different holding time periods, b Prepared at different hot-pressing temperatures

64

Y. Wang et al.

0.95

120

0.94 0.93

80

0.92 15

7.5

30

60

90

350

0.98

0.97

300 250

0.96

T1/T2

0.96

Compressive strength, N

0.97

160

Compressive strength T1/T2

400

0.98

200

40

(b)

0.99

Compressive strength T1/T2

240

T1/T2

Compressive strength, N

(a)

200 0.95

150 100

0.91

648

673

Holding time period, min

698

723

748

773

798

0.94

Hot-pressing temperature, K

Fig. 6 Relationship between the compressive strength and T 1 /T 2 of the hot-pressed coal briquettes. a Prepared at different holding time periods, b Prepared at different hot-pressing temperatures

temperatures. A series of decomposition and polycondensation reactions take place inside the coal particles during the heating process, which means that the aliphatic structures of the hot-pressed coal briquettes decrease, while the aromatic structures increase with the increase of the temperature, leading to an increase in the degree of aromatization. Therefore, the compressive strength of the hot-pressed coal briquettes can be improved by increasing the hot-pressing temperature. However, extending the holding time period can hardly change the degree of aromatization so that the compressive strength of the hot-pressed coal briquettes was not improved.

Raman Spectra Analysis Raman spectra analysis is a non-destructive analytical way to acquire the detailed information on the chemical structure, phase and morphology, crystallinity, and molecular interactions of the sample [16]. Figure 7 shows the Raman spectra of the

(a)

(b)

90min

60min 30min

Intensity, a.u.

Intensity, a.u.

798 K

773 K 748 K 723 K 698 K

15min

673 K 648 K

7.5min

800

1000

1200

1400

-1

Raman shift, cm

1600

1800

800

1000

1200

1400

1600

1800

-1

Raman shift, cm

Fig. 7 Raman spectra of the hot-pressed coal briquettes. a Prepared at different holding time periods, b Prepared at different hot-pressing temperatures

Characterization of the Hot-Pressed Coal Briquettes Prepared …

65

hot-pressed coal briquettes, where D-band (1350–1380 cm−1 ) and G-band (1580– 1600 cm−1 ) represent the amorphous carbon and the graphitized carbon, respectively. Peak-fitting is very necessary to obtain the accurate data, so the original Raman spectra are divided into five peaks, where Fig. 8 is an example [17]. The intensity ratio of I G /I D is always used to characterize the degree of graphitization. Figure 9 shows the relationship between the compressive strength and I G /I D of the hot-pressed coal briquettes. As shown in Fig. 9, the intensity ratio of I G /I D is basically consistent with the compressive strength of the hot-pressed coal briquettes, where I G /I D peaked at 15 min and had a low slope at 673–748 K. Compared with the relationship between the transmittance ratio of T 1 /T 2 and the compressive strength, the intensity ratio of I G /I D has a better correspondence with the compressive strength. However, they both prove that increasing the hot-pressing temperature can improve the compressive strength of the hot-pressed coal briquettes while extending the holding time period can hardly change the compressive strength. Fig. 8 Raman spectra of the hot-pressed coal briquettes after peak-fitting

200

Observed data Generated data

Intensity/a.u.

150

G

D

100

D2

D1

50

D3 0 800

1000

1200

1400

1600

1800

-1

Raman shift/cm

(b)

1.46

250

200

1.42 150 1.40 100 1.38 50 0

20

40

60

Holding time period, min

80

1.36 100

Compressive strength, N

1.44

IG/I D

IG/ID

Compressive strength, N

Compressive strength

400

Compressive strength

350

IG/I D

1.6

1.5

300 250

1.4

IG/ID

(a)

200 1.3

150 100 640

1.2 660

680

700

720

740

760

780

800

Hot-pressing temperature, K

Fig. 9 Relationship between the compressive strength and I G /I D of the hot-pressed coal briquettes. a Prepared at different holding time periods, b Prepared at different hot-pressing temperatures

66

Y. Wang et al.

Conclusions Different analytical tools including the ultra-high temperature laser confocal microscope, high resolution Raman spectrometer and Fourier infrared spectrometer, were used to characterize the hot-pressed coal briquettes prepared with the HyperCoal in this paper. The thermoplasticity of raw materials is an important performance that affects the mechanical strength of the hot-pressed coal briquettes, so the state changes of HyperCoal and GD-coal during the heating process were observed in situ. The results indicate that KL-HyperCoal has an excellent thermoplasticity, which contributes to the preparation of the hot-pressed coal briquettes with high mechanical strength. The material structure of the hot-pressed coal briquettes was characterized by the infrared spectra analysis and Raman spectra analysis. Both the transmittance ratio of T 1 /T 2 and the intensity ratio of I G /I D have a good correspondence with the compressive strength of the hot-pressed briquettes, which prove that increasing the hot-pressing temperature can improve the compressive strength while extending the holding time period can hardly change the compressive strength. Acknowledgements The present work was supported by the National Natural Science Foundation of China (No. 51574023) and the National Key Research and Development Program (2016YFB0600701).

References 1. Zhi GR, Peng CH, Chen YJ, Liu DY, Sheng GY, Fu JM (2009) Deployment of coal briquettes and improved stoves: possibly an option for both environment and climate. Environ Sci Technol 43(15):5586–5591 2. Blesaa MJ, Fierro V, Miranda JL, Moliner R, Palacios JM (2001) Effect of the pyrolysis process on the physicochemical and mechanical properties of smokeless fuel briquettes. Fuel Process Technol 74(1):1–17 3. Blesaa MJ, Miranda JL, Izquierdo MT, Moliner R (2003) Curing temperature effect on mechanical strength of smokeless fuel briquettes prepared with molasses. Fuel 82(8):943–947 4. Tosun YI (2007) Clean fuel-magnesia bonded coal briquetting. Fuel Process Technol 88(10):977–981 5. Loison R, Foch P, Boyer A (1989) Coke: quality and production. Butterworths, London 6. Benk A, Coban A (2011) Investigation of resole, novalac and coal tar pitch blended binder for the production of metallurgical quality formed coke briquettes from coke breeze and anthracite. Fuel Process Technol 92(3):631–638 7. Mollah MM, Jackson WR, Marshall M, Chaffee AL (2015) An attempt to produce blast furnace coke from Victorian brown coal. Fuel 148:104–111 8. Mollah MM, Jackson WR, Marshall M, Chaffee AL (2016) Attempts to produce blast furnace coke from Victorian brown coal. 2. Hot briquetting, air curing and higher carbonization temperature. Fuel 173:268–276 9. Mollah MM, Marshall M, Sakurovs R, Jackson WR, Chaffee AL (2016) Attempts to produce blast furnace coke from Victorian brown coal. 3. Hydrothermally dewatered and acid washed coal as a blast furnace coke precursor. Fuel 180:597–605

Characterization of the Hot-Pressed Coal Briquettes Prepared …

67

10. Zhong Q, Yang YB, Jiang T, Li Q, Xu B (2016) Xylene activation of coal tar pitch binding characteristics for production of metallurgical quality briquettes from coke breeze. Fuel Process Technol 148:12–18 11. Bell DA, Towler BF, Fan MH (2011) Coal gasification and its applications. William Andrew, Oxford 12. Wang YJ, Zuo HB, Zhao J, Zhang WL (2019) Using HyperCoal to prepare metallurgical coal briquettes via hot-pressing. Int J Min Met Mater 26(5):547–554 13. Ferraro JR, Basile LJ (1985) Fourier transform infrared spectroscopy: applications to chemical systems. Academic Press, Pittsburgh 14. Ferraro JR, Krishnan K (1990) Practical fourier transform infrared spectroscopy: industrial and laboratory chemical analysis. Academic Press, Pittsburgh 15. Sun XG, Chen JP, Hao DH (2001) Micro-ftir spectroscopy of macerals in coals from the tarim basin. Acta Sci Natur Univ Pekinensis 6:832–838 16. Larkin PJ (2018) Infrared and Raman spectroscopy: principles and spectral interpretation. Elsevier, New York 17. Su XB, Si Q, Song JX (2016) Characteristics of coal Raman spectrum. J China Coal Soc 41(5):1197–1202

The Co-extraction of Low-Rank Coal and Biomass by Polar Solvent at Mild Conditions Jun Zhao and Haibin Zuo

Abstract In order to effectively and widely use low-rank coal and biomass, the co-extracted product was extracted from KL coal and biomass by polar solvent N-2methyl-2-pyrrolidinone (NMP) at high temperature and high pressure. The structures of the co-extracted product and residues were characterized by element analysis, FTIR spectra. The effects of temperatures, solid-liquid ratio, and extraction time on the extraction yields of co-extracted product were studied, respectively. Moreover, the extraction mechanism of the co-extracted product at mild conditions is proposed. The results show that the ash contents of the co-extracted products from KL coal and biomass are significantly lower than KL coal, and the volatile contents and H/C are significantly improved. For KL coal, with the increase of biomass/coal ratio, the extraction yield of KL is improved evidently from 56.25 to 71.44% and the ash content is decreased from 0.68 to 0.59% at suitable conditions. Keywords Low-rank coal · Extraction yield · Biomass · Thermal extraction

Introduction The ironmaking production consumes a large amount of high-quality coking coal and injection coal. The high-quality coal resources are becoming scarce and expensive, and unfortunately, a large majority of cheap low-rank coal cannot be effectively used. The successful utilization of low-rank coal in blast furnace production would broaden the energy source of ironmaking while increasing the economic benefits [1–3]. The thermal dissolution treatment of coal with an organic solvent is a promising method that would upgrade the utilization of low-rank coal [4–6]. There are many issues that restrict the development of thermal dissolution of low-rank coal when using an organic solvent, such as strict operation conditions, complex production processes, and high-cost. China is as an agricultural country that produces about 400 million J. Zhao · H. Zuo (B) State Key Laboratory of Advanced Metallurgy, University of Science and Technology Beijing, Beijing 100083, People’s Republic of China e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_7

69

70

J. Zhao and H. Zuo

tons of agricultural waste (equivalent to 200 million tons of standard coal) per year, which could be used for biomass energy [7]. Biomass accounted for about 70% of the total energy consumption of countryside; a majority of them are directly burned or discarded, and only a few are used to fuel ethanol, biodiesel, and biomass power generation. The biomass presents a serious environmental issue, as 4.9% of SO2 and 7.7% of NOx are discharged by direct burning [8]. The reasonable development and utilization of biomass and low-rank coal remain critical for reducing the environment population and promoting the energy structure [9]. There is an increasing interest in the co-thermal dissolution of coal and biomass with organic solvents to produce ash-free coal, which contributes to the highefficiency cleaning utilization of low-rank coal and biomass. Shui et al. [10] investigated the co-thermal dissolution of coal and biomass with nonpolar solvent 1methylnaphthalene (1-MN) at various temperatures, which found that the highest thermal dissolution yield (58.7%) was obtained at 320 °C. Increasing the temperature caused the thermal dissolution yield to decrease. This demonstrated the synergistic effect between the coal and the biomass during the process of co-thermal dissolution. The synergistic effects vary with the changes of coal type, solvent type, reaction condition, and the structural composition of the thermal dissolution products. It is necessary to perform in-depth research on this process and discuss the mechanism of the function between the biomass and the coal. In this study, the extraction behaviors of KL coal and biomass by NMP were investigated. The effects of temperature, biomass/coal ratio, and extraction time on the extraction yield were preliminary determined. The structures of thermally dissolve coals and residues were characterized by elemental- analysis and FT-IR spectra and the reaction mechanism of KL coal and biomass by NMP was proposed. This study aims to provide theoretical guidance for its industrial application.

Experimental Materials A sub-bituminous coal (KL) was used in this study. The biomass was obtained from the sawdust of China fir residue which is the most common type of biomass in Northern China. The coal samples and the biomasses were dried at 80 °C for 12 h in a vacuum and then grounded until they were less than 200 mesh (74 µm). The samples were deposited into sealed bags until they were used in the experiments. NMP (AR, >99.0%) and ethanol were purchased from Shanghai Aladdin Bio-Chem Technology Co., LTD. All solvents were used as commercial pure chemical reagents, without further purification. Gases were obtained from the cylinders with purities greater than 99%.

The Co-Extraction of Low-Rank Coal and Biomass by Polar Solvent …

71

Thermal Dissolution The thermal dissolution of KL coal and biomass was conducted using an apparatus for thermal dissolution, as shown in Fig. 1. A certain amount of coal sample, biomass, and 400 mL NMP was charged into a stain steel tuber cell. High-purity nitrogen was introduced into the autoclave reactor with a flow of 400 mL/min for 15 min to ensure inert atmosphere. The cell was heated to required temperature at a rate of 10 °C/min and held at that temperature for 60 min. The cell was then cooled to room temperature for 2–3 h. About 200 mL ethanol was put into the cell to wash the residue, and the filtrate was also collected into round-bottom flask; the residual solid was recovered, washed with ethanol, and dried at 80 °C for 12 h in vacuum. The filtrate was added to a rotary evaporator to recover organic solvent and precipitate soluble constituents. The soluble constituents were flushed repeatedly by ethanol, and then, the thermal extraction constituents (TECs) were dried at 80 °C for 12 h in vacuum. The yield of soluble constituents was calculated according to Eq. (1): Y, % =

Mr − Mc × 100% Mr × (1 − Ad)

(1)

where Mr (g), Mc (g), and Ad (%) were the initial mass of the coal, the mass of the residue, and the ash content of the initial coal, respectively.

Fig. 1 Schematic diagram of apparatus for thermal dissolution. 1 Temperature control unit, 2 Gas cylinder, 3 Heating jacket, 4 Thermocouple, 5 Gas outlet, 6 High-pressure reactor, 7 Pressure gauge, 8 Mechanical agitator

72

J. Zhao and H. Zuo

Results and Discussion Physical and Chemical Property of Raw Samples The results of proximate analysis, the ultimate analysis, the particle size distribution, the XRD, and the FTIR spectra demonstrated that there were some important differences between the chemical and the structural properties of the coal and the biomass. As shown in Table 1, the biomass obtained a higher volatile content and a lower fixed carbon than the coals. The biomass samples had higher H/C and O/C than the coals, which suggested that the biomass had more aliphatic and obtained thermal dissolution easier. It was proposed that the blending of biomass and coal enhanced the thermal dissolution yield. The particle sizes of the biomass and KL coal are shown in Fig. 2. The particles of biomass ranged from 1 to 120 µm, while KL coal was small, falling in between 1 and 60 µm.

Effect of Temperature on Extraction Yield and Ash Content Figure 3 shows the extraction yield of KL coal and biomass in NMP at different temperatures. Along with the raising of thermal dissolution temperature, both the extraction yield of KL coal and biomass increased rapidly, and the maximum yield (90.05%, 56.25%) of biomass and KL coal was, respectively, obtained at 380 and 350 °C. NMP was a polar solvent, the high extraction yield of biomass by NMP might be attributed to the result of heat-induced and solvent-induced structural relaxation followed by dissolution of biomass in the NMP. While KL coal showed a relatively low extraction yield compared with biomass, it might be ascribed to the high degree of aromatic, which was hard to be extracted by NMP. With an increasing of temperature, the ash content of biomass and KL coal slightly decreased, and the minimum value of ash content for biomass and KL coal was, respectively, achieved at 350 and 380 °C. Based on the above results, the best extraction behaviors of biomass and KL coal could be found at 350 °C; the effect of the coal/biomass ratio and thermal dissolution time on the co-thermal dissolution yield of KL coal and biomass was investigated at 350 °C.

Effect of Biomass/Coal Ratio on the Extraction Yield and Ash Content In order to investigate the amount of biomass used and improve the economy of the thermal dissolution of KL coal, effects of the biomass/coal ratio on the extraction yield and ash content of KL coal were studied, and the results are shown in Fig. 4. The thermal dissolution was carried out at 350 °C for 1 h with 400 mL NMP and a

10.37

KL coal

32.2

80.76

57.34

17.87

db dry basis, daf dry ash-free basis, diff difference

1.37 76.87

50.19 5.28

5.86 15.01

43.51

Odiff

C

H

Ultimate analysis/wt% (daf)

FC

A

V

Proximate analysis/wt% (db)

Biomass

Samples

Table 1 Proximate and ultimate analysis of coal samples

1.77

0.39

N

1.07

0.06

S

0.82

1.40

H/C

0.15

0.65

O/C

Atomic ratio N/C 0.02

0.01

The Co-Extraction of Low-Rank Coal and Biomass by Polar Solvent … 73

74

J. Zhao and H. Zuo

75

Cumulative distribution (%)

6

Cumulative distribution interval distribution

4

Biomass

50

2

25

0

0 100

KL coal

4

75 50

2

interval distribution (%)

100

25 0

0 0.1

1

10

100

1000

particle diameter (um)

Fig. 2 The particle diameter distributions of the different samples

50

80

Extraction yield/%

60

0.3

70 60 50

0.2

40 30

0.1

20

1.8

Extraction yield Ash content

1.6 1.4

40

1.2 1.0

30

0.8

20

0.6 0.4

10

10 0

2.0

KL coal

280

300

330

350

380

0.0

0

Ash content/%

0.4

Extraction yield Ash content

Biomass

Ash content/%

90

Extraction yield/%

100

0.2 0.0

280

300

330

350

380

TemperatureoC

o

Temperature C

Fig. 3 Effect of temperature on extraction yield and ash content

Extraction yield/%

90

1.0

Extraction yield Ash content

biomass-KL

0.9

80

0.8

70

0.7

60

0.6

50

0.5

40

0.4

30

0.3

20

0.2

10

0.1

0

0.0 0

20

50

Biomass/%

80

100

Ash content/%

100

Fig. 4 Effect of biomass/coal ratio on the extraction yield and ash content

The Co-Extraction of Low-Rank Coal and Biomass by Polar Solvent … Fig. 5 The FT-IR of co-thermal dissolution of the biomass and KL coal at various ratios

75

Absorbance

KL 20% biomass 50% biomass 80% biomass Residue (50% biomass)

500

1000

1500

2000

2500

3000

3500

4000

Wavenumber/cm-1

varied amount of biomass. It can be found that the extraction yield and ash content of KL coal reached to 71.44% and 0.59%, respectively, during the ratio of biomass/coal was 1/1. With an increasing of biomass/coal ratio, the extraction yield of KL coal significantly increased and the ash content of KL coal slightly decreased. This was because the additional free radicals required from the biomass could be used to loosen the carbon structure of the KL coal, and a great deal of light hydrocarbon components were released. The FT-IR spectra of the co-thermal dissolution of biomass and KL coal at various ratios of KL coal to biomass are shown in Fig. 5. The intensity of absorption peaks at 1700 cm−1 (C=O) and 2858 cm−1 (aliphatic CH2 and CH3 ) increased with the rising of biomass/coal ratio. The maximal intensity of these absorption peaks were found at 80% biomass, while the intensity of the narrow bands at 3050 cm−1 , which often attributed to aromatic C–H, decreased as the content of the biomass increased. Comparing the FT-IR spectra of soluble constituents to the residue at equal percentage of KL coal and biomass showed that intensity of the absorption peaks for aliphatic and oxygen-containing groups weakened in residue, and the absorption peak at 3600 cm−1 (miner) and 3050 cm−1 (aromatic C–H) were weakened in TECs, suggesting that the light compounds were dissolved in the NMP, and the heavy compounds and minerals were left in the residue at the co-thermal dissolution process.

Effect of Thermal Dissolution Time on the Extraction Yield and Ash Content As shown in Fig. 6, the effect of thermal dissolution time on the extraction yield and ash content was investigated at 50% biomass. It can be found that the thermal

76

J. Zhao and H. Zuo 80

Extraction yield Ash content

70

Extraction yield/%

60

1.0

0.8

50 0.6

40 30

0.4

Ash content/%

Fig. 6 Effect of thermal dissolution time on the extraction yield and ash content

20 0.2 10 0

30

45

60

75

90

0.0

Time/min

dissolution time significantly affected the extraction yield and ash content. When the thermal dissolution time is less than 60 min, the extraction yield increased and ash content decreased with the prolonging of time. While continuing to increase the time, the extraction yield and ash content were basically maintained at about 71% and 0.55%. Thus, it can be concluded that the optimum thermal dissolution time is 60 min. In conclusion, the optimized process conditions were: thermal dissolution temperature 350 °C, thermal dissolution time 1 h, the ratio of biomass/coal 1:1.

Conclusions (a) The ash content can be significantly reduced by the thermal dissolution process, and the combustion property of TEC can be effectively improved based on the increasing of aliphatic. (b) KL coal and biomass have a good reaction activity with NMP. The thermal dissolution yield of KL coal increased from 56.25% to 71.44% and the ash content of KL coal decreased from 0.68% to 0.59% as the addition of biomass (c) The optimized process conditions are: thermal dissolution temperature 350 °C, thermal dissolution time 1 h, the ratio of biomass/coal 1:1. (d) The proposed mechanism of co-thermal dissolution between KL coal and biomass by NMP is deduced, and the main reaction is the formation of hydrogen bonds between NMP and light micromolecular components. Acknowledgements This work was financially supported by the National Natural Science Foundation of China (NO. 51574023) and National Key Research and Development Program (2016YFB0600701).

The Co-Extraction of Low-Rank Coal and Biomass by Polar Solvent …

77

References 1. Xiong G, Li Y, Jin L (2015) In situ FT-IR spectroscopic studies on thermal decomposition of the weak covalent bonds of brown coal. J Anal Appl Pyrol 115:262–267 2. Kashimura N, Takanohashi T, Saito I (2006) Upgrading the solvent used for the thermal extraction of sub-bituminous coal. Energy Fuel 20(5):2063–2066 3. Shui H, Zhao W, Shan C (2014) Caking and coking properties of the thermal dissolution soluble fraction of a fat coal. Fuel Process Technol 118:64–68 4. Takanohashi T, Shishido T, Saito I (2008) Effects of HyperCoal addition on coke strength and thermoplasticity of coal blends. Energy Fuel 22(3):1779–1783 5. Rahman M, Samanta A, Gupta R (2013) Production and characterization of ash-free coal from low-rank Canadian coal by solvent extraction. Fuel Process Technol 115:88–98 6. Masaki K, Kashimura N, Takanohashi T (2005) Effect of pretreatment with carbonic acid on “HyperCoal” (ash-free coal) production from low-rank coals. Energy Fuel 19(5):2021–2025 7. Wang YN, Wei XY, Li ZK (2017) Extraction and thermal dissolution of Piliqing subbituminous coal. Fuel 200:282–289 8. Pan CX, Liu HL, Liu Q (2017) Oxidative depolymerization of Shenfu subbituminous coal and its thermal dissolution insoluble fraction. Fuel Process Technol 155:168–173 9. Takanohashi T, Iino M (1988) Extraction of argonne premium coal samples with carbon disulfide-N-methyl-2-pyrrolidinone mixed solvent at room temperature and ESR parameters of their extracts and residues. Fuel 67(12):1639–1647 10. Kim SD, Woo KJ, Jeong SK (2008) Production of low ash coal by thermal extraction with N-methyl-2-pyrrolidinone. Korean J Chem Eng 25(4):758–763

Discussion on the Application of Rooftop Photovoltaic Power Plant in the Steel Enterprise Xiancong Zhao, Huanmei Yuan, Yuzhao Han, Zefei Zhang and Hao Bai

Abstract In recent years, sustainable energies such as the solar and wind energy were widely applied to substitute traditional energies in the industrial sector. An emerging trend is that plenty of photovoltaic (PV) power plants were installed on the roof of the factory buildings in steel enterprises. In this paper, we reviewed the recent development of rooftop PV power plants and discussed the collaborative scheduling between rooftop PV power plant and existed power system. A case study was conducted in a steel enterprise with annual capacity of 8 million tons of steel in China. The design, cost, and benefits of installing rooftop PV power plant are investigated. Results demonstrate that the annual power output is around 20 million kWh, which can cover 5–10% of the total power consumption of the plant. The payback period of the PV power plant is seven years which is economically feasible. Keywords Rooftop photovoltaic power plant · Collaborative scheduling · Payback period · Steel enterprise

Introduction In recent years, with the aim of developing the low-carbon economy, sustainable energies such as the solar and wind energy were widely used to substitute traditional energies from the energy intensive industries (EII) [1]. As one of the most EII in the world, the steel making industry is seeking for opportunities to integrate renewable energy into its current energy system [2]. An emerging trend is that many photovoltaic X. Zhao Department of Industrial Engineering & Management, Peking University, Beijing 100871, China e-mail: [email protected] H. Yuan (B) · Y. Han · Z. Zhang · H. Bai State Key Laboratory of Advanced Metallurgy, University of Science and Technology Beijing, 30# Xueyuan Road, Beijing 100083, China e-mail: [email protected] School of Metallurgical and Ecological Engineering, University of Science and Technology Beijing, 30# Xueyuan Road, Beijing 100083, China © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_8

79

80

X. Zhao et al.

(PV) systems are installing on the roof of the factory’s buildings and connected to the power networks of steel plants [3, 4]. In developed countries such as the US, some small-scaled PV systems have already installed steel plants in Alabama and Minnesota [5, 6]. Taken Louis Industries as an example, the family-owned steel manufacturing plant is using more than 1200 solar panels to power its facility. The 498 kW PV system, which was activated in 2016, is capable of producing about 625,000 kWh of electricity per year. Some PV systems were equipped with battery banks for energy storage to solve the problem of fluctuated and periodic nature of PV generation. For example, the IISCO Steel Plant in Burnpur (India) had installed a group of tubular lead–acid batteries together with the 160 kW PV system [7]. Some larger PV systems were installed in developing countries such as India and China. The Rourkela Steel Plant in India had set up a 1 MW PV generation unit in 2015 [8], at a cost of 1.03 million dollars. In addition, the plant is also in the process of setting up a 15 MW hydropower project in the downstream of Mandira dam in collaboration with a local construction company. On the other hand, the capacity of PV system installed in China’s steel plants is even higher because of larger production scale and building area. Thus, more roofs can be used for PV installation. The Baosteel had set up a 50 MW PV system in 2012, which belongs to a governmental project which aims to increase the application of solar energy, with 50% of construction cost compensated by the national energy administration [9]. The PV system in Baosteel generated more than 15 million kWh of electricity in 2013. Other large-sized PV systems were installed or under construction in WISCO, Xuan Steel, and Mei Steel [10–12], with a total capacity of 84 MW. The list of steel companies installed with PV system is given in Table 1. Driven by technological improvement that reduced the solar PV cost and the policy support from the government’s side, it is expected that the business of PV installation will continue expanding in steel industry in the following years [13]. Table 1 List of steel companies installed with PV system

Company

Country/Region

Year

Capacity (MW)

Baosteel

Shanghai, China

2012

50

Apel Steel Corporation

Alabama, USA

2014

0.34

IISCO Steel Plant

Paschim Bardhaman, India

2014

0.16

Rourkela Steel Plant

Sundargarh, India

2015

1

Louis Industries

Minnesota, USA

2016

0.498

WISCO

Wuhan, China

2017

14

Discussion on the Application of Rooftop Photovoltaic …

81

The Application of PV System Challenges the PLS in Steel Plant One of the advantages of integrating PV system in steel plant is that the power purchased from the main grid would be reduced. In addition, the peak load shifting (PLS) of power generation would be realized through the installation of battery banks for the storage of electricity generated by PV systems. Generally, the byproduct gas system include byproduct gas production system, main process gas consumers, storage system, and cogeneration system; the on-site power generation is mainly relied on the boiler system and the CCPP system; the composition of on-site power generation becomes more complicated after the introduction of PV power generation system, as shown in Fig. 1. The PLS in the steel plants can be implemented through the optimal collaboration between the on-site power plant (OSPP) and the byproduct gasholders under timeof-use (TOU) power price. The TOU tariff is a kind of dynamic tariff that is widely implemented in industrial sector [14]. Through storing more byproduct gases in gasholders when power price was cheap and releasing them for power production when power price was expensive, the PLS can help decrease the electricity purchasing cost by 10–30% [14]. Furthermore, if the application of PV system and battery storage are considered, better performance of PLS would be realized through optimal allocation of the battery storage and scheduling of the combined power generation. Some studies had been conducted on PLS topic in the field of steel making process. He et al. [15] proposed a static model to evaluate the economic and environmental

Fig. 1 Schematic view of the byproduct gas and power system with PV and BESS

82

X. Zhao et al.

advantages of applying PLS in the steel industry in China. Results demonstrated that the economic benefits can reach 0.93 billion dollars annually for such a Chinese steel industry. Maneschijn et al. [16] assessed the load shifting potential in a steel plant in South Africa. In such a study, the time-of-use (TOU) energy tariff was considered. Results indicated that a 2 MW of additional generation capacity can be shifted from the valley price period (VPP) to the peak price period (PPP), with an annual saving of approximately 2.3 million dollars. Zhao and Bai [17] considered the effect of load operation on boiler’s efficiency and proposed an optimal scheduling model for the PLS operation under TOU tariff. The regression data demonstrated that the efficiency of gas-fired boiler could decrease from 80% to 40% if the operation load decreased from 100 to 30%. Therefore, it is unreasonable to assume a fixed efficiency for a gas-fired boiler operating under PLS. The optimal scheduling results show that the electricity cost can be reduced by 29.7%.

Case Study The above content discussed the application of rooftop PV power plant and PLS in steel enterprises. However, very limited study had been done on the capacity selection and economic evaluation of rooftop PV power plant in steel industry. In this section, the design, cost, and benefits of installing rooftop PV power plant will be investigated. Case study was conducted in a steel enterprise with annual capacity of 8 million tons of steel in China, named as M company for short. The company is located in Qian’an city with global solar irradiance of around 1600 kW/m2 and annual sunshine hours of around 2500 h, as can be seen in Fig. 2. The Qian’an city has concentrated a large number of steel enterprises with annual total steel production almost equals to Germany. The city also has good solar energy resources, and there is no extreme weather condition such as the typhoon. Therefore, the investigation on the feasibility of rooftop PV systems for steel enterprises in this area is very representative.

Capacity Evaluation for Rooftop PV System The available roof area should be calculated first in order to determine the installed capacity of a rooftop PV power plant. Steel companies have a large number of production plants and other building roof resources, but not all plants are suitable for building rooftop PV power generation projects. For example, plants with high dust content are not suitable for PV installation, such as the pellet plant and sinter plant. On the other hand, thermal processing plants such as continuous casting plant and hot rolling plant will reduce the conversion efficiency of PV modules due to the high roof temperature of the plant.

Discussion on the Application of Rooftop Photovoltaic …

83

Fig. 2 Distribution of solar energy sources in China [18]

According to our investigation, the available roof area of M company is about 150,000 m2 . Because the structure of roofs for steel enterprises is almost colored steel, for the consideration of roof load, the PV modules are not installed at the best angle towards the sun, but ties with the roof. Therefore, there is no shadow zone in the front and rear PV modules. It is essential to leave 500–600 mm distance for maintenance channel. In brief, the total installed capacity can be estimated by Eq. (1). P =C×A

(1)

where P stands for capacity, Wp , C represents the constant coefficient of the roof structure, and A stands for effective area of the roof, m2 . It can be calculated that the capacity of rooftop PV modules for M company is 15 MW.

84

X. Zhao et al.

Selection of PV Module Serial Number and Inverter The rooftop PV power system consists of PV array, combiner box, inverter, transformer, controller, and power grid, as illustrated in Fig. 3. In order to install a 15 MW PV power system, 546 pieces of 275 Wp polycrystalline silicon solar modules are required. In the PV array, the electrical performance parameters of each PV module in the same PV module string should be consistent, and the series number of PV module strings should be calculated according to the following Eqs. (2) and (3): N≤

Vdc max

Voc × [1 + (t − 25) × K V ]

V

V

mppt min mppt max ≤N≤    Vpm × 1 + (t  − 25) × K V Vpm × 1 + (t − 25) × K V

(2) (3)

where K v represents the temperature coefficient of the open circuit voltage of the PV module; K V stands for the temperature coefficient of the working voltage of the PV module; N represents the number of series connection of the PV module; t stands for the ultimate low temperature under the working conditions of the PV module, °C; t  represents the maximum temperature under the working conditions of the component, °C; V dc max stands for the maximum DC input voltage allowed by the inverter, V; V mmpt max represents the maximum MPPT voltage of the inverter, V; V mmpt min stands for the minimum value of the MPPT voltage of the inverter, V; V oc stands for the open circuit voltage of the PV module, V; V pm represents the operating voltage of the PV module, V. In the rooftop PV power generation system, the configuration capacity of the inverter should match the installation capacity of the PV array. The maximum DC input power allowed by the inverter should not be less than the actual maximum DC output power of the corresponding PV array. Meanwhile, the maximum power operating voltage variation range of the PV module string should be within the maximum power tracking voltage range of the inverter.

Fig. 3 Grid-connected rooftop PV power generation system

Discussion on the Application of Rooftop Photovoltaic …

85

Calculation of Power Generation The power generation of PV system can be calculated through simulation software. For example, the PVsyst and RETScreen can be calculated through Eq. (4). E P = HA ×

PAZ ×K ES

(4)

where E p is the amount of generated electricity, kW h; H A is the global horizontal solar irradiance of the location, kW h/m2 ; E s is the irradiance under standard conditions, kW h/m2 ; PAZ is the component installation capacity, kWp ; K is the comprehensive efficiency coefficient. The overall efficiency coefficient was set as 78.7% [19], and the power generation for the first year was 19.127 million kWh after calculation.

Economic and Environmental Benefits Analysis In recent years, with the continuous improvement of China’s PV production technology, the investment cost of PV power plants has also dropped rapidly. In 2017, the cost of PV power generation has dropped to around 7 yuan/W, and the component cost has dropped to around 3 yuan/W. According to the latest market survey in 2018, the cost of polycrystalline silicon PV power generation is now around 5–6 yuan/W, which is 5.5 yuan/W for average. The total investment of M company’s 15 MW PV power generation system is about 82.5 million yuan. Generally, the generation period of the PV power system is set as 25 years. The maximum limit of crystalline silicon PV modules is calculated according to the system’s 25-year power output attenuation, which is set as 20%, and the average annual attenuation is calculated by 0.8%. The PV power plant attenuation of the first year is 2.3%, and the annual attenuation is 0.7%. In 2018, the national subsidy price for electricity is 0.37 yuan/kW h, and the period is 20 years. However, for PV power plants connected to the grid after the end of 2017 will no longer receive provincial subsidy. The local desulfurization benchmark price of M company is 0.372 yuan/kW h, and the annual operating cost is 15%. The annual net income of distributed PV power plants is shown in Fig. 4. The total power generation and total benefit in 25 years of this PV power plant are 440.017 million kWh and 251.78 million yuan, respectively, while the annual average power generation and average net income are 17.601 million kWh and 100.71 million yuan, respectively. The investment recovery period is seven years. Meanwhile, the project can save 7040.4 t/a of standard coal, reduce carbon emissions by 4787.5 t/a, CO2 emissions by 17548.2 t/a, SO2 emissions by 528.0 t/a, and NOx emissions by 264.0 t/a.

86

X. Zhao et al.

Fig. 4 Annual power generation and net benefit of PV power station

Conclusions This paper reviewed the emerging trends of PV installation in steel plant and discussion the challenges that brought to the PLS. Due to the environmental pressure and policy encouragement, it can be expected that the installation of PV system will continue expanding in steel industry in the following years. In this paper, the installation of rooftop distributed PV power plants in iron and steel enterprises, including analysis of solar energy resources, selection of PV operation modes, determination of installed capacity of PV power plants, series connection of PV modules and inverters, is investigated. Case study was taken in a typical steel company in Qian’an city in North China. The city concentrated a large number of steel enterprises with annual total steel production almost equals to Germany. Results demonstrate that the annual power output is around 20 million kWh, which can cover 5–10% of the total power consumption of the plant. The payback period of the PV power plant is seven years which is economically feasible. Acknowledgements This research was supported by Boya post-doctoral project of Peking University.

References 1. Birgit F, Nagore S, Neil S (2016) The critical role of the industrial sector in reaching long-term emission reduction, energy efficiency and renewable targets. Appl Energ 162:699–712 2. Abdul Q, Norhuda AM, Ali A (2017) Analysis of the integration of a steel plant in Australia with a carbon capture system powered by renewable energy and NG–CHP. J Clean Prod 168:97–104

Discussion on the Application of Rooftop Photovoltaic …

87

3. The Hindu (2016) Steel Secretary inaugurates solar power plant at Visakhapatnam steel plant (Online). Available: http://www.thehindu.com/news/cities/Visakhapatnam/SteelSecretary-inaugurates-solar-power-plant-at-VSP/article16914968.ece 4. Zhang ZB (2015) Unidirectional slope type photovoltaic roof of steel structure factory building, China Patent, 204 418 460, Jun. 24, 2015 5. Clover I (2014) Off-grid solar system brings energy transition to Alabama steel company (Online). Available: https://www.pv-magazine.com/2014/08/18/off-grid-solar-system-bringsenergy-transition-to-alabama-steel-company_100016125/ 6. XcelEnergy (2016) Steel factory flips switch on solar array (Online). Available: https://www. xcelenergy.com/company/media_room/news_releases/steel_factory_flips_switch_on_solar_ array 7. Singh R (2015) The 43rd annual report of the steel authority of India. Steel authority of India ltd., New Delhi, Rep. 168122, Aug 2015 8. Canyon Consultancy (2013) Technical specifications for 1 MW solar photovoltaic power project at RSP. Canyon Consultancy ltd, Bhubaneswar, Rep. RSP/383/18109/12080420/TS/R-2, Oct 2013 9. National Energy Administration (2014) The 50 MW industrial plant roof photovoltaic power project for Bao steel, Shanghai (Online). Available: http://www.nea.gov.cn/2014-09/03/c_ 133617308.htm 10. Qian F (2014) The application of PV generation technology in the Bao steel. Shanghai Energy Conserv 12:3–6 (in Chinese) 11. China Daily (2017) Wuhan iron and steel cooperation enables the largest photovoltaic power plant in central China (Online). Available: http://hb.chinadaily.com.cn/2017-06/15/content_ 29753031.htm 12. Liu GL, Sun CQ, Hang QZ, Bao ZQ, Li J (2015) The application of distributed photovoltaic power generation technology in steel enterprises. Energ Res Util 4:31–34 (in Chinese) 13. Can S, Vasilis F (2014) Energy policy and financing options to achieve solar energy grid penetration targets: accounting for external costs. Renew Sust Energ Rev 32:854–868 14. Hao JX, Zhao XC, Bai H (2017) Collaborative Scheduling between OSPPs and gasholders in steel mill under time–of–use power price. Energies 10(8):1–10 15. He K, Zhu H, Wang L (2015) A new coal gas utilization mode in China’s steel industry and its effect on power grid balancing and emission reduction. Appl Energ 154:644–650 16. Maneschijn R, Vosloo JC, Mathews MJ (2016) Investigating load shift potential through the use of off-gas holders on South African steel plants. In: International conference on the industrial and commercial use of energy, Cape town, pp 104–111 17. Zhao XC, Bai H, Shi Q, Lu X, Zhang ZH (2017) Optimal scheduling of a byproduct gas system in a steel plant considering time-of-use electricity pricing. Appl Energ 195:100–113 18. World Bank Group (2017) Global solar atlas (Online). Available: https://globalsolaratlas.info/ downloads/china?c=36.633162,108.034923,4 19. Zhou XS (2017) Application of distributed photovoltaic power generation system in civil buildings. Telecommun Power Technol 4:38–40 (in Chinese)

Performance of Anodes with Proper Active Metal Elements Added to the Al–0.16wt%In in Alkaline Electrolyte for Al-Air Batteries Huimin Lu, Neale Neelameggham, Leng Jing and Jianxue Liu

Abstract Wind energy and solar energy are stored in aluminum through lowtemperature aluminum electrolysis, and then the distributed energy generation by metallic fuel cells realizes renewable energy utilization. Aluminum is an ideal material for metallic fuel cells. In this research, the performance of Al-air batteries based on pure Al, Al–0.16 wt%In, Al–0.16 wt%In–0.1 wt%Ga, Al–0.16 wt%In–0.5 wt%Bi, Al–0.16 wt%In–0.12 wt%Sn, and Al–0.16 wt%In–3 wt%Zn anodes in 4 M NaOH solution was investigated by galvanostatic discharge test. The electrochemical properties of the anodes were investigated in the same electrolyte using electrochemical impedance spectroscopy (EIS) and polarization curves. Battery performance was tested by constant current discharge at 20 mA cm−2 current density. The characteristics of the anodes after discharge were investigated by scanning electron microscopy (SEM) and energy dispersive analysis of X-ray (EDAX). Results confirm that compared with pure Al and Al–0.16 wt%In in 4 M NaOH, the electrochemical properties of Al–0.16 wt%In–0.1 wt%Ga anode restrains hydrogen evolution, improves electrochemical activity, and increases anodic utilization rate. Keywords Al-air battery · Self-corrosion · Aluminum alloy · Distributed energy generation

Introduction The Al-air battery which primarily comprises aluminum has numerous favorable properties, such as high energy density, low cost, innocuousness, abundance, and recyclability [1]. As promising power and energy storage devices, Al-air batteries can be widely applied in fields such as electric vehicles, navigation, and portable H. Lu (B) · L. Jing · J. Liu School of Materials Science and Engineering, Beihang University, Beijing 100191, China e-mail: [email protected] N. Neelameggham IND LLC, 9895 Dream Cir, South Jordan, UT 84095, USA e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_9

89

90

H. Lu et al.

sources [2–4]. However, super-pure aluminum is inapplicable for use as the anode of an Al-air battery in an uninhibited alkaline electrolyte, since (a) its surface is covered by a passive hydroxide layer resulting in high overpotential during anodic dissolution and (b) it suffers from high corrosion currents as water reduces on preferential surface sites evolving hydrogen. Alloying super-pure (99.999 wt%) aluminum with particular elements improves its electrochemistry or activates it. Studies on alloys for Al-air batteries have focused on alloying elements such as Mg, Zn, Pb, Sn, Ga, In, Mn, Hg, and Tl, because these elements possess a degree of solubility in aluminum matrix, have lower melting temperatures than aluminum (except for Mn), are more noble than aluminum (except for Mg), are soluble in alkaline electrolytes, and have a high hydrogen overpotential. The purpose of activation is (a) to increase the overpotential for the reduction in the water on the surface and (b) to reduce the overpotential for the oxidation by breaking down the passive hydroxide layer. Previous papers have investigated many kinds of alloys trying to analyze the mechanism of activation, such as the metal dissolution–deposition process and point defect mechanisms [5–8]. We studied the micro-grid system for renewable energy. As a renewable energy carrier, aluminum metal is discharged in an Al-air battery. In the low-temperature aluminum electrolytic cell (working temperature of 700–800 °C, the energy consumption of 9000 kWh/t aluminum), alumina electrolysis is used to charge into metal aluminum, and wind energy and solar energy are stored in aluminum. The low cost of aluminum produced with such a renewable energy system makes Al-air batteries more versatile. In this paper, Al-air batteries based on pure Al and Al–0.16 wt%In anodes are prepared. The reason for choosing Al–0.16 wt%In is that the upper limit for the indium concentration in a binary aluminum alloy for the use in Al-air batteries is 0.16%, which is close to the solid solubility limit for indium in aluminum at a heat treatment temperature of 640 °C. The corrosion behavior and battery performance are studied in 4 M NaOH solution. In contrast to pure Al, Al–0.16 wt%In possesses better performance considering the self-discharge and battery performance. Based on the above test, the purpose of our study is to research the feasibility of alloy anodes as the Al-air battery with proper active metal elements added to the Al–0.16 wt%In in an aqueous 4 M NaOH solution to further inhibit the self-corrosion and improve battery performance.

Materials and Methods Material preparation: Raw materials are super-pure (99.999 wt%) aluminum, indium ingots, zinc ingots, bismuth ingots, gallium particle, and tin particle (>99.9%), for casting the experiment alloys. The nominal compositions of the experiment alloys are Al–0.16 wt%In, Al–0.16 wt%In–0.1 wt%Ga, Al–0.16 wt%In–0.5 wt%Bi, Al– 0.16 wt%In–0.12 wt%Sn, and Al–0.16 wt%In–3 wt%Zn. The amount of the activating elements is small so that they are in solid solution with the aluminum matrix. Raw material ingots were cut, dried, and weighed the required amount of materials

Performance of Anodes with Proper Active Metal Elements Added …

91

and melted in a resistance furnace with a corundum crucible in the temperature range of 760 ± 10 °C. The molten alloys were poured in a graphite crucible. All alloys were heat treated at 650 ± 10 °C for at least 12 h followed by a water quench to achieve a solid solution alloy by suppressing precipitation of alloying elements. Electrochemical measurements: The samples were encapsulated in epoxy with only 1 cm2 exposed, ground with emery paper (grade 400-800-1000-2000 mesh sandpaper), and then cleaned with doubly distilled water. Next, they were dehydrated using absolute alcohol and dried with a hair dryer. The electrochemical tests were conducted using a standard three electrodes system employing a Gamry reference 3000 electrochemical workstation. The counter electrode was a platinum sheet 20 × 20 mm in size, and the reference electrode was a mercury/mercuric oxide (Hg/HgO) electrode. The solution used in this study was 4 M NaOH. Tafel polarization curves were measured at a rate of 1 mV s−1 after open-circuit potential (OCP) was measured more than 2 h to ensure the stability of the system. The voltage scan range of the anodic polarizing curve was −0.5 to 1 V versus OCP. EIS measurements were carried out at open-circuit potential with a 5 mV sine wave perturbation. The measuring frequency range was 100 kHz–0.1 Hz. Hydrogen evolution rate: The samples of self-corrosion tests were cut to 20 mm × 5 mm, then ground with emery paper (grade 400-800-1000-2000 mesh sandpaper), and immersed in 4 M NaOH solution for 1 h. Hydrogen collection was used by hydrogen collection equipment which was described elsewhere. The corrosion rate was calculated using the formula: Corrosion rate = Hydrogen volume/time of immersion/surface area (ml min1 cm−2 ) Battery performance tests: The batteries consist of anodes, cathodes, and electrolytes. The test samples constituted the anodes. The cathodes were air electrodes with a double-layer structure of gas diffusion and catalytic layers, laminated with a nickel mesh current collector. The catalyst was MnO2 powder. The electrolytes were 4 M NaOH. The experimental test unit was described elsewhere. Al-air batteries were tested by constant current discharge for 3 h at 20 mA cm−2 current densities using the LAND test system. After each test, the anode current efficiency and capacity were calculated based on the measurement of weight change before and after discharge. The fuel efficiency was calculated using the following formula: Fuel efficiency, η = I t/(m F/9.0) where η is the anodic efficiency, %; I is the discharge current, A; m is the weight loss, g; t is the time, s. F is the Faraday constant. The samples surface after discharge was examined using a CamScan 3400 scanning electron microscope (SEM). All the measurements were taken at room temperature (25 ± 5 °C) and repeated at least three times to ensure the test results would reproduce.

92

H. Lu et al.

Results and Discussion Hydrogen evolution rate: Fig. 1 and Table 1 show the hydrogen evolution rates of different alloys in 4 M NaOH solution. The following reaction explains why aluminum anodes in alkaline solution can generate hydrogen: 2Al + 6H2 O → 2Al(OH)3 ↓ +3H2

(1)

Hydrogen evolution rate of the Al–In alloy in 4 M NaOH solution is evidently inhabited to some extent in contrast to that of pure Al. The phenomenon could be explained by the fact that the hydrogen evolution overpotential of indium is higher than that of aluminum. In addition, hydrogen evolution rate of the Al–0.16In alloy can be further reduced by the addition of Ga. Hydrogen evolves slower because Ga possesses high overpotential for the hydrogen evolution reaction. Another possible reason is that the Ga present in the ternary alloy is oxidized to Ga2 O3 imparting some sort of passivity. However, hydrogen evolution rate of the other three alloys is higher than that of Al–0.16In alloy to varying degrees.

Fig. 1 Hydrogen evolution rates of different anodes in 4 M NaOH solution

Table 1 Hydrogen evolution rates of different anodes in 4 M NaOH solution for 1 h

Anode

Hydrogen evolution volume/ml

Hydrogen evolution rate/ml min−1 cm−2

Pure Al

28.1

0.468

Al–In

17.7

0.295

Al–In–Ga

14.2

0.237

Al–In–Bi

19.8

0.330

Al–In–Sn

20.6

0.343

Al–In–Zn

23.5

0.390

Performance of Anodes with Proper Active Metal Elements Added …

93

Potentiodynamic polarization: Figs. 2 and 3; Table 2 show the potentiodynamic polarization curves and corresponding corrosion parameters of different anodes measured in 4 M NaOH solution, respectively. The anodic polarization curves of all anodes exhibit current fluctuations between −1.32 and −1.05 V versus Hg/HgO, indicating that the alloy is alternating between a more active state and the state exhibited by pure aluminum. These fluctuations could have been caused by successive destruction and build-up of a passive hydroxide layer due to local variations in pH at the active sites. In addition, Al–In–Ga has the most negative corrosion potential and the least corrosion current density among these ternary alloys, which indicates that Al–In–Ga has the highest electrochemical activity with the most corrosion resistant and the smallest self-corrosion rate than that of the others. The polarization resistance of Al–In–Zn is the lowest, for the addition of zinc and indium to aluminum can produce holes in the oxide film of aluminum anode, and hence, decrease the resistance. Fig. 2 Potentiodynamic polarization curves for pure Al and Al–In alloy in 4 M NaOH solution

Fig. 3 Potentiodynamic polarization curves for Al–In-X (X = Ga, Bi, Sn, Zn) anodes in 4 M NaOH solution

94 Table 2 Corrosion parameters of different anodes in 4 M NaOH solution

H. Lu et al. Sample

E corr (V vs. Hg/HgO)

icorr (A cm−2 )

Pure Al

−1.55

2.83 × 10−2

Al–In

−1.78

1.18 ×

10−3

Al–In–Ga

−1.79

6.89 × 10−4

27.9

Al–In–Bi

−1.72

1.64 × 10−3

38.98

Al–In–Sn

−1.68

1.88 ×

10−3

Al–In–Zn

−1.68

5.69 × 10−3

Rp ( cm2 ) 6.492 38.12

33.8 11.01

Battery performance: Fig. 4 presents the discharge behavior of metal-air battery with different anodes at current density of 20 mA cm−2 . Table 3 summarizes the performance of the above batteries at 20 mA cm−2 . As shown in Table 3, the battery with Al–In–Ga displays higher operating voltage and anodic utilization than those of the others except for pure Al. The anodic utilization of pure Al is higher than that of the others and this is because of the low self-corrosion rate of Al resulted from the formation of the protective oxide film on Al surface. It should be noted that the operating voltage fluctuations of most alloys are higher than that of pure Al, which Fig. 4 Discharge behavior of different anodes in 4 M NaOH solution

Table 3 Discharge performance parameter with different anodes at 20 mA cm−2 current density in 4 M NaOH solution

Anodes

Operating voltage/V

Anodic efficiency/%

Pure Al

1.11

82.3

Al–In

1.24

37.1

Al–In–Ga

1.38

51.3

Al–In–Bi

1.20

35.6

Al–In–Sn

1.30

34.6

Al–In–Zn

1.35

32.0

Performance of Anodes with Proper Active Metal Elements Added …

95

indicates that the alloys are alternating between a more active state and the state by aluminum. A good Al-air battery should possess low anode self-corrosion, high negative corrosion potential, weak anodic polarization, and high anodic efficiency. Pure aluminum is still unsuitable as the anode of the alkaline Al-air battery due to its severe self-corrosion, even if it possesses a high anodic efficiency. As shown in Table 3, the Al-air battery based on the Al–In–Ga anode has an operating voltage of 1.38 V and anodic efficiency of 51.3% in 4 M NaOH. Comprehensively considering the above four factors, the best performance of the Al-air battery based on the Al–In–Ga anode in 4 M NaOH is obtained among all the samples. SEM analysis after discharge: The morphology of the anodes after discharge for 3 h is shown in Fig. 5, which is obtained by SEM examination. In Fig. 5, it can be seen that the morphology of pure Al after discharge is very flat. The pits on the pure Al surface were broad, shallow, and interconnected indicating the anodic current density was reasonably distributed across the whole aluminum surface. However, the morphology of the Al–In anode shows more pitting, indicating that the anodic current is concentrated at very small sites on the surface. As shown in Fig. 5c, the surface of Al–In–Ga alloy after discharge is porous and mainly consisted of aluminum, gallium, indium, and oxygen. These deposits are found at the base of pits on the alloy surface and on top of the surface hydroxide layer. This result indicates that the presence of both In and Ga deposits may facilitate the formation of a porous film of discharge product instead of the dense passivation film on the discharged alloy anode surface. The porous surface film promotes the transfer of reactive species between the electrolyte and the anode surface, and thus, the alloy anode might discharge stably in 4 M NaOH. Irregular pits were uniformly distributed onto the electrode surface, in agreement with operating voltage fluctuation. It should be noted that In deposits via the dissolution–deposition process are found on the surfaces of Al–In–Ga, Al–In–Bi, and Al–In–Sn anodes, but not on the surface of Al–In–Zn anode. It can be explained by the fact that the addition of zinc to the aluminum alloy decreases the number of the indium-enriched segregated phases and promotes indium alloying with aluminum.

Conclusions In this paper, the hydrogen evolution rate, electrochemical behavior, and discharge performance of Al–0.16 wt%In alloy with proper active metal elements are added to have been studied in 4 M NaOH solution. The results disclose that the self-corrosion of Al–In alloy in 4 M NaOH can be further inhibited by adding a proper amount of Ga into the alloy. The Al-air battery based on the Al–0.5 wt%In–0.1 wt%Ga anode has an operating voltage of 1.38 V and anodic efficiency of 51.3% in 4 M NaOH solution. By analyzing the morphology result of the Al–0.5 wt%In–0.1 wt%Ga anode in 4 M NaOH solution after discharge, it can be illustrated that the presence of both In and Ga deposits may facilitate the formation of a porous film, which could inhibit

96

H. Lu et al.

Fig. 5 SEM micrographs of different anodes in 4 M NaOH solution after discharge: a pure Al, b Al–In, c Al-n-Ga, d Al–In–Bi, e Al–In–Sn, f Al–In–Zn

the corrosion of the alloy anode. From the above discussion, it can be concluded that Al–In–Ga alloy is a promising candidate as the anode of Al-air batteries in alkaline solution. Acknowledgements This work was supported by a grant from the China Aerospace Science Fund.

Performance of Anodes with Proper Active Metal Elements Added …

97

References 1. Liu Y, Sun Q, Li W, Adair KR, Li J, Sun X (2017) A comprehensive review on recent progress in aluminum–air batteries. Green Energy Environ 2(2017):246–277 2. Egan DR, De Leon CP, Wood RJK (2013) Developments in electrode materials and electrolytes for aluminium-air batteries. J Power Sources 236:293–310 3. Zaromb S (1962) The use and behavior of aluminum anodes in alkaline primary batteries. J Electrochem Soc 109(12):1125–1130 4. Zhang X, Yang SH, Knickle H (2004) Novel operation and control of an electric vehicle aluminum/air battery system. J Power Sources 128(2):331–342 5. Kraytsberg A, Ein-Eli Y (2013) The impact of nano-scaled materials on advanced metal–air battery systems. Nano Energy 2(4):468–480 6. Vlaskin MS, Shkolnikov EI, Bersh AV (2011) An experimental aluminum-fueled power plant. J Power Sources 196(20):8828–8835 7. Smoljko I, Gudi´c S, Kuzmani´c N (2012) Electrochemical properties of aluminium anodes for Al/air batteries with aqueous sodium chloride electrolyte. J Appl Electrochem 42(11):969–977 8. Abiola OK, Otaigbe JOE (2009) The effects of phyllanthus amarus extract on corrosion and kinetics of corrosion process of aluminum in alkaline solution. Corros Sci 51(11):2790–2793

Theoretical and Experimental Research on the Mass Changes of Elements in Molten Steel with CO2 Used as RH Lifting Gas Baochen Han, Rong Zhu, Guangsheng Wei, Chao Feng and Jianfeng Dong

Abstract CO2 injection into RH as lifting gas was recently proposed instead of Ar. In this study, FactSage software was used for calculating the mass changes of elements in thermodynamic equilibrium with CO2 injection under vacuum condition. Compared with Ar, CO2 as RH lifting gas can be used for a small amount of decarburization without a significant increase in oxygen content of molten steel. And the carbon content of alloys can be theoretically increased by more than 12% if all CO2 participates in the reaction between CO2 and [C]. Furthermore, the industrial application research of CO2 injection into RH as lifting gas was carried out in a commercial 150t RH. The results agreed with the above theoretical trends. And the problem of aluminum loss can be solved by reducing the additive amount of aluminum alloy in the ladle furnace (LF) and replenishing the aluminum during the RH refining later stage. Keywords Carbon dioxide · RH refining · Mass change · Thermodynamic calculation · Industrial test

B. Han (B) · R. Zhu · G. Wei · C. Feng · J. Dong School of Metallurgical and Ecological Engineering, University of Science and Technology, Beijing, Beijing 100083, China e-mail: [email protected] R. Zhu e-mail: [email protected] G. Wei e-mail: [email protected] B. Han · R. Zhu · G. Wei Beijing Key Laboratory of Research Center of Special Melting and Preparation of High-End Metal Materials, University of Science and Technology, Beijing, Beijing 100083, China © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_10

99

100

B. Han et al.

Introduction Ruhrstahl–Heraeus (RH) refining reactor, which was originally invented for dehydrogenation from molten steel, is now being developed for decarburization, desulfurization, degassing, and removal of non-metallic inclusions of multi-function refining process. During RH refining process, molten steel is circulated between vacuum vessel and ladle owing to the lifting gas injected through snorkel nozzles. In the process of practical production, it should be paid more attention to shorten the degassing time, reduce the temperature drop of molten steel, and therefore, increase the yield and benefit of steel work [1, 2]. The flow and mixing behavior of molten steel plays a very important role in affecting the refining efficiency and the productivity [3–5]. Generally, decreasing the mixing time is accepted to enhance the refining efficiency [6–10]. Relevant research focuses on the optimization of refining operation parameters and dimensions of RH equipment. Park [5] and Korneev [11] reported that it can accelerate the molten steel circulation with RH equipped with large-diameter round snorkel, according to the results of numerical calculations and tests, respectively. Some scholars investigated the refining effect with conventional snorkels replaced by the oval snorkels [12, 13] or multi-snorkels [14–16]. However, it is difficult to modify the existing equipment for steel enterprises due to the technical limit. Compared with equipment modification, the optimization of refining operation parameters is much easier accepted to factories due to its characteristics of more effective, more practical, and lower cost. Silva [8] found that auxiliary gas injection through the ladle bottom could improve the liquid circulation rate. And then, research [4, 17, 18] on bottom blowing gas was launched by using physical and numerical simulation. In our recent work [19], the method of CO2 injection instead of Ar as RH lifting gas was proposed. On the basis of meeting refining requirements, the utilization of CO2 can achieve two advantages, one of which is lowering the cost and another is that the CO2 captured from steel factories will reduce CO2 emissions. In present study, FactSage 7.1 software was used to investigate the changes of [PctC] and [PctO] in molten steel before and after RH refining, the composition of refining off-gas and the influence on carbon content in alloys with the introduction of CO2 gas. In order to better control the variables, the molten steel without decarburization during RH refining was chosen as test material. And industrial trials were carried out to investigate the variation of composition of molten steel during RH refining with Ar and CO2 injected as lifting gas, respectively.

Thermodynamic Calculations with CO2 Injection During RH Refining Process In this section, the variation of carbon and oxygen content in molten steel with the volume and type of lifting gas was calculated and analyzed in equilibrium state under

Theoretical and Experimental Research on the Mass Changes … Table 1 FactSage 7.1 database setting list

101

Option

Database setting

Calculation mode

Equilib

Database

FactPS, FToxide, FSstel

Solution module

FSstel-LIQU, FToxide-SLAGA

Compound module

Gaseous phase: Ar, CO2 , CO, O2 Liquid steel phase: Fe, Al, O, C Slag phase: Al2 O3 , FeO

Table 2 FactSage 7.1 menu list

Reactants and final conditions

Data

Fe

100 g

C

0.08 g

Al

0.05 g

O

0.001 g

Ar/CO2

0–0.042816/0.04728 g

Final conditions

Bubble pressure 67 Pa, Temperature 1873 K

vacuum cyclic refining condition. Contrastive study was conducted on the effect of Ar and CO2 as lifting gas on RH refining. Thermodynamic calculations were performed with the FactSage software version 7.1. Equilibrium calculations were performed with the EQUILIB module and FACT database. The solution modules in the calculations are FSstel-LIQU, FToxideSLAGA, and Gaseous phase (Ar, CO2 , CO, and O2 ). A few parameters in molten steel were used to reproduce the evolution of elements mass in molten steel based on industrial sampling data for the process conditions [20, 21]. The mass of Ar and CO2 injected into molten steel was set to 0–0.042816 g and 0–0.04728 g, respectively, which equals that the injection flow rate was from 0 to 120 Nm3 /h. To simplify the calculation, the oxygen content in steel was set to 10 ppm during the calculation. And the specific settings during the calculation are shown in Tables 1 and 2.

Effect on [C] and [O] Content in Molten Steel When smelting non-ultra-low carbon steel, the molten steel of RH refining process comes from ladle furnace (LF) refining process, and therefore, initial oxygen content in molten steel is very low during RH treatment. Furthermore, the variation of various elements oxidized by free-oxygen in molten steel can be excluded. And after RH vacuum cycle degassing, the amount of oxidation of various elements in liquid steel is very small. It should be noticed that Si and Mn elements were not oxidized obviously

102

B. Han et al.

Fig. 1 Mass changes of [C] and [O] in molten steel

according to previous experimental results [19], so this section mainly focuses on the changes of [C] and [O] content in molten steel. Figure 1 shows the mass changes of [C] and [O] in molten steel with CO2 and Ar injection in molten steel, respectively. As shown in Fig. 1, there is almost no variation in the [C] content in molten steel due to the insert gas injection and ultra-low freeoxygen content. However, owing to the weak oxidizability of CO2 , a significant decrease of [C] content in molten steel occurs with CO2 injection. In addition, twice as much gas can be produced by the reaction of CO2 and [C] compared with the same amount of Ar injection, which can strengthen molten bath stirring and accelerate the removal of impurities [22]. It can also be observed from Fig. 1 that there is no change in [O] content in molten steel with CO2 or Ar injection, which indicates that CO2 injection will not cause the increasing of [O] content and reoxidation of molten steel under vacuum treatment. Hornby [23] believed that a CO2 molecule would decompose into a CO molecule and an oxygen atom, which could cause an increase in [O] content in molten steel. However, due to the short contact time, the equilibrium decomposition amount of CO2 is less than 1%, which means that the increasing amount of [O] content is significantly less than 2 × 106 . In addition, Bruce [24] and Gu [25] introduced CO2 gas into LF refining process and found that CO2 gas does not affect the quality of the molten steel.

Effect on Composition of Off-Gas The metallurgical behavior of gas entering molten steel can be inferred by monitoring the composition of off-gas. In this calculation process, the removal of hydrogen and nitrogen was not considered because of the small amount in molten steel. Figure 2 shows the variation of the main composition in off-gas with CO2 and Ar injection, respectively. When Ar injection in molten steel, the main composition in off-gas is Ar according to the results in Fig. 2. In addition, a very small amount of CO gas

Theoretical and Experimental Research on the Mass Changes …

103

Fig. 2 Mass changes of gases in off-gas

exists in the off-gas, which is due to the decarbonization of free-oxygen in molten steel. When CO2 injection in molten steel, the main composition in off-gas is CO gas. This is because CO2 can react with some elements in molten and generate CO gas entering off-gas. And with the increase of CO2 injection rate, the more CO gas in off-gas is produced under equilibrium state. Furthermore, there is absolutely no possibility of Ar generation in the off-gas.

Effect on Carbon Content in Alloys During conventional RH refining process, several alloys should be added into molten steel to adjust the composition of molten steel. In this section, the trend of the change of carbon content in alloys with the CO2 ratio of reacting with [C] in molten steel was studied, which was under the condition of keeping the [C] content of molten steel unchanged. As shown in Fig. 3, carbon content in alloys increases with the higher CO2 ratio of reacting with [C]. Furthermore, the carbon content in alloys can be theoretically increased by more than 12% if all CO2 gas participates in the reaction CO2 (g) + [C] = 2CO(g). Based on the above analysis, the cost of alloys can be reduced by increasing the carbon content in alloys with CO2 as the RH lifting gas.

Industrial Tests Test Scheme In order to compare the refining effects of RH vacuum treatment with CO2 and Ar used as lifting gas more obviously, two grades of non-ultra-low carbon steel were

104

B. Han et al.

Fig. 3 Increase in carbon content of alloys

selected as test material. And the composition of molten steel before RH refining in ladle is shown in Table 3. As shown in Table 3, the main difference between steel A and B was the different initial [Al] content. Besides, the purity of CO2 and Ar used in test was 99.9%. In addition, the results before and after the RH refining indicate that no obvious changes in the silicon, manganese, nickel, chromium, and oxygen content in the steel occurred. It can be shown from Table 4 that the tests were conducted on the basis of not changing the original operation process as far as possible. Consequently, the gas flow rate, refining time, and vacuum degree were controlled to 100 Nm3 /h, 28 min, and 67 Pa, respectively. And the data of ten random heats for each scheme was chosen and analyzed in Sect. 4.

Results and Discussion Variation of [C] Content in Molten Steel CO2 injection is expected to be used for a small amount of decarburization during RH refining process. Accordingly, the change of carbon content is the most important index to be investigated. Figure 4 shows the average amount of decarburization of schemes 1, 2, and 3. It can be shown that the average amounts of decarburization of three schemes are 25, 42, and 94 ppm, respectively. The oxidation of [C] with the three schemes differs because of the different steel species, gas types, and gas flow rates applied. Based on a comparative analysis of schemes 1, 2, and 3, as indicated in Tables 3 and 4, Fig. 4, CO2 injection caused 17 and 69 ppm more [C] oxidation with schemes 2 and 3 than Ar at the same gas flow rate during the refining process. As analyzed in

C

0.1300

0.1310

Component

A

B

0.0210

0.0415

Al 0.0050

0.0041

O

Table 3 Steel composition of ladle pulling in RH (mass%) Si 0.2436

0.2230

Mn 1.2955

1.4054

P 0.0147

0.0130

S 0.0053

0.0039

Ni 0.0186

0.0280

Cr 0.0310

0.0400

Theoretical and Experimental Research on the Mass Changes … 105

106

B. Han et al.

Table 4 Gas control strategy and test schemes Scheme

Steel

Dragging gas

Gas flow rate (Nm3 /h)

Refining time (min)

Vacuum degree (Pa)

Heats

1

A

Ar

100

20

67

10

2

A

CO2

100

20

67

10

3

B

CO2

100

20

67

10

Fig. 4 Oxidation amount of [C] in RH refining

Sect. 2.1, this is because CO2 can react with [C] in molten steel. However, the amount of decarburization is obviously different between schemes 2 and 3, which can be explained by the following reason: The initial [Al] content in steel A is higher than that in steel B, and [Al] is oxidized much faster than [C] by CO2 with scheme 2. Most of CO2 gas reacts with [Al] in molten steel, which causes less oxidation of [C]. However, the results of scheme 3 are opposite to scheme 2. Besides, the decarburization rate is affected by the total pressure of the vacuum chamber which can be explained by the change in interfacial area of the reaction between the gas and liquid phases [26]. With scheme 3, more CO2 molecules react with [C] in molten steel and generate more CO molecules, which resulting in the increase in interfacial area of the reaction between the CO2 gas and liquid phases.

Variation of [Al] Content in Molten Steel In order to ensure a low oxygen content in molten steel during RH refining, a certain acid-soluble aluminum content should be maintained in the molten steel. CO2 can oxidize [Al] so easily that this section mainly investigates the oxidation of [Al] in molten steel by CO2 injection. Figure 5 shows the average amount of aluminum loss of industrial tests with the three schemes applied. As shown in Fig. 5, the average aluminum loss of scheme 1 through 3 is 42, 172, and 49 ppm, respectively. And

Theoretical and Experimental Research on the Mass Changes …

107

Fig. 5 Average aluminum loss in RH refining

according to the comparative analysis using schemes 1 and 2, CO2 injection causes 130 ppm more aluminum loss than Ar at the same gas flow rate during the refining process of steel A. The aluminum loss of scheme 3 is 49 ppm, which is 122 ppm less than that of scheme 2, which is due to the initial [Al] content in steel B being 0.021%. In previous study [19], the influence of initial [Al] content on amount of oxidation of [C] and [Al] has been analyzed in detail. Thus, the initial [Al] content and gas type are the major factors affecting the loss of aluminum. Besides, the difference of aluminum loss under different schemes is very obvious, based on which a large amount of aluminum loss can be avoided by reduce the initial [Al] content with CO2 injection.

Variation of [O] Content in Molten Steel As analyzed in Sect. 2.1, dissolved oxygen is very sensitive to molten steel during RH refining process. To verify the calculation results, the change of dissolved oxygen content with schemes applied has been detected, and the test results are shown in Fig. 6. Comparing the change of oxygen content before and after refining, there is no obvious increase in [O] content with scheme 1 applied, which is because no oxidant was added in ladle during RH refining. However, CO2 injection causes 3 and 1 ppm increase in [O] content with schemes 2 and 3 applied. Because aluminum is the main deoxidizer in the tests, the change of [Al] content in molten steel is thus an important factor affecting the [O] content. Combining the results in Sect. 4.2, CO2 gas injection into molten steel causes more aluminum loss. Therefore, owing to the reaction between [Al] and CO2 , the change of dissolved oxygen of the three schemes differs. Nevertheless, compared with scheme 1, the amount of increase in [O] content of schemes 2 and 3 is so small. Above all, it can be concluded that CO2 injection

108

B. Han et al.

Fig. 6 Change of dissolved oxygen content before and after RH refining

will not cause a significant increase of [O] content and reoxidation of molten steel under vacuum treatment.

Conclusion In the present study, FactSage software was used to study the mass changes of elements in molten steel with Ar and CO2 as RH lifting gas, respectively. The results of industrial tests carried out at a 150 ton RH refining device were shown for discussing the effect of practical application of CO2 as lifting gas. Synthesizing the theoretical calculation and industrial trials, the following conclusions can be drawn. (1) Compared to Ar gas, CO2 injected as RH lifting gas can be used for a small amount of decarburization without a significant increase in oxygen content of molten steel. (2) Keeping the carbon content of molten steel unchanged, the carbon content of alloys can be increased, which can be theoretically increased by more than 12% if all CO2 gas participates in the reaction CO2 (g) + [C] = 2CO(g). (3) It may be due to equipment or process operation that nitrogen removal during RH refining process is still difficult in industrial tests even with CO2 injection as lifting gas.

References 1. Li YH, Bao YP, Wang M, Wang R, Tang DC (2015) Influence of process conditions during Ruhrstahl-Hereaeus refining process and effect of vacuum degassing on carbon removal to ultra-low levels. Ironmaking Steelmaking 42:366–372

Theoretical and Experimental Research on the Mass Changes …

109

2. Demaglie GR, Tangari P, Fera S, Colla V (2013) Improving manufacturing of ULC steel grades by revamping of RH degasser in steelmaking shop No. 2 of ILVA, Taranto Works. Ironmaking Steelmaking 37:257–261 3. Feng K, Wang HB, Xu AJ, He DF (2013) Endpoint temperature prediction of molten steel in RH using improved case-based reasoning. Int J Miner Metall Mater 20:1148–1154 4. Geng DQ, Lei H, He JC (2012) Simulation on flow field and mixing phenomenon in RH degasser with ladle bottom blowing. Ironmaking Steelmaking 39:431–438 5. Park YG, Yi KW, Ahn SB (2001) The effect of operating parameters and dimensions of the RH system on melt circulation using numerical calculations. ISIJ Int 41:403–409 6. Yang GW, Wang XH, Huang FX, Wang WJ, Yin YQ (2014) Transient inclusion evolution during RH degassing. Steel Res Int 85:26–34 7. Zhang J, Liu L, Zhao X, Lei S, Dong Q (2014) Mathematical model for decarburization process in RH refining process. ISIJ Int 54:1560–1569 8. da Silva CA, da Silva IA, de Castro Martins EM, Seshadri V, Perim CA, Vargas Filho GA (2004) Fluid flow and mixing characteristics in RH degasser of Companhia Siderurgica de Tubarao, and influence of bottom gas injection and nozzle blockage through physical modelling study. Ironmaking Steelmaking 31:37–42 9. Zhang LF, Li F (2014) Investigation on the fluid flow and mixing phenomena in a RuhrstahlHeraeus (RH) steel degasser using physical modeling. JOM 66:1227–1240 10. Ajmani SK, Dash SK, Chandra S, Bhanu C (2004) Mixing evaluation in the RH process using mathematical modeling. ISIJ Int 44:82–90 11. Korneev VM, Ovsyannikov VG, Burmistrova EV, Frolov VI, Samoilin SA (2005) Improving the steel degassing technology by use of large-diameter snorkels. Refract Ind Ceram 46:153–156 12. Kuwabara T, Umezawa K, Mori K, Watanabe H (1988) Investigation of decarburization behavior in RH-reactor and its operation improvement. Trans ISIJ 28:305–313 13. Ling H, Zhang LF, Liu C (2018) Effect of snorkel shape on the fluid flow during RH degassing process: mathematical modelling. Ironmaking Steelmaking 45:1–12 14. Jiang F, Cheng GG (2012) Effects of gas injection with multi-hole orifices in up-leg snorkel on bubble behaviour and decarburisation rate during RH refining. Ironmaking Steelmaking 39:386–390 15. Kishan PA, Dash SK (2009) Prediction of circulation flow rate in the RH degasser using discrete phase particle modeling. ISIJ Int 49:495–504 16. Obata F, Waka R, Uehara K, Ito K, Kawata Y (2000) Circulation characteristics of RH degassing vessel water model with multi-legs. Tetsu-to-Hagane 86:225–230 17. Chen GJ, He SP, Li YG, Guo YT, Wang Q (2016) Investigation of gas and liquid multiphase flow in the Rheinsahl-Heraeus (RH) reactor by using the Euler-Euler approach. JOM 68:1–11 18. Chen GJ, He SP (2016) Mixing behavior in the RH degasser with bottom gas injection. Vacuum 130:48–55 19. Han BC, Zhu R, Zhu YQ, Liu RZ, Wu WH, Li Q, Wei GS (2018) Research on selective oxidation of carbon and aluminum with introduction of CO2 in RH refining of low-carbon steel process. Metall Mater Trans B 49:3544–3551 20. Cho MK, Van Ende MA, Eun TH, Jung IH (2012) Investigation of slag-refractory interactions for the Ruhrstahl-Heraeus (RH) vacuum degassing process in steelmaking. J Eur Ceram Soc 32:1503–1517 21. Bale CW, Bélisle E, Chartrand P, Decterov SA, Eriksson G, Gheribi AE, Hack K, Jung IH, Kang YB, Melancon J, Pelton AD, Petersen S, Robelin C, Sangster J, Spencer P, Van Ende MA (2016) FactSage thermochemical software and databases, 2010–2016. Calphad-Comput Coupling Phase Diagrams Thermochem 54:35–53 22. Wei GS, Zhu R, Wu XT, Dong K, Yang LZ, Liu RZ (2018) Technological innovations of carbon dioxide injection in EAF-LF steelmaking. JOM 70:969–976 23. Hornby S, Doulas L, Bermel L (1990) Use of CO2 in the AOD. In: Electric furnace conference proceedings 24. Bruce T, Weisang F, Allibert M (1987) Effects of CO2 stirring in a ladle. In: Electric furnace conference proceeding, Chicago

110

B. Han et al.

25. Gu YL, Wang HJ, Zhu R, Wang J, Lv M, Wang H (2014) Study on experiment and mechanism of bottom blowing CO2 during the LF refining process. Steel Res Int 85:589–598 26. Kishimoto Y, Yamaguchi K, Sakuraya T, Fujii T (1993) Decarburization reaction in ultra-low carbon iron melt under reduced pressure. ISIJ Int 33:391–399

Hydrogen as a Fuel and Ramifications Ashok Khandkar and Neale R. Neelameggham

Abstract There are expectations that hydrogen when produced economically will act as a clean fuel. This paper analyzes the ramifications on this perception. Thermochemical computations are made to show what may happen during combustion of hydrogen and associated reactions in this analysis. Keywords Hydrogen fuel · Thermal emissions · Radiative forcing · Thermal NOx · Climate change

Introduction About 400 years ago, the world witnessed a transformative event—conversion of coal to mechanical power and energy from a hydrocarbon source, coal, via the steam engine. Since then, we in the USA particularly find ourselves at an interesting crossroad—we have about 5% of the world’s population, and we consume 20% of the world’s oil. With the advances and bounty from fracking, a technology which we have developed and embraced, we have become an exporter versus an importer of oil. This of course seems to have alleviated one major problem, that of energy security, as long as the bounty lasts. But what of that source runs out? Or if there are unforeseen environmental issues that make this unacceptable. What are our alternatives?

Hydrogen Economy Expectation Some might argue for a hydrogen economy. This is based on three expectations: (1) that hydrogen can be produced from domestic energy sources in an economically, A. Khandkar University of Utah, Salt Lake City, USA N. R. Neelameggham (B) IND LLC, South Jordan, USA e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_11

111

112

A. Khandkar and N. R. Neelameggham

so that its applications are affordable, (2) production of hydrogen is environmentally benign, and (3) that various applications that employ hydrogen as a fuel, such as fuel cell vehicles, can gain market share in competition with the incumbent and other emerging alternatives. If these expectations can be met, the USA, and the rest of the world as well, could benefit from reduced vulnerability attribute to lack of energy security, improved environment, especially due to lower or no carbon emissions.

Challenges But is hydrogen really an environmentally benign fuel? For this, we restrict our comments only to the transportation sector, since this consumes the bulk of the hydrocarbon sources of energy in the USA. The challenges to wide-scale adoption of a hydrogen economy are many. To begin with, let us consider the “engine”, i.e. the conversion of hydrogen into a form that enables mechanical power to be harnessed for transportation. Despite substantial R&D expenditures, fuel cell-based power plants for automotive applications are still a factor of 10 times too expensive, are not robust enough, and have an efficiency that is still too low for light-duty-vehicle applications. Hence, in the absence of some government incentives, economic viability remains distant. Another challenge is hydrogen storage: Unlike gasoline or diesel, high-pressure compressed hydrogen or cryogenic liquid hydrogen storage have shortcomings related to safety and power density on a volumetric basis that impede long-term commercial viability. Technological breakthroughs in storage solutions that provide a vehicle with a 200–300 mile driving range, that are compact, lightweight, and inexpensive; and that comply with safety standards are sorely needed.

Emissions—Water Vapor Some hydrogen-on-demand technologies—such as Powerball—where plastic encapsulated sodium is cut in water to make pressurized hydrogen for fuel cells. But all such hydrogen-on-demand and storage technologies, other than electrolysis of water using photovoltaic generated electricity, inherently consume more energy than that delivered by the oxidation of hydrogen. In other words, processes for converting water or hydrocarbon fuels to hydrogen will always generate thermal emissions and will not provide a solution to control warming of atmospheric air mass. In addition, the radiative forcing effect from another triatomic molecule—water vapor will continue to add to the anthropogenic thermal effects proportional to population—even though it is clean in the sense of not having particulate emissions. Water vapor can condense by loosing its heat to outer space—into mist droplets and ice in clouds—and come down to earth as precipitation. Increased energy conversion

Hydrogen as a Fuel and Ramifications

113

using hydrogen as a fuel will continue the climate change by changes and increases in precipitation pattern worldwide in a lifeless fashion. For life molecules to form the essential life, forming elements are essential. If hydrogen economy becomes the way of life, we will be having a different popular outcry at that time regarding this different climate change. Another important issue is that of a nationwide, ubiquitous, and cost-effective hydrogen filling infrastructure for the hydrogen economy to be real nationwide, although local infrastructures are developing. The production, delivery, and dispensing of hydrogen are very logistically, technically. and economically complex, as it depends on many variables associated with available production facilities and cooperation between private industry and public entities. Successful transition to a hydrogen economy will not only require addressing the aforementioned challenges, but also a concerted effort in developing policy and enacting rules and regulations alongside technical breakthroughs to address the limitations outlined above. Assuming that ingenuity can be marshalled to tackle the technical problems, and lower capital costs, allows safe and effective hydrogen production, storage, the basic thermodynamic question of whether hydrogen is a eco-friendly fuel still remains. It is clear that wide-scale substitution of hydrogen for hydrocarbon fuels will lead to major reductions in CO2 emissions. It could also lead to enhanced energy security. But what is the source of hydrogen? Unlike hydrocarbon sources like oil or natural gas, which are recycled over the long term by nature, from biomass sources and our demand for energy from burning fossil fuels, hydrogen is neither abundant nor readily “mined”. Thus, typically, hydrogen is produced by electrolysis of water or from hydrocarbon sources. For the former, when considering the “well to wheel” efficiency and generation of greenhouse gases, when one accounts for the power required for electrolysis, whether derived from renewable energy, nuclear energy or surplus energy (e.g. energy from a conventional power plant that is operating an optimal efficiency but generating power in excess of demand, say at night time), hydrogen production is not carbon neutral.

Emissions—Nitrogen Oxides Benefits of a central power generating plant to produce power that in turn produces hydrogen is that carbon capture and sequestration technologies may be more efficiently and economically deployed. This thus may enhance the probability of achieving reduction in CO2 and sulfur and nitrogen oxide emissions through scrubbing, treatment, and sequestration at the source where hydrogen is generated. The same is true for hydrogen generated from hydrocarbon fuel sources. A recent study, “Eco-stoichiometry of anthropogenic CO2 that returns to earth” discusses the thermochemistry of reactive nitrogen formation attributed to thermal combustion reactions [1]. The study shows that the nitrogen in air converts to nitric oxide at a minimum rate of about 3 mol for every 2100 mol of nitrogen in the air used

114

A. Khandkar and N. R. Neelameggham

for the oxidation of the hydrocarbon (assuming 10% excess air over that required for stoichiometric combustion) irrespective of the fuel. The fuel may be a hydrocarbon or even free hydrogen. This is due to thermally formation of nitric oxide, which takes place at combustion temperatures of about 2000 K. Increasing the excess air creates more thermal nitric oxide formation. The flame temperature of hydrogen combustion is higher than that for combustion of hydrocarbon fuels, making it more challenging to employ a strategy of using excess air to reduce combustion temperatures, in an effort to minimize thermal nitric oxide formation. What is important to recognize is that even using hydrogen as a fuel results in generating water vapor and undesirable side NOx products due to the nitrogen present in air as an oxidation agent.

Conclusions Hydrogen as a fuel will continue to be one of many technologies that will contribute to alternative energy technologies for a secure and environmentally friendly option. This will require a concerted research and development effort and improve our understanding of the underlying science to address the technical hurdles and working with policy and lawmakers to make it a reality.

Reference 1. Neelameggham NR (2018) Eco-stoichiometry of anthropogenic CO2 that returns to earth. https://www.smashwords.com/books/view/905256

Applying Biochar Composite Briquette for Energy Saving in Blast Furnace Ironmaking Kai Fan, Zi Yu and Huiqing Tang

Abstract In this research, carbon composite briquette (CCB) was prepared using ultrafine iron-oxide fine (Size = 2.0 µm), and biochar fines (Size = 45.0 µm) by cold briquetting followed by heat treatment. Anti-pulverization capacity and reduction kinetics of the prepared biochar composite briquette (BCB) under simulated blast furnace (BF) conditions were investigated. The coke saving effect of charging BCB was analyzed by numerical simulation. Results showed that the prepared BCB was with a chemical composition of 0.77 wt% metallic iron, 72.59 wt% magnetite, 11.25 wt% wustite, 4.66 wt% gangue, and 11.10 wt% carbon. BCB could keep a crushing strength after partial reaction of more than 1900 N/briquette under the simulated BF conditions. Model simulations indicated that for a BF with a productivity of 6250 ton hot metal (tHM)/day, a coke rate reduction of 20 kg/tHM could be realized by replacing 10% sinter with BCB and moreover, the status of the BF was negligibly influenced. Keywords Biochar · Carbon composite briquette · Blast furnace · Coke rate reduction

Introduction In the foreseeable future, the traditional manufacturing route (blast furnace (BF) ironmaking and basic oxygen furnace (BOF) steelmaking) would continue to be predominant for producing iron and steel all over the world [1]. Although this process is well established and highly efficient, it is facing challenges to lower energy consumption and to minimize CO2 emissions for a more efficient and more sustainable iron and steel industry [2, 3]. As BF ironmaking accounts for 75–80% of the total energy consumption and generating most CO2 emissions [4], the contribution of BF ironmaking to achieve the target is crucial. K. Fan · Z. Yu · H. Tang (B) State Key Laboratory of Advanced Metallurgy, University of Science and Technology Beijing, 30 Xueyuan Rd., Beijing 100083, China e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_12

115

116

K. Fan et al.

Utilization of biochars in iron and steel industry is expected to lower carbon and energy consumption as well as greenhouse gas (GHG) emissions because biochars are considered to be a renewable energy source. Biochar utilization in ironmaking and steel-making has been tried in several sectors including coking, sintering, BF fuel injection, EAF slag foaming, reduction reactions, and RHF direct reduction. Nowadays, carbon composite briquette (CCB) with high mechanical strength is considered to be a future raw material for BF ironmaking as it could significantly reduce the coke consumption in BF process. Coal is the main carbon-bearing material in preparing CCB for BF application. However, as using biochar has the fundamental meanings for the sustainable development of BF ironmaking. It is considered that using biochar to prepare composite briquette for BF application should be tried. In this study, high-strength BCB was prepared by cold briquetting followed by heat treatment. Anti-pulverization capability and reaction behavior of BCB under the simulated BF conditions were examined. Thereafter, the application of BCB in the blast furnace was evaluated.

Materials and Methods BCB Preparation Raw materials for preparing BCB were iron oxide, biochar, and quartz. The Fe2 O3 powders and SiO2 powders (analytical reagent grade) were purchased from Sinopharm Chemical Reagent Co. (Shanghai, China). The biochar was obtained from a local farm. The biochar was crushed, grounded, and was further carbonized under 1173 K. The proximate analysis of the prepared biochar particles is volatile matter: 3.91 wt%, fixed carbon: 88.23 wt%, ash: 4.91 wt% and moisture: 2.95 wt%. The mean diameters of hematite quartz and biochar particles were 2.02, 2.52, and 41.9 µm, respectively. Iron oxide, biomass char powders and silica powders were fully mixed with an addition of 10% distilled water and 2% organic binder (cellulose). The mass ratio of the iron oxide, the quartz, and the biochar in the mixture was iron oxide: quartz: biochar = 82:3:15. The moistened fines were pressed into agglomerates using a die under a pressure of 15 MPa. The agglomerates were air-dried followed roasting under 423 K. Then the agglomerates were subjected to heat treatment. The thermal route of the heat treatment was the following. The furnace was heated from room temperature to 1073 K at a rate of 5 K/min. After holding at 1073 K for 30 min, the furnace was cooled naturally.

Applying Biochar Composite Briquette for Energy Saving …

117

Fig. 1 Simulated BF conditions

Tests Under Simulated BF Conditions A tubular furnace was used. The diameter of the reaction tube is 45 mm. The tests were conducted under a simulated BF gas composition-temperature profile from the burden surface to the cohesive zone (CZ) (Fig. 1). In all individual runs, the furnace was heated from room temperature with BCB samples of approximately 25.0 g being loaded. After the predetermined time, BCB samples were withdrawn from the tube and quenched using an N2 stream. The total gas flow rate into the tube was 2000 cm3 /min (standard temperature and pressure (STP)).

Gasification Tests Gasification tests of BCB were conducted using a custom-built thermalgravimetric analysis device. The following is a brief introduction. The furnace was heated using super-canthal (MoSi2 ) elements, producing a 50-mm hot zone in the reaction tube (Diameter = 55 mm). The sample holder was made of a heat-resistance alloy (Fe– Cr–Al) wire. In the tests, the furnace was firstly preheated to the reaction temperature under N2 by introducing N2 at the inlet till the temperature became stable. After the sample (one single BCB) was preheated under 773 K for approximately 5 min in the upper part of the reaction tube, it was introduced into the constant temperature zone and N2 was switched to N2 –CO–CO2 mixture. When the sample was introduced into the hot zone, its mass loss was measured by the balance and recorded every two seconds by the logged computer. After the predetermined time, the N2 –CO–CO2 gas mixture was switched back to N2 and the briquette sample was withdrawn from the furnace and quenched using N2 . The total gas flow rate was 3000 ml/min (STP).

118

K. Fan et al.

Analysis and Characterization The selected samples were analyzed and characterized using the following techniques. Cold crushing strength was measured using an MJDW-10B electronic universal testing machine (MAIJIE Co., China). The loading speed of the jig was set at 2.0 mm/min. The load at which the briquette suddenly failed was recorded. An average value of three briquette tests was accepted. Morphology on the sample crosssection was observed using a Quanta-250 scanning electron microscope (SEM, FEI Co., US). Phase identification was performed using an M21X X-ray diffractometer (XRD, MAC Science Co., Japan).

Results and Discussion BCB Characteristics Figures 2 and 3 show the XRD pattern of the sample and the morphology on the crosssection of BCB, respectively. It could be observed that the main iron-bearing phases were magnetite. No single particle could be identified in Fig. 4. After heat treatment, the particles were sintered together. The binding force of BCB was originated from the sintered texture in the briquette. Chemical analysis showed that the prepared BCB had a chemical composition of 0.77 wt% metallic iron, 72.59 wt% magnetite, 11.25 wt% wustite, 4.66 wt% gangue, and 11.10 wt% carbon. Fig. 2 XRD pattern of BCB

Applying Biochar Composite Briquette for Energy Saving …

119

Fig. 3 SEM image on the cross-section of BCB

Fig. 4 Serial images of BCB after partial reaction

Anti-pulverization Capability of BCB Under Simulated BF Conditions BCB would experience mechanical, thermal, and chemical destruction if charged in BF. The anti-pulverization capacity of BCB under the simulated BF conditions was then investigated. Figure 4 shows the serial photos of BCB after partial reaction. It could be seen that the briquette kept a cylindrical shape and retained its compactness in the test. Figure 5 depicts the change of the briquette crushing strength after partial reaction. BCB kept a crushing strength after partial reaction of more than 1900 N/briquette through the Fig. 5 Change of BCB crushing strength after partial reaction with reaction time

120

K. Fan et al.

whole testing period and reached a maximum strength of 5500 N/briquette at 1473 K (180 min). Generally, the briquette kept a permittable crushing strength after partial reaction that is required for practical blast furnace iron-ore bearing burden in the whole period. The above analysis indicates the prepared BCB had a good anti-pulverization capability to avoid damage in the BF.

BCB Reaction Model in BF Isothermal reaction tests were conducted under the temperatures (T ) of 1073, 1173, 1273, and 1373 K with their, respectively, corresponding atmospheres as shown in Fig. 1. Figure 6 shows the results. Since the briquette mass loss was solely owing to the gasification of carbon and the removal of iron-oxide oxygen, the mass-loss fraction in Fig. 6 was defined as (m0 − mt )/(mC + mO ), where m0 is the initial mass of BCB, (g); mt is BCB mass at time t, (g); mC is the mass of carbon in the briquette, (g); and mO is the mass of iron-oxide oxygen, (g). A BCB reduction model was developed. Several assumptions were made in modeling and they are (1) shape of the briquette is assumed to be a sphere with a diameter (d) of 0.014 m; (2) the involved reactions are reactions (R1)–(R3); (3) the briquette reaction proceeds under a quasi-steady state; and (4) distribution of pressure, temperature, and composition of gas-phase inside the briquette is uniform.

Fig. 6 Mass-loss fraction curves under different BF conditions

Fe3 O4 + CO(g) = 3FeO + CO2 (g)

(R1)

FeO + CO(g) = Fe + CO2 (g)

(R2)

C + CO2 (g) = 2CO(g)

(R3)

Applying Biochar Composite Briquette for Energy Saving …

121

Based on the above assumptions, conservation of CO and CO2 are expressed as Eqs. 1 and 2, respectively.   Sbriquette PCO2 out − PCO2 in kCO2 /(8.314T ) + (r1 + r2 − r3 )Vbriquette = 0

(1)

Sbriquette (PCO out − PCO in )kCO /(8.314T ) + (2r3 −r1 +r2 )Vbriquette = 0

(2)

where, S briquette is the surface area of BCB, (m2 ); V briquette is the volume of BCB, (m3 ); PCO in and PCO2 in in are the partial pressures of CO and CO2 inside BCB, respectively, (Pa ); PCO out and PCO2 out are the partial pressure of CO and CO2 in the furnace atmosphere, respectively, (Pa ); k CO and K CO2 are the mass transfer coefficients of CO and CO2 on the briquette surface, respectively, (m/s); r i is the reaction rate of reaction i (i = R1, R2, and R3), (mol/(m3 s)). The rates of reactions (R1)–(R2) are Eqs. (3) and (4), and the rate of reaction (R3) is Eqs. (5) and (6). ri = Ai PCO in − Ai /K i PCO2 in

(3)

Ai = ags /(8.314T )/(K i /(ki (1 + K i ))(1 − f i )−2/3 )

(4)

r3 = B1 PCO2 ,in

(5)

B1 = k3 (1 − f 3 )2/3 (1/1.01 × 105 )ρCO /0.012

(6)

where ags is the specific area of iron-oxide particles, (m2 /m3 ); K i is chemical equilibrium constant of reaction i, (−); k i is the rate constant of reaction i, (m/s or l/s); and f i is reduction fraction or carbon conversion of reaction i, (−); ρCO is the initial carbon density of the briquette, (kg/m3 ). Data of K i and k i are given in Ref. [5]. k CO and kCO2 in Eqs. (1) and (2) are determined by Eqs. (7) and (8). kCO2 = DCO2 −N2 (2.0 + 0.6Re1/2 Sc1/3 )/d

(7)

kCO = DCO−N2 (2.0 + 0.6Re1/2 Sc1/3 )/d

(8)

where Re is Reynolds number, (−); and Sc is Schmidt number, (−); DCO2 −N2 and DCO−N2 are diffusivity coefficients of CO2 and CO in N2 , respectively, (m2 /s). From the above equations, PCO2 in and PCO in can be solved explicitly and they are Eqs. (9) and (10). PCO in = (b2 d1 − b1 d2 )/(a1 b2 − a2 b1 )

(9)

PCO2 in = (a1 d2 − a2 d1 )/(a1 b2 − a2 b1 )

(10)

122

K. Fan et al.

where a1 = −A1 − A2 , a2 = C1 + A1 + A2 , b1 = C2 + A1 /K 1 + A2 /K 2 , b2 = −(A1 /K 1 + A2 /K 2 + 2B1 ),C1 = 6kCO2 /(8.314T d), C2 = 6kCO /(8.314T d), d1 = C2 PCO2 ,out and d2 = C1 PCO,out . For a single BCB, the mass of each component at a given time is Eqs. (11)–(14). t m Fe3 O4 = m Fe3 O4 ,0 + Vbriquete

(−0.232r1 )dt

(11)

0.072(3r1 − r2 )dt

(12)

0

t m FeO = m FeO,0 + Vbriquete 0

t m Fe = m Fe,0 + Vbriquete

0.056r2 dt

(13)

−0.012r3 dt

(14)

0

t m C = m C,0 + Vbriquete 0

where m Fe3 O4 , mFeO , mFe , and mC are the mass of Fe3 O4 , FeO, Fe and C in the briquette, respectively, (kg); and the subscript, 0, denotes the initial value of the assigned variable. PCO2 ,out , PCO,out , Sc, and Re could be determined by the experimental conditions and are constants under the given experimental conditions. DCO2 −N2 and DCO−N2 are calculated using the formulas suggested in Ref. [5]. Owing to the sintering of the ironoxide particles, it is difficult to determine ags in Eq. (4). The try and error method was used and it was found that an ags of 800 m2 /m3 was adequate. Measured and modelpredicted mass-loss fraction curves under different conditions were compared, and the results are shown in Fig. 7. Figure 7 indicates that the agreement between model prediction and measurement is satisfying. Fig. 7 Model-predicted and measured mass-loss fraction curves under different BF conditions

Applying Biochar Composite Briquette for Energy Saving …

123

Evaluation of Applying BCB in BF The effect of BF energy saving by charging BCB was analyzed by numerical simulation. The simulations were carried out by integrating Eqs. (11)–(14) and the total BF model developed previously [6]. In the simulation, normal BF operation data in Ref. [6] was used. k CO and kCO2 were calculated using the gas flow rate given under the above experimental conditions as it was observed that further increasing the gas flow rate had a negligible influence on the mass-loss fraction curves in Fig. 7. Two cases (case A and case B) were simulated and compared. Case A was BF operation under normal conditions, and case B was with BCB charging. In case B, 10% of sinter was replaced with BCB. Figure 8 shows the behavior of BCB in BF. BCB iron oxide reaches a high reduction above the CZ (Fig. 8a), reflecting that the reducibility of BCB iron oxide is higher than that of sinter, and the overall gasification degree of BCB carbon is 81% (Fig. 8b), indicating that 19.0% of CCB carbon enters the BF lower part. The ungasified carbon could be consumed by the direct reduction of molten FeO and the carburization of BCB iron. Table 1 lists some BF indexes of the two cases. It is seen that replacing 10% sinter with BCB results in a drop of 3 K of BF top gas temperature, an increase of 0.3% of the BF top gas utilization, a BF productivity increase of 58 tHM/day, and a coke rate decrease of 19.7 kg/tHM. The change of BF productivity is owing to the different total iron content between BCB and the sinter. Table 2 lists the distribution of coke consumption in both cases. It is seen that by replacing 10% sinter with BCB, the coke consumption below the CZ was considerably reduced. Fig. 8 Reaction behavior of BCB in BF: a profile of reduction fraction of BCB iron oxide, and b profile of gasification rate of BCB carbon

124

K. Fan et al.

Table 1 Simulation results of some BF indexes BF index

Case A

Case B

Productivity (tHM/day)

6250

6308

Top gas temperature (K)

463

460

Top gas utilization (%)

51.3

51.6

Fuel rate (kg/THM)

Coke: 335.0, PC: 180.0, Carbon from BCB: 0

Coke: 314.6, PC: 178.3, BCB carbon: the gasified above CZ: 14.30, the ungasified above CZ: 3.46

Table 2 Distribution of coke consumption in BF

Item/(kg/tHM)

Case A

Case B

Combustion

175.0

173.4

54.3

51.6

Gasification above CZ Consumption below CZ

47.3

31.2

Carburization of molten iron

45.0

45.0

Other reactions

13.4

13.4

335.0

314.6

Total

Note: (1) The carbon for carburization of the briquette iron is from the briquette carbon; (2) other reactions represent the carbothermic reductions of non-ferrous oxides as SiO2 , MnO and so on, and the carbon consumption of these reactions is not influenced by charging BCB

Conclusions (1) BCB was prepared using cold briquetting followed by heat treatment. The main phases in the briquette were magnetite and wustite. The prepared BCB was with a chemical composition of 0.77 wt% metallic iron, 72.59 wt% magnetite, 11.25 wt% wustite, 4.66 wt% gangue, and 11.10 wt% carbon. The binding force was originated from the sintered texture inside the briquette. (2) Under the simulated BF conditions, BCB kept the cylindrical shape and retain its compactness; moreover, BCB kept a crushing strength after partial reaction of more than 1900 N/briquette and reach 4500 N/briquette at 1473 K (180 min). (3) Simulation using the briquette reaction kinetic and BF model showed that replacing 10% sinter with the BCB, coke rate of the BF could be decreased from 335 kg/tHM to 314.6 tHM. Acknowledgements The authors thank the National Natural Science Foundation of China for supporting this work (Project No. U1960205).

Applying Biochar Composite Briquette for Energy Saving …

125

References 1. Geerdes M, Chaigneau R, Kurunov I (2015) Modern blast furnace ironmaking: an introduction. Ios Press, Amsterdam 2. An R, Yu B, Li R, Wei YM (2018) Potential of energy savings and CO2 emission reduction in China’s iron and steel industry. Appl Energy 226:862–880 3. Tan X, Li H, Guo J, Gu B, Zeng Y (2019) Energy-saving and emission-reduction technology selection and CO2 emission reduction potential of China’s iron and steel industry under energy substitution policy. J Clean Prod 222:823–834 4. Zhou JC, Zhang CX, Li XP, Fan B (2010) Status of energy consumption and review and prospect of energy saving technologies of blast furnace procedure in China. J Iron Steel Res (China) 22(9):1–12 5. Tang HQ, Yun ZW, Fu XF, Du S (2018) Modeling and experimental study of ore-carbon briquette reduction under CO–CO2 atmosphere. Metals 8(4):205–215 6. Tang HQ, Rong T, Fan K (2019) Numerical investigation of applying high-carbon metallic briquette in blast furnace ironmaking. ISIJ Int 59(5):810–819

Properties and Microstructure of Copper and/or Nickel Supported on GO, rGO, and NGO Xiangyong Lv, Guangfen Liang, Yandong Li, Huamei Duan, Dengfu Chen and Mujun Long

Abstract The modified Hummers method is used to prepare graphite oxide (GO). Then reduced graphene oxide (rGO) and modified graphite oxide (NGO) are prepared by ascorbic acid reduction and ammonia solution modification, respectively. With GO, rGO, and NGO as the support, Cu or Ni mono-metal and CuNi bi-metals are support as M-GO, M-rGO, and M-NGO (M: Cu, Ni, and CuNi), respectively. FTIR, XPS, and TEM are used to analyze the properties and microstructure of the resulting composites. Up to 60% of Cu2+ can be reduced, and copper grains are evenly dispersed on the support. Cu2+ reduction can be promoted by support reduction, ammonia modification, and nickel addition. The reduction degree of copper is related to the size of copper crystal. Keywords Hummers method · Carbon support · CuNi bi-metals · Mono-metal

Introduction GO is a quasi-two-dimensional layered solid with covalent bonds in layers. And the layers are bonded by various oxygen-containing functional groups and weak hydrogen bonds. This special structure makes its layer spacing and specific surface area larger than graphite [1], and its ion exchange capacity stronger than graphite [2, 3]. Meanwhile, GO has both hydrophilicity and oleophilicity. These advantages give it good intercalation composite ability [4]. All kinds of intercalation composites prepared with GO as intercalation matrix have shown superior performance [5], and relevant researches have attracted extensive attention. Muzyka et al. [6] successfully X. Lv · G. Liang · H. Duan (B) · D. Chen · M. Long College of Materials Science and Engineering, Chongqing University, Chongqing 400044, China e-mail: [email protected] Chongqing Key Laboratory of Vanadium-Titanium Metallurgy and New Materials, Chongqing University, Chongqing 400044, China Y. Li College of Materials Science and Engineering, Yangtze Normal University, Chongqing 408100, China © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_13

127

128

X. Lv et al.

prepared GO and found that GO contains a large number of oxygen-containing functional groups, including hydroxyl group, carboxyl group, ether group, and epoxy group. Zhang and Liu [7] prepared a series of graphite oxide composites with highly dispersed mono-metal, bi-metal, and multi-metal nanoparticles supported on graphite oxide. It is found that metal nanoparticles are evenly dispersed on the support and the particles are not agglomerated. The obtained composite materials had high catalytic activity and stability. Sharma et al. [8] found that a large number of oxygen-containing functional groups introduced into GO could be anchors to complexation with metal ions, and multilayer GO as adsorbent could remove heavy metals in water. Dutta and Verma [9] found that using ammonia water to modify graphite oxide to reduce the COOH group of graphite oxide to CONH2 could improve the adsorption capacity of graphite oxide to metal ions. In this paper, GO products are successfully synthesized by Hummers and Offeman [10] method and modified with ammonia to obtain NGO. rGO is obtained by reducing GO with strong reducing agent. In addition, GO, rGO, and NGO are used as supports to load Cu and/or Ni metal particles to obtain composite materials. By means of XRD, FTIR, Raman, XPS, and TEM, the types of oxygen-containing functional groups on the GO layer and the distribution of metal particles on the support are investigated. By comparison, the influence of the different preparation methods on the property of the resulting composite materials is obtained.

Preparation of Supports Preparation of GO Graphite oxide (GO) is prepared by Hummers method. 460 ml concentrated sulfuric acid is added to a large beaker, and then 10 g graphite powder and 10 g NaNO3 are slowly added into concentrated sulfuric acid and stirred for 2 h in ice-water bath. Then, 60 g KMnO4 is slowly added into the beaker and the reaction is treated with ultrasound for 70 min. Then put the beaker into a water bath at 35 °C, add 1000 ml of distilled water to the mixture and stir for 1 h. Then transfer the beaker to a water bath at 98 °C and stir for 20 min. After the completion of the above steps, the beaker is transferred to a water bath at 35 °C, and an appropriate amount of 30% H2 O2 is added to the beaker until the mixture changed from dark brown to yellow color. After the color remained unchanged for a period of time, the temperature is maintained at 35 °C and 1000 ml of distilled water is added to the beaker, stirring for 1 h, and then ultrasonic treatment for 30 min. After ultrasonic treatment, the solution is filtered while hot to obtain filter cake. Then wash the filter cake with 10% HCl and filter it, and repeat these steps until there is no SO2− 4 in the filtrate (detected by BaCl2 ). Then use distilled water to wash, filter for many times. The washed filter cake is put into a vacuum drying box and dried at 40 °C for 12 h under vacuum. The prepared material is denoted as GO.

Properties and Microstructure of Copper …

129

Preparation of rGO 0.6 g GO is added to a beaker containing 100 ml of N-methyl pyrrolidone and ultrasonic dispersion is performed for 1 h to obtain evenly dispersed brown-yellow GO dispersion solution. Add 6 g vitamin C, stir thoroughly for 1 h at 65 °C. After filtering and drying, rGO is obtained.

Preparation of NGO Add 5 g GO and 100 ml ammonia hydroxide to a beaker. Ultrasonic stirring and maceration at 80 °C for 12 h. Filter while hot to obtain filter cake. The filter cake is dried under vacuum at 40 °C for 12 h and the material is labeled as NGO.

Preparation of Composite Materials 1 g of support (GO, rGO, or NGO) is placed in anhydrous ethanol solution. According to the 2:1 molar ratio of copper nitrate to nickel nitrate, the total mass of Cu ion and Ni ion is taken as 10% of the support mass of copper nitrate (0.260 g) and nickel nitrate (0.157 g), which are dissolved in 20 ml ethanol. And the mixture is added to a rotary evaporator. Ethanol is removed under vacuum. Then calcined in a tubular furnace under a protective atmosphere (5% H2 in He). Under the heating rate of 5 °C/min, it is roasted to 300 °C at room temperature, which is kept at 300 °C for two hours and then naturally cooled to room temperature. The supported composite materials loaded with CuNi bi-metals are obtained. The composites are denoted as CuNi-R (R = GO, rGO, NGO). The mono-metal is loaded with the similar method as above. The same amount of copper nitrate (or nickel nitrate) dissolves it in ethanol solution and followed by the rotary evaporation and calcination. The composite materials are denoted as Cu-R and Ni-R (R = GO, rGO, NGO).

Results and Discussion FTIR Analysis Figure 1 is the infrared spectrum of GO, rGO, and NGO. It can be seen from the figure that GO and NGO have the same functional groups at 1056 cm−1 , 1415 cm−1 and 1610 cm−1 , respectively, corresponding to the stretching vibration peak of C–OH, the bending vibration peak of O–H in hydroxyl [11], and the stretching vibration peak of

130

X. Lv et al.

Fig. 1 FTIR data of GO, rGO and NGO

C=C of benzenoid ring. Different from NGO, GO also has two functional groups at 972 and 3046 cm−1 , which are the out-of-plane bending vibration peak and stretching vibration peak of C–H in aromatic hydrocarbon, respectively. It can be seen that the surface properties of GO were significantly changed by ammonia modification. The peaks of rGO at 1400 cm−1 , 1636 cm−1 , 2360 cm−1 , and 3440 cm−1 are, respectively, the stretching vibration peak of C–O–C, the stretching vibration peak of O–H, the stretching vibration peak of O–H of phenol group and the bending vibration peak of –COOH.

XPS Analysis Figure 2a–c is XPS graph of Cu2p for CuNi-GO, CuNi-rGO, and CuNi-NGO. Figure 3a–c are the XPS graph of Cu2p for Cu-GO, Cu-rGO, and Cu-NGO. XRD results show that only Cu2+ in CuNi-NGO, Cu-rGO, and Cu-NGO can be reduced to Cu0 . In Fig. 2, the content of reduced copper in CuNi-GO, CuNi-rGO, and CuNiNGO are 60.7%, 49.9%, and 39.9%, respectively. Consider only the reduction of copper content, when CuNi bi-metal is loaded, the GO had the greatest impact on the reduction of Cu2+ to reduced copper, while the NGO had the least impact. In

Fig. 2 XPS analysis of Cu2p: a CuNi-GO, b CuNi-rGO, c CuNi-NGO

Properties and Microstructure of Copper …

131

Fig. 3 XPS analysis of Cu2p: a Cu-GO, b Cu-rGO, c Cu-NGO

Fig. 3, the content of reduced copper in Cu-GO, Cu-rGO, and Cu-NGO is 44.3%, 52.0%, and 49.5%, respectively. Therefore, when loading Cu mono-metal, rGO has the greatest influence on the reduction of Cu2+ to reduced copper, while GO has the least influence. When the supports were the same, the effect of GO loaded CuNi bi-metal on the reduction of Cu2+ to reduced copper is greater than that of loaded Cu mono-metal. However, rGO and NGO load Cu mono-metal have greater influence on the reduction of Cu2+ to reduced copper than load CuNi bi-metal. Figure 4a–c is the XPS graph of Ni2p for CuNi-GO, CuNi-rGO, and CuNi-NGO. Figure 5a–c are the XPS diagram of Ni2p for Ni-GO, Ni-rGO, and Ni-NGO. In

Fig. 4 XPS analysis of Ni2p: a CuNi-GO, b CuNi-rGO, c CuNi-NGO

Fig. 5 XPS analysis of Ni2p: a Ni-GO, b Ni-rGO, c Ni-NGO

132

X. Lv et al.

Ni2p spectra (Figs. 4 and 5), the peaks that appeared at 855.8 and 873.6 eV can be assigned to Ni2p2/3 and Ni2p1/2 spin–orbits of NiO, respectively [12, 13]. There is an associated peak at the distance of about 17.7 eV from 855.8 eV, which is the satellite peak of NiO [14]. And the nickel element in the composite material only exists in the form of Ni2+ .

Microstructure Analysis Figure 6a–c is the TEM image of CuNi-GO, CuNi-rGO, and CuNi-NGO. It can be seen from Fig. 6 that the crystals are evenly dispersed on the supports. Some crystals in the three composites are agglomerated and grew into large size particles. The phenomenon of CuNi-GO and CuNi-rGO are the most obvious. Figure 7a–d is TEM image of CuNi-GO and the EDS mappings of elements Cu, Ni, and O. It can be seen from Fig. 7a that the crystals are evenly distributed on the support and the crystal size ranges from 10–500 nm. It can be seen that a few crystals are larger than 100 nm, but most crystals are smaller than 100 nm. According to the EDS mapping in Fig. 7b–d, large crystals are mainly Cu and O elements, and Cu is aggregated. While Ni element is well distributed on the support. According to XPS the spectrum of Cu2p, the main components of the crystal are Cu2 O and/or CuO. It can also be seen from Fig. 7c that Ni element has high dispersion in GO and is hardly forms aggregation on GO support. Figure 7e–h is TEM image of CuNi-rGO and a EDS mappings of elements Cu, Ni, and O. In Fig. 7e, the crystals are uniformly distributed. Most crystals are less than 10 nm in size, and a few are more than 10 nm, which may be due to crystal aggregation. Figure 7f–h is an elemental sweep of CuNi-rGO, where the interaction between Cu and Ni results in the formation of large crystals. Other smaller crystals contain only nickel. According to Fig. 4(b), the crystals are NiO. Figure 7i–l is TEM image of CuNi-NGO and EDS mapping of elements Cu, Ni, and O. It can be seen from Fig. 7i that the crystal size is all less than 200 nm, and a small number of crystal size is more than 100 nm, and the crystal is evenly scattered

Fig. 6 TEM image of CuNi-GO (a), CuNi-rGO (b), and CuNi-NGO (c)

Properties and Microstructure of Copper …

133

Fig. 7 TEM images of CuNi-GO (a), CuNi-rGO (e), and CuNi-NGO (i), the EDS mappings of CuNi-GO (b-d), CuNi-rGO (f-h), and CuNi-NGO (j-l)

on the carrier, most of the crystal is thin and a small number is granular. The surface scanning photos of the elements show that the granular crystals are mainly composed of Cu and O elements and contain a small amount of Ni elements, while the flake crystals only contain Cu elements. According to Figs. 2c and 4c, granular crystals are composed of copper crystal (Cu0 , Cu2 O, and/or CuO) and nickel crystal (NiO), while flake crystals are composed of Cu0 , Cu2 O and/or CuO. This property is consistent with CuNi-GO and Ni elements are dispersed. Among the three composite materials, the largest crystal size is CuNi-GO, followed by CuNi-NGO, and the smallest is CuNi-rGO. Figure 8a–i shows TEM images of Cu-GO, Cu-rGO, and Cu-NGO, as well as faceswept photos of Cu and O elements. By comparing TEM images of Cu-GO, Cu-rGO, and Cu-NGO (Fig. 8a, d, g), it can be seen that the crystal size in Cu-GO is small, most crystal size ranges from 10 to 100 nm, and a few crystals are more than 100 nm. Cu-rGO is relatively large in size, with most crystals between 150 and 300 nm in size. The crystal size of Cu-NGO is between Cu-GO and Cu-NGO, with a size range of 50–150 nm. The Cu crystals on NGO support are hollow dots. Therefore, the order of crystal size of the three composite materials is: Cu-GO < Cu-NGO < Cu-rGO.

Conclusions In this experiment, graphite oxide (GO) is prepared, and the GO is reduced and modified to obtain rGO and NGO. CuNi bi-metallic, Cu mono-metal, and Ni mono-metal

134

X. Lv et al.

Fig. 8 TEM images of Cu-GO (a), Cu-rGO (d), and Cu-NGO (g), the EDS mappings of Cu-GO (b, c), Cu-rGO (e, f), and Cu-NGO (h, i)

were loaded with GO, rGO, and NGO as supports to obtain the loaded composite material. The properties of composite materials were analyzed by means of FTIR, XPS, and TEM, and the results were as follows: GO-based composite materials contain the most reduced copper. When loading Cu mono-metal, reduced copper content in Cu-rGO is the highest, while the content of reduced copper in Cu-GO is the lowest. When CuNi bi-metallic is loaded, the content of reduced copper in GO increased, the content of reduced copper in rGO and NGO decreased, and the content of reduced copper in CuNi-GO is the highest. Nickel only exists in the form of NiO compound in composite materials, and nickel is evenly distributed on the support, making it difficult to form crystals. The crystal sizes of CuNi-rGO are the smallest when CuNi bi-metallic is loaded. When loading Cu as a mono-metal, the crystal size in Cu-GO is the smallest. It can be seen that

Properties and Microstructure of Copper …

135

the crystal size in the composite material is related to the content of reduced copper. The higher content of reduced copper, the smaller the crystal size. Acknowledgements The authors gratefully acknowledge the financial support by the National Natural Science Foundation of China. Project No. 51704048. Thanks to the financial supported by the Fundamental Research Funds for the Central Universities. Project No. 2019CDXYCL0031. The project name: Cutting-edge Technological Innovation of New Materials and New Metallurgical Technologies. We would like to thank Analytical and Testing Center of Chongqing University for FTIR, XPS, and TEM analysis. We would like to thank Dr. Zhang Bin at Analytical and Testing Center of Chongqing University for their assistance with TEM analysis.

References 1. Geim AK, Novoselov KS (2007) The rise of graphene. Nat Mater 6(3):183–191 2. Ôçôôëâ Â, Úâôõß ÂÎ, Ëð Ñ, Åóâ ÑÑ, Øñóñûñ ÇÐÇ, Çóïë Ö, Óâêæçîâõß Æ (2012) Physical properties of graphene (Scientific session of the Physical Sciences Division of the Russian Academy of Sciences, 28 March 2012). Physics–uspekhi 55(11):1223–1234 3. Wei BW, Qu D, Hu CF, Li FZ, Zhou TL, Xie RJ, Zhou ZM (2014) Synthesis and physical properties of graphene nanosheets reinforced copper composites. Adv Mater Res J 833:310–314 4. Bissessur R, Liu PKY, Scully SF (2006) Intercalation of polypyrrole into graphite oxide. Synth Met 156(16):1023–1027 5. Xiao P, Xiao M, Liu P, Gong K (2000) Direct synthesis of a polyaniline-intercalated graphite oxide nanocomposite. Carbon 38(4):626–628 6. Muzyka R, Kwoka M, Smedowski D, Díez N, Gryglewicz G (2017) Oxidation of graphite by different modified hummers methods. New Carbon Mater 32(1):15–20 7. Zhang H, Liu C (2013) Polyethylene/graphite oxide nanocomposites obtained by in situ polymerization using modified graphite oxide–supported metallocene catalysts. J Polym Res 20(1):39–46 8. Sharma RK, Sharma A, Sharma S, Dutta S (2018) An unprecedented ester-amide exchange reaction using highly versatile two-dimensional graphene oxide supported base metal nanocatalyst. Ind Eng Chem Res 57(10):3617–3627 9. Dutta R, Verma S (2015) A facile method of synthesizing ammonia modified graphene oxide for efficient removal of uranyl ions from aqueous medium. RSC Adv 94(5):77192–77203 10. Hummers WS Jr, Offeman RE (1958) Preparation of graphitic oxide. J Am Chem Soc 80(6):1339–1339 11. Szab T, Berkesi O, Forg P, Josepovits K, Sanakis Y, Petridis D, Dkny I (2006) Evolution of surface functional groups in a series of progressively oxidized graphite oxides. Chem Mater 18(11):2740–2749 12. Badrayyana S, Bhat DK, Shenoy S, Ullal Y, Hegde AC (2015) Novel Fe–Ni-graphene composite electrode for hydrogen production. Int J Hydrogen Energy 40(33):10453–10462 13. Gui J, Zhang J, Liu T, Peng Y, Jie C (2017) Two-step controllable preparation of NiO nanocrystals anchored reduced graphene oxide sheets and their electrochemical performance for supercapacitors. New J Chem 41(19):10695–10702 14. Eliche-Quesada D, Mérida-Robles J, Maireles-Torres P, Rodr´ıGuez-Castellón E, Busca G, Finocchio E, Jiménez-López A (2003) Effects of preparation method and sulfur poisoning on the hydrogenation and ring opening of tetralin on NiW/zirconium-doped mesoporous silica catalysts. J Catal 220(2):457–467

Investigation of H2 Addition Effects on CO/CO2 /H2 -Air Flames by a Combustion Diagnostic System Based on TDLAS Yu Liu, Jingsong Wang, Qingguo Xue, Haibin Zuo and Xuefeng She

Abstract In this study, a novel combustion diagnostic system based on tunable diode laser absorption spectroscopy (TDLAS) is developed to measure the temperature in flames on a turbulent partly premixed burner. This system simulates harsh industrial combustion environment and enables in situ measurements of non-uniform temperature in CO/CO2 /H2 -air flames. The effects of H2 addition (0, 4, and 8%, v/v) in CO/CO2 gas flow on multigas flames are investigated by this system. Two tunable continuous-waves near 1996 and 2004 nm are employed as the light sources of laser. The flame temperature distributions at several different locations along the axial direction at the burner centerline are reported. The results showed that the thickness of the flame decreases and the flame zone becomes a deeper blue color with the increasing addition of H2 in multigas flow. The length of the visible flame increases with the increasing addition of H2 . An increase in the addition of H2 in CO/CO2 /H2 gas leads to an increase of flame temperature. Keywords Combustion diagnostic system · TDLAS · H2 addition · CO/CO2 /H2 -air flames · Temperature

Introduction Converter gas and producer gas are important secondary energy resources for iron and steel enterprises and contain a large amount of carbon monoxide (CO), some carbon dioxide (CO2 ), and a few other gases such as nitrogen (N2 ), hydrogen (H2 ), and methane (CH4 ) [1, 2]. In China, converter gas and producer gas are normally used as fuel in hot-blast stoves, steel rolling heating furnaces, and power plants [3, 4]. In order to get a primarily understanding of the combustion behavior of these gases in harsh industrial environment, it is necessary to measure the temperature in flames. Y. Liu · J. Wang (B) · Q. Xue · H. Zuo · X. She State Key Laboratory of Advanced Metallurgy, University of Science and Technology Beijing, 30 Xueyuan Road, Haidian District, Beijing 100083, People’s Republic of China e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_14

137

138

Y. Liu et al.

Tunable diode laser absorption spectroscopy (TDLAS) is a novel spectroscopic monitoring technique and has been proven to be useful for temperature measurements. Zhou Xin et al. compared two tunable diode laser temperature sensors for nonintrusive determination of gas temperature and temperature fluctuations [5]. Kamimoto et al. discussed high temperature field application of two dimensional temperature measurement technology using CT-TDLAS[6]. Zhang et al. investigated a novel technique for characterizing temperature non-uniformity based on measurements of line-of-sight TDLAS[7]. Ma et al. reported in situ measurements of non-uniform temperature, H2 O and CO2 concentration distributions in a premixed methane–air laminar flame using TDLAS[8]. TDLAS has been widely used for temperature measurement in harsh ambient conditions due to its high temperature sensitivity. Additions (H2 , N2 , and CH4 ) in converter gas and producer gas cannot be avoided. The effects of CO, H2 O, and H2 addition on gas flames have been investigated. Jeong Park et al. studied addition effects of H2 and H2 O on flame structure and NOx emission behavior in methane–air counterflow diffusion flames [9]. S. S. Shy et al. investigated effects of H2 or CO2 addition, equivalence ratio, and turbulent straining on turbulent burning velocities for lean premixed methane combustion [10]. Babak Kashir et al. studied the effects of H2 addition in non-premixed turbulent combustion of C3 H8 – H2 –N2 mixture [11]. Jie Liu et al. examined effects of H2 and CO addition on the characteristics of methane laminar flame [12]. The effects of additions in multigases are mainly about the flame structure, temperature, and velocity, which characterize the thermodynamics and kinetics of flames. In this study, a novel combustion diagnostic system based on tunable diode laser absorption spectroscopy (TDLAS) was developed for measuring the temperature in flames. This system simulates harsh industrial combustion environment and enables in situ measurements of non-uniform temperature distributions in CO/CO2 /H2 -air flames. The effects of H2 addition in CO/CO2 /H2 gas flow on multigas flames were investigated by this system.

Experimental Combustion Diagnostic System Based on TDLAS A novel combustion diagnostic system based on TDLAS was designed and built for simultaneous measurements of the temperature and CO2 concentration in CO/CO2 /H2 -air flames on a turbulent partly premixed flame burner. The simplified schematic of the diagnostic system is shown in Fig. 1. This system has five parts: gas supply, gas transport, gas combustion, combustion chamber, and flame detection. Gas supply part consists of CO, H2 , and CO2 gas cylinders, pressure relief valves, an air compressor, mass flow controllers (MFCs), a mixing chamber, digital manometers, and an operation screen. Gas transport part

Investigation of H2 Addition Effects …

139

Fig. 1 Schematic of the combustion diagnostic system

comprises a ball valve, pressure stabilizing valves, a gas solenoid valve, digital manometers, and a butterfly valve. Gas combustion section includes a partly premixed burner and a fire controller. Combustion chamber part consists of a horizontal furnace and a window. Flame detection part is composed of laser diode controllers, DFB lasers, a splitter, laser probes, extension tubes, a detector, a computer, a displacement platform, an N2 gas cylinder, a pressure relief valve, and LZB gas flow meters. Figure 2 shows a partly premixed industrial burner. Figure 2a shows the structure and size of the burner. Figure 2b shows the size of the fuel and air outlets. The detection device is used for monitoring the flame. The combustion state in the burner

Fig. 2 Schematic of the structure of the partly premixed industrial burner. a Shows the structure and size of the burner. The burner consists of a flame detection device, an observation port, fuel and air inlets, a flange plate, and a burner sleeve. b Shows the size of the fuel and air outlets inside the burner sleeve

140

Y. Liu et al.

is observed using the observation port. The burner sleeve is inside the combustion chamber and has a length of 270 mm and a diameter of 40 mm. The center of the burner sleeve is fuel outlet and has a diameter of 9 mm. Air is the co-flow and partly premixed with the fuel. The burner is fixed on the outer wall of the combustion chamber using a flange plate. The fire controller controls the ignition switch. The calorific value range of the gas burned by the burner is in the range of 2000–3000 kcal/Nm3 . The designed gas flow ranges are 0–20 L/min, 0–30 L/min, and 0–100 L/min for H2 , CO2 , and CO, respectively. The supplied CO, H2 , and CO2 gases from Praxair have a purity of 99.9%, 99.999%, and 99.99%, respectively. The gas flow rates are controlled by mass flow controllers (NAURA Technology D07 series). The accuracy of MFC is ±1% S.P., and the response time is ≤0.5 s. The operating screen is utilized to control MFCs and display flow rates and pressure of fuel and airlines. The ball and butterfly valves are used as the switch of the two lines. The pressure stabilizing valve is used to stabilize the air pressure and limit the pressure of the two lines to a working pressure range 0–5 kPa. The gas solenoid valve is an automatic shutting-off device used to ensure the security. Digital manometers (Zhong De innovation ZD2008GN series) are set up to monitor the pressure. The accuracy of digital manometers is 0.25% F.S, and the range is 0–16 kPa. The horizontal furnace consists of a stainless steel casing and a double-layer high temperature refractory. The space size inside the furnace is 300 × 300 × 2000 mm3 , which is sufficient for the simulation of industrial furnace. The laser diode controllers (ILX 3724C) exactly control the current and temperature. Two DFB lasers (NEL KELD1G5BAAA) generate two tunable continuouswaves. The splitter splits the light sources into six laser probes corresponding to six measurement positions. Each position is 35 mm apart. The window with a length of 300 mm and a width of 140 mm is designed for the measurement and observation of flames. Two scales are placed on the window to calibrate the length of the flame and coordinates of the measurement. The axial and radial scales are in the ranges of 0–205 mm and −10 to 10 mm, respectively. Light is emitted by the laser probe and passes through the extension tube and flame. The detector receives the light and converts it into an electrical signal. Two BNC adapters (National Instruments BNC 2110) are used for signal input and output. The signal is transmitted to the data acquisition card in the computer and demodulated by laboratory view-based software. The extension tube is placed on the laser probe and connected to an N2 gas cylinder, a pressure relief valve, and LZB gas flow meters. The burner and extension tube inside the combustion chamber is shown in Fig. 3. The flow rate of N2 transmitted into the tube keeps 100 mL/min aiming to blow off the air inside the tube. The extension tube is adjusted to near the flame to reduce the effect of environment around the flame. The laser detection device is supported by a displacement platform, which is controlled to move along the axial and radial directions by UartAssist software with a minimum moving distance of 1 mm and a maximum moving distance of 1000 mm.

Investigation of H2 Addition Effects …

141

Fig. 3 The burner and extension tube inside the combustion chamber

TDLAS Theory and Parameters TDLAS is governed by Beer–Lambert law. When a laser beam passes through a gas medium with a length of L, the fractional transmission can be expressed as Eq. (1), I = exp[−P S(T )(υ)X L] I0

(1)

where I 0 and I are the incident and transmitted laser intensity, S(T ) is the line strength of the transition, which is only a function of temperature, P is the static pressure, X is the concentration of the absorbing gas, (υ) is the line-shape function. The integrated absorbance across the absorption feature can be obtained by Eq. (2),   +∞ I dν = P X S(T )L − ln A= I0

(2)

−∞

The temperature is obtained by comparing the line strength of two different transitions with different temperature dependence. The temperature-dependent line strength S(T ) can be expressed in terms of the known line strength at a reference temperature T 0 , which is defined by Eq. (3), S(T ) = S(T0 )

     hcE  1 Q(T0 ) T0 1 exp − − Q(T ) T k T T0

142

Y. Liu et al.





−hcv0 1 − exp kT





−hcv0 1 − exp kT

−1 (3)

where Q(T ) is the molecular partition function, h is Planck’s constant, c is the speed of light, k is Boltzmann’s constant, and E  is the lower-state energy. The temperature can be inferred from the measured ratio of the integrated absorbance for two different temperature-dependent transitions, and the ratio of these two integrals is given by Eq. (4),  Pabs Lv1 S1 (T )dv S1 (T ) A1  = = R= A2 S2 (T ) Pabs Lv2 S2 (T )dv      1 1 hc   S(T0 , v1 )  E1 − E2 − exp − = S(T0 , v2 ) k T T0

(4)

where Pabs is the partial pressure of the absorbing species, v is the line-shape function of a particular transition, S(T0 , vi ) is the line strength of the transition centered at vi , for the reference temperature T0 , and T is the temperature of the gas and can be obtained using Eq. (5), T =

hc k



E 2 − E 1

(T0 ) ln AA21 + ln SS21 (T + 0)



  hc ( E 2 −E 1 ) k T0

(5)

The quantity hc/k has a numerical value of 1.438 cm K. A1 and A2 are the integrated areas of the absorption lines. This study utilized two-line scanned-wavelength direct absorption spectroscopy to measure the temperature in CO/CO2 /H2 -air flames. Two tunable continuous-waves near 1996 and 2004 nm were employed as the light sources. A spectral simulation of the selected absorption lines using the HITRAN database was plotted, as shown in Fig. 4. The strong absorbance of CO2 and high temperature sensitivity were chosen to guarantee high signal-to-noise ratio (SNR) and minimal interference from ambient water vapor. The scanning rate of the laser is 106 Hz. The signal frequency is 100 Hz. The fluctuation of the results decreased with the increasing average number of the signalshots and finally kept a steady value in the range of 9000–10,000. 10,000 signal-shots were finally averaged to obtain accurate enough results.

Operating Conditions and Analysis of the Data Table 1 shows the measurement conditions of the experiments. Figures 5 and 6 show the representative measurements (Plot 0) of the absorption spectra of CO2 along with the corresponding Voigt-fitting profiles (Plot 1). Plot 2 shows the residuals of the two

Investigation of H2 Addition Effects …

143

Fig. 4 Simulation of the CO2 and H2 O absorption lines at different temperatures. The total pressure and path length were set at 1 atm and 45 mm, respectively Table 1 Combustion conditions of CO/CO2 /H2 -air flames Parameter

Multigas pressure

Air pressure

CO volume flow rate

CO2 volume flow rate

H2 volume flow rate

Air volume flow rate

Case

kPa

kPa

L/min

L/min

L/min

L/min

1

0.5

0.4

32.0

8.0

0.0

320.0

2

0.5

0.4

32.0

8.0

1.7

320.0

3

0.6

0.4

32.0

8.0

3.5

320.0

Fig. 5 Measured absorption spectra of CO2 near 1996 nm. The Voigt fit to the experimental data is plotted for comparison along with the residual

144

Y. Liu et al.

Fig. 6 Measured absorption spectra of CO2 near 2004 nm. The Voigt fit to the experimental data is plotted for comparison along with the residual

lines. The platform moved four times along the axial direction at a speed of 7 mm/s. 30 positions in total were selected for temperature measurement.

Results and Discussion Photos of the three flames were directly captured using a digital camera and shown in Fig. 7. The color and flame length of each flame were observed. All the flames seemed blue and resembled a cone. The flame of case 1 appeared bright blue, while flames of case 2 and case 3 appeared deep blue. It was observed that the thickness of the flame decreased and the flame zone became deeper blue color with the increasing addition of H2 in multigas flow. The increasing length of the visible flame with increasing addition of H2 is because the combustible component of the multigas is increased resulting in an increase in the flame temperature and reaction rate. The online TDLAS temperature measurements as a function of distance along the centerline axis were shown in Fig. 8. In the entire flame zone, the TDLAS temperature

Fig. 7 Photos of the three flames. a The flame of 0 L/min addition of H2 in CO/CO2 /H2 gas flow. b The flame of 1.7 L/min (4%) addition of H2 in CO/CO2 /H2 gas flow. c The flame of 3.5 L/min (8%) addition of H2 in CO/CO2 /H2 gas flow

Investigation of H2 Addition Effects …

145

Fig. 8 The online TDLAS temperature measurements of three flames. a Comparision between the effects of 0% H2 and 4% H2 addition on flame temperature. b Comparision between the effects of 4% H2 and 8% H2 addition on flame temperature

remained constant between 660 and 1300 K. The temperature uncertainty for all the points was ≤40 K. The maximum non-uniform temperature recorded in the flames for 0, 4%, and 8% (v/v) was 1260.8 K, 1268.3 K, and 1276.1 K, respectively. The whole temperature variation and maximum temperature variation indicate that an increase in the addition of H2 in CO/CO2 /H2 gas leads to an increase of flame temperature. This can be explained by the increasing amount of combustible component and calorific value. In addition, the temperature fluctuations become intense and flame temperature decreases rapidly on the tip of the flame when the temperature falls to 774 K. With the addition of H2 , the distance of this intense temperature fluctuation from the burner gets longer. It can be inferred that the temperature below 774 K belongs to unburned zone.

Conclusions In this study, a novel combustion diagnostic system based on tunable diode laser absorption spectroscopy (TDLAS) is developed to measure the temperature in flames on a turbulent partly premixed burner. This system simulates harsh industrial combustion environment and enables in situ measurements of non-uniform temperature in CO/CO2 /H2 -air flames. The effects of H2 addition (0, 4%, and 8%, v/v) in CO/CO2 gas flow on multigas flames are investigated by this system. Two tunable continuouswaves near 1996 and 2004 nm are employed as the light sources of laser. The flame temperature distributions at several different locations along the axial direction at the burner centerline are reported. The thickness of the flame decreases and the flame zone becomes deeper blue colour with the increasing addition of H2 in multigas flow. The length of the visible flame increases with the increasing addition of H2 . An increase in the addition of H2 in CO/CO2 /H2 gas leads to an increase of flame temperature.

146

Y. Liu et al.

Acknowledgements This work was supported by the State Key Laboratory of Advanced Metallurgy, University of Science and Technology Beijing.

References 1. Wang A-H, Jiu-ju X-P, Wang D, Zhou Q-A (2007) Affecting factors and improving measures for converter gas recovery. J Iron Steel Res Int 14:22–26 2. Dubovikov OA, Brichkin VN, Loginov DA (2016) Study of the possible use of producer gas of coal gasification as fuel. In: XVIII International coal preparation congress. Springer, pp 593–599. https://doi.org/10.1007/978-3-319-40943-6_91 3. Yikun W, Xiaomiao L, Lei D, Xiaoxu W, Changan W, Defu C (2014) A review on utilization of combustible waste gas(I):blast furnace gas, converter gas and coke oven gas. Thermal Power Gener 7(43):01–09. https://doi.org/10.3969/j.issn.1002-3364.2014.07.001 4. Zhi-Jun LI, Shao-Hua S (2010) Gas reheating furnace of low energy consumption technology. Mach Build Autom 39(1):163–164. https://doi.org/10.19344/j.cnki.issn1671-5276.2010. 01.048 5. Zhou X, Jeffries JB, Hanson RK, Li G, Gutmark E (2007) Wavelength-scanned tunable diode laser temperature measurements in a model gas turbine combustor. AIAA J 45(2):420–425 6. Kamimoto T, Deguchi Y, Kiyota Y (2015) High temperature field application of two dimensional temperature measurement technology using ct tunable diode laser absorption spectroscopy. Flow Meas Instrum 46:51–57 7. Zhang G, Liu J, Xu Z, He Y, Kan R (2016) Characterization of temperature non-uniformity over a premixed CH4 —air flame based on line-of-sight tdlas. Appl Phys B-Lasers O 122(1):3 8. Ma LH, Lau LY, Ren W (2017) Non-uniform temperature and species concentration measurements in a laminar flame using multi-band infrared absorption spectroscopy. Appl Phys B-Lasers O 123(3):83 9. Park J, Keel SI, Yun JH (2007) Addition effects of H2 and H2 O on flame structure and pollutant emissions in methane—air diffusion flame. Energy Fuel 21(6):3216–3224 10. Shy SS, Chen YC, Yang CH, Liu CC, Huang CM (2008) Effects of H2 or CO2 addition, equivalence ratio, and turbulent straining on turbulent burning velocities for lean premixed methane combustion. Combust Flame 153(4):510–524 11. Kashir B, Tabejamaat S (2013) A numerical study on the effects of H2 addition in non-premixed turbulent combustion of C3 H8 –H2 –N2 mixture using a steady flamelet approach. Int J Hydrogen Energy 38(23):9918–9927 12. Jie L, Xin Z, Tao W, Hou X, Zhang J, Zheng S (2015) Numerical study of the chemical, thermal and diffusion effects of H2 and co addition on the laminar flame speeds of methane–air mixture. Int J Hydrogen Energy 40(26):8475–8483

MnOx -Decorated Fe–Zr-Based Nano-Catalysts for Low-Temperature NH3 -SCR: Improvement of Catalytic Activity Chen Yang, Jian Yang, Qingrui Jiao, Yuanmeng Tian, Qingcai Liu, Shan Ren and Jiangling Li Abstract MnOx -modified Fe–Zr-based nano-catalysts (denoted as Mn(co)/Fe–Zr and Mn(im)/Fe–Zr) were synthesized with the co-precipitation impregnation method and impregnation method, respectively, and then were used for low-temperature selective catalytic reduction of NO with NH3 (SCR). Among these catalysts, the Mn(co)/Fe–Zr catalyst exhibits the highest NH3 -SCR activity (94%) at 225 °C when WHSV = 300,000 mL g−1 h−1 . Meanwhile, the properties of catalysts had been characterized by XRD, BET, XPS, and H2 -TPR. As determined by BET, the addition of MnOx increased the surface area and pore volume of the catalyst. The XRD results suggest that the Mn(co)/Fe–Zr possessed highly dispersed MnOx on the surface of the catalyst. The results of XPS revealed that the Mn(co)/Fe–Zr had highest concentrations of Mn4+ and Fe3+ . Keywords Mn/Fe–Zr · Low-temperature SCR · De–NO

Introduction In recent years, with the increasingly strict environmental protection policies in China, the controlling focus of nitrogen oxides on industrial flue gas has gradually shifted from thermal power plants to industrial furnaces. The selective catalytic reduction NOx with ammonia (NH3 -SCR) is the most promising approach for reducing NOx emission [1–3]. However, the commercial V2 O5 –WO3 (MoO3 )/TiO2 is almost active in the temperature window of 300–400 °C [4, 5]. Therefore, it is necessary to develop novel catalysts with high NOx conversion at low temperature. Iron-based catalysts have been widely studied due to its lower cost and nontoxicity [6, 7]. For example, the α-Fe2 O3 catalyst displayed good SCR activity and good thermal stability [8]. Liu et al. synthesized FeTiOx catalyst by using coprecipitation impregnation method, and the results showed there is the synergistic C. Yang · J. Yang (B) · Q. Jiao · Y. Tian · Q. Liu · S. Ren · J. Li College of Materials Science and Engineering, Chongqing University, Chongqing 400030, China e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_15

147

148

C. Yang et al.

reaction between Fe and Ti [9]. Meanwhile, in previous studies, the addition of zirconium oxide could enhance the thermal stability and improve the SCR activity on catalysts [10, 11]. In addition, it can interact with active components to promote the removal efficiency of NO [12]. Furthermore, on this basis, MnOx has been demonstrated as active components for increasing the SCR activity at low temperature [13, 14]. However, MnOx -decorated Fe–Zr-based catalyst has not been reported. Therefore, in this study, Mn(co)/Fe–Zr and Mn(im)/Fe–Zr catalysts are prepared by co-precipitation impregnation method and impregnation method, respectively, and then, the physicochemical characteristics of this catalyst were explored by XRD, BET, XPS, and H2 -TPR. Finally, the SCR performance of catalyst is evaluated in the temperature of 75–350 °C when WHSV = 300,000 mL g−1 h−1 .

Experiment Catalyst Preparation In this work, all chemicals were of analytical purity. Fe–Zr support was synthesized through the hydrothermal method. First, 0.09 mol Fe(NO3 )3 ·9H2 O and 0.01 mol Zr(NO3 )4 ·5H2 O were dissolved in 20 mL deionized water. Then, 2 mol/L NaOH (15 mL) solution was added into mixture and stirred for 30 min. After aging 72 h, the mixture was poured into a 50 mL Teflon-lined autoclave and then heated at 170 °C for 24 h. Finally, the product was washed with deionized water and ethanol two times, respectively. Finally, the Fe–Zr support was obtained after dried at 105 °C for 12 h. The Mn(im)/Fe–Zr catalyst with MnOx (10.0 wt%) loadings was prepared by impregnation method. During impregnation, 0.2 g Fe–Zr support was mixed with 20 mL deionized water and then ultrasonically treated at 60 °C until the liquid was eliminated, followed by drying at 105 °C for 12 h. The Mn(co)/Fe–Zr catalysts with MnOx (10.0 wt%) loadings were prepared by co-precipitation impregnation method. Typically, 0.2 g Fe–Zr support and a proper amount of Mn(NO)3 solution were mixed in 40 mL deionized water and stirred for 3 h. During the stirring process, adjusted the PH of mixture to 10 by adding a proper amount of NH3 ·H2 O. The resulting mixture was washed with deionized water and ethanol two times, respectively, followed by drying at 105 °C for 12 h. All catalysts were calcined at 400 °C for 3 h before the SCR reaction.

Characterization The X-ray diffraction (XRD) tests were recorded on a PANalytical X’Pert Powder diffractometer with Cu Kα radiation. The BET surface areas of the samples were

MnOx -Decorated Fe–Zr-Based Nano-Catalysts …

149

measured by N2 adsorption-desorption at −196 °C using a Quadrasorb 2MP apparatus. The X-ray photoelectron spectra (XPS) were performed on a Thermo Scientific ESCALAB 250Xi electron spectrometer. The base pressure was 5 × 10−10 mbar under testing conditions. H2 temperature-programmed reduction (H2 -TPR) was carried out on an AutoChem II 2920 auto-adsorption apparatus. During H2 -TPR, the samples were exposed to 10% H2 /Ar with a flow rate of 50 mL/min, and then, the temperature was raised to 900 °C at a rate of 10 °C/min. 100 mg of sample was used for each test.

Catalytic Activity Test The activity of the catalysts was tested in a fixed-bed quartz reactor (diameter: 8 mm) containing 100 mg of catalyst, and the weight hourly space velocity (WHSV) was 300,000 mL g−1 h−1 . The typical reactant gas was composed of 500 ppm NO, 500 ppm NH3 , 5 vol.% O2 , and balanced N2 , with the total gas flow rate 500 mL/min. A non-dispersive infrared flue gas analyzer (Thermo 60i) was used for detecting the concentrations of NO gas. NO conversion was calculated according to Eq. (1) NO convension(%) =

[NO]in − [NO]out × 100% [NO]in

(1)

Results and Discussion NH3 -SCR Activity The catalytic activity performance of the obtained catalysts is presented in Fig. 1. It was shown that the NO conversion of Fe–Zr support was less than 30% in the low Fig. 1 SCR performance of the obtained samples

150

C. Yang et al.

Fig. 2 XRD patterns of the obtained samples

temperature ( F, Adj R2 , C.V, Adeq Precision of purity model is 0.0290, 0.9642, 0.15%, and 21.503, respectively, which show the extremely high confidence level and precision. The polynomial equations derived from analysis for the factors on the responses are represented below: Yparticle size = 11.02 + 4.38A − 1.49B − 0.87C − 3.71D − 1.04E + 2.58H (2) Ypurity = 98.39 + 0.16A + 0.11B + 0.082C − 0.55D + 0.14E − 0.36F + 0.092G + 0.058H

(3)

Conclusions The statistical experimental design was used to investigate the influence of eight factors on performance of iron oxalate. We found that reaction temperature has a significant effect on both material’s particle size and purity. The influence of reaction factors could be attributed to the change of thermodynamics and kinetics which leads to different crystal nucleation and growth process. The model has a well fitted response and a good liner correlation with the data of Adj R-Squire, C.V, and Adeq Precision. The analytical results in this paper demonstrate that the preparation of

Controllable Synthesis of Battery-Grade Iron Oxalate with Waste …

257

battery-grade iron oxalate is a new way to utilize waste ferrous sulfate, which can offer an opportunity for safe industrial production. Acknowledgements This work was financially supported by the National Natural Science Foundation of China (Grant No. 51364021) and the Program for Innovative Research Team at the University of Ministry of Education of China (No. IRT_17R48).

References 1. Vondruska M, Bednarik V, Sild M (2001) Stabilization/solidification of waste ferrous sulphate from titanium dioxide production by fluidized bed combustion product. Waste Manag 21(1):11– 16. https://doi.org/10.1016/S0956-053X(00)00075-1 2. Huang P, Deng S, Zhang Z, Wang X, Chen X, Yang X, Yang L (2015) A sustainable process to utilize ferrous sulfate waste from titanium oxide industry by reductive decomposition reaction with pyrite. Thermochim Acta 620:18–27. https://doi.org/10.1016/j.tca.2015.10.004 3. Zhu XY, Xu GJ, Liu CH (2011) Upgrading of China’s Titanium Dioxide industry from the perspective of clean production. In: 2011 international conference on remote sensing, environment and transportation engineering, pp 8723–8726. https://doi.org/10.1109/rsete.2011.5964212 4. Kang B, Ceder G (2009) Battery materials for ultrafast charging and discharging. Nature 458(7235):190. https://doi.org/10.1038/nature07853 5. Xu B, Qian D, Wang Z, Meng YS (2012) Recent progress in cathode materials research for advanced lithium ion batteries. Mater Sci Eng: R: Rep 73(5–6):51–65. https://doi.org/10.1016/ j.mser.2012.05.003 6. Wang J, Yang J, Zhang Y, Li Y, Tang Y, Banis MN, Sun X (2013) Interaction of carbon coating on LiFePO4 : a local visualization study of the influence of impurity phases. Adv Func Mater 23(7):806–814. https://doi.org/10.1002/adfm.201201310 7. Wang J, Sun X (2012) Understanding and recent development of carbon coating on LiFePO4 cathode materials for lithium-ion batteries. Energy Environ Sci 5(1):5163–5185. https://doi. org/10.1039/C1EE01263K 8. Yang K, Deng Z, Suo J (2012) Synthesis and characterization of LiFePO4 and LiFePO4 /C cathode material from lithium carboxylic acid and Fe3+ . J Power Sources 201:274–279. https:// doi.org/10.1016/j.jpowsour.2011.11.019 9. Hu J, Hu X, Chen A, Zhao S (2014) Directly aqueous synthesis of well-dispersed superparamagnetic Fe3 O4 nanoparticles using ionic liquid-assisted co-precipitation method. J Alloy Compd 603:1–6. https://doi.org/10.1016/j.jallcom.2014.02.022 10. Wang P, Wang Z, Wu Z (2012) Insights into the effect of preparation variables on morphology and performance of polyacrylonitrile membranes using Plackett-Burman design experiments. Chem Eng J 193:50–58. https://doi.org/10.1016/j.cej.2012.04.017 11. Plackett RL, Burman JP (1946) The design of optimum multifactorial experiments. Biometrika 33(4):305–325. https://doi.org/10.2307/2332195 12. Sastry SV, Khan MA (1998) Aqueous based polymeric dispersion: Plackett–Burman design for screening of formulation variables of atenolol gastrointestinal therapeutic system. Pharm Acta Helv 73:105–112. https://doi.org/10.1016/s0031-6865(97)00052-6 13. Vatanara A, Najafabadi A R, Gilani K, Asgharian R, Darabi M, Rafiee-Tehrani M (2007) A Plackett–Burman design for screening of the operation variables in the formation of salbutamol sulphate particles by supercritical antisolvent. J Supercrit Fluids. 40(1):111–116. https://doi. org/10.1016/j.supflu.2006.03.028 14. Zhang K, Yang X, Wu J, Huang X, Yao Y (2016) Optimization of the process parameters for the synthesis process of battery-grade ferrous oxalate by response surface method. NANO 11(11):1650123. https://doi.org/10.1142/S179329201650123X

258

K. Zhang et al.

15. Zhang K, Liang F, Wang Y, Dai Y, Yao Y (2019) Multilayer iron oxalate with a mesoporous nanostructure as a high-performance anode material for lithium-ion batteries. J Alloy Compd 779:91–99. https://doi.org/10.1016/j.jallcom.2018.11.011 16. Guzun AS, Stroescu M, Jinga SI, Voicu G, Grumezescu AM, Holban AM (2014) PlackettBurman experimental design for bacterial cellulose–silica composites synthesis. Mater Sci Eng, C 42:280–288. https://doi.org/10.1016/j.msec.2014.05.031 17. Elsanhoty RM, Al-Turki IA, Ramadan MF (2012) Screening of medium components by Plackett-Burman design for carotenoid production using date (Phoenix dactylifera) wastes. Ind Crops Prod 36(1):313–320. https://doi.org/10.1016/j.indcrop.2011.10.013 18. Rahman Z, Zidan AS, Habib MJ, Khan MA (2010) Understanding the quality of protein loaded PLGA nanoparticles variability by Plackett-Burman design. Int J Pharm 389(1–2):186–194. https://doi.org/10.1016/j.ijpharm.2009.12.040 19. Bahloul L, Ismail F, Samar MEH, Meradi H (2014) Removal of AY99 from an aqueous solution using an emulsified liquid membrane. Application of Plackett-Burman design. Energy Procedia 50:1008–1016. https://doi.org/10.1016/j.egypro.2014.06.120 20. Veintemillas-Verdaguer S, Esteban SO, Herrero MA (2007) The effect of stirring on sodium chlorate crystallization under symmetry breaking conditions. J Cryst Growth 303(2):562–567. https://doi.org/10.1016/j.jcrysgro.2007.01.014 21. Yan FW, Zhang SF, Guo CY, Zhang XH, Chen GC, Yan F, Yuan GQ (2009) Influence of stirring speed on the crystallization of calcium carbonate. Crystal Res Technol: J Exp Ind Crystallogr 44(7):725–728. https://doi.org/10.1002/crat.200900190 22. He M, Addai-Mensah J, Beattie D (2009) The influence of polymeric dispersants on sericite– chalcocite particle interactions in aqueous media. Chem Eng J 152(2–3):471–479. https://doi. org/10.1016/j.cej.2009.05.010 23. Loginov M, Larue O, Lebovka N, Vorobiev E (2008) Fluidity of highly concentrated kaolin suspensions: influence of particle concentration and presence of dispersant. Colloids Surf A 325(1–2):64–71. https://doi.org/10.1016/j.colsurfa.2008.04.040

Mechanical Beneficiation of End-of-Life Lithium-Ion Battery Components Haruka Pinegar and York R. Smith

Abstract Recycling of end-of-life (EOL) lithium-ion batteries (LIBs) is imperative to resolve resource issues and manage waste streams of EOL LIBs. A mechanicalhydrometallurgical recycling route is preferable regarding material recovery and energy consumption, while cost reduction is challenging. Effective mechanical pretreatment of EOL LIBs to beneficiate LIB components is key to reduce operating cost. In this study, the primary mechanical treatment (shredding and sieving) of LIB packs and the subsequent attrition milling were performed to investigate the liberation and separation of LIB components. The results demonstrated that the separation of LIB components was significantly improved by attrition milling of shredded LIB pieces. Active electrode materials (lithium metal oxide and graphite), metal foils (copper and aluminum), and low-value components (cell/pack casing, separator, plastic, and paper) were enriched in 2 mm fractions, respectively. A combination of these treatments demonstrates to be a promising method for LIB component beneficiation. Keyword Lithium-ion battery · Recycling · Comminution · Beneficiation · Attrition milling

Introduction Since its commercial release in the early 1990s, lithium-ion batteries (LIBs) have become the most essential energy storage device for high-performance portable electronics attributing to their superior electrochemical performance. Over the last decade, the application of LIBs for electric vehicles (EVs) has become more commonplace with increasing demands on mitigation of anthropogenic global warming. It is expected that the largest LIB market will shift to EVs from portable electronics, which will significantly increase the overall production of LIBs [1]. On the other H. Pinegar · Y. R. Smith (B) Department of Materials Science and Engineering, University of Utah, 122 South Central Campus Drive #304, Salt Lake City, UT, USA e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_25

259

260

H. Pinegar and Y. R. Smith

hand, soaring LIB production and application will lead to a dramatic increase in consumption of raw materials and the waste stream of end-of-life (EOL) LIBs. The soaring LIB production increases the risk of resource depletion of valuable materials, especially of the electrode materials [2]. Generally, a fine powder of lithium metal oxide coated on aluminum current collector foil with support of polymer binder is used as a cathode while that of graphite coated on copper foil is used as an anode. These are separated by a porous plastic separator filled with an electrolyte composed of an organic solvent and lithium salts [3]. Materials such as lithium, cobalt, and graphite are concerned for limited reserves, resource concentration, geopolitical issues, and low battery-grade product yield [2, 4–7]. While recycling rates of transition metals are more than 50% (cobalt: 68%, nickel: 58%, manganese: 53%, and copper: 53%), recycling rates of lithium and graphite are nearly 0% [2]. Considering resource issues, it will be imperative for LIB consuming countries to establish LIB recycling system which enables independent close-loop LIB production. Recycling of LIBs is also beneficial for those countries as it eliminates the environmental issues caused by disposed LIBs in landfills (i.e., toxic chemical release and fire hazard [8, 9]) and reduces energy consumption and greenhouse gas emissions in LIB production [10, 11]. However, the present global LIB recycling capacity is barely sufficient to resolve resource and waste management issues in the future. The LIB recycling rate based on recycled portable/household electronics in the North America was estimated to be only 3% [12]. Presently, the main commercial LIB recycling processes are pyrometallurgy- or hydrometallurgy-based, or a combination of thereof. Considering the recovery of LIB materials, including lithium, graphite, and aluminum and lower energy consumption, the technical improvement and capacity increase in mechanicalhydrometallurgical recycling processes are more desirable. The major problem of this route is high operating cost for hydrometallurgical processes (i.e., leaching, solvent extraction and precipitation) to refine recovered LIB materials to battery-grade [13]. Effective mechanical pre-treatment of LIBs, such as shredding and sieving, to separate LIB component is the first critical step to simplify the subsequent refining treatments, leading to reduction of operating cost. Increasing the yield and concentration of liberated active electrode materials (i.e., lithium metal oxide and graphite) from current collector foils (i.e., aluminum and copper) before hydrometallurgical treatments is key to increase profitability of a recycling process as these electrode materials are most valuable. A prior study done by Diekmann et al. reported an active electrode materials yield of 75% by a two-step crushing/air classification of EV battery modules followed by sieving with 500 µm screen [14]. However, the number of studies done for mechanical treatment of LIBs is limited up to date despite its importance in development of LIB recycling technology. In this study, the effect of the mechanical and physical treatments of LIB packs was investigated on component separation, distribution of different size fractions, and liberation of active electrode materials. The highlight of this work is the novel application of attrition milling as a secondary treatment of shredded LIB packs.

Mechanical Beneficiation of End-of-Life …

261

Fig. 1 Pictures of a investigated LIB packs, b shredded LIB packs separated into different size fractions, c manually sorted components, and d layered composite chunks

Experimental Methods Shredding and Sieving of LIB Packs First, shredding and sieving of randomly selected LIB packs from portable electronics with three different cell types (prismatic, cylindrical, and pouch cell as shown in Fig. 1a) were performed as a primary mechanical treatment. LIB cells were removed from the packs and discharged completely by external electrical circuit and placed back to the packs. LIB packs were shredded in a laboratory-scale twin shaft shredder which was connected to a motor and gearbox at 28 RPM. After keeping in the fume hood overnight, the collected particles were screened by vibration sieving, which separated the shredded packs into six size fractions: >8 mm, 4.75–8 mm, 2–4.75 mm, 1–2 mm, 500 µm to 1 mm, and 8 mm, 4.75 mm to 8 mm, 2 mm–4.75 mm, 1.18 mm to 2 mm, 500 µm to 1.18 mm, 212 µm to 500 µm, and 1.18 mm), small fractions (212 µm to 1.18 mm), and fine fraction (500 µm indicates that significant amounts of these materials were remained attached on the current collector foils. The fraction of 1 mm). However, it is obvious from the graph that large fractions are composed of mixture of all LIB components, and the loss of active electrode materials into these fractions is significant. It was attributed to layered composite

264

H. Pinegar and Y. R. Smith

Fig. 3 Component and size distributions for LIB packs with a prismatic cells, b cylindrical cells, and c pouch cells

chunks contained in these fractions. These results demonstrated that additional treatment would be required to improve beneficiation of LIB components before refining treatments.

Mechanical Beneficiation of End-of-Life …

265

Attrition Milling Figure 4 shows the results of attrition milling at the solid–liquid ratio of 100 and 150 g/L. As shown in Fig. 4, attrition milling of shredded LIB packs demonstrated effective liberation and separation of LIB components. By attrition milling, more than 95% separation efficiency of layered composite chunks was achieved, which would solve component separation issue after the primary mechanical treatment. The

Fig. 4 Component distribution of shredded LIB pack after attrition milling and size distribution of shredded LIB pack before and after attrition milling at a solid–liquid ratio of a 100 g/L and b 150 g/L

266 Table 1 Main LIB component in the fractions of 2 mm after 1 h attrition milling of shredded LIB packs

H. Pinegar and Y. R. Smith Solid–liquid ratio (100 g/L)

Solid–liquid ratio (150 g/L)

81.59

79.05

68.33

65.54

96.25

96.69

>2 mm Low-value components (%) 500 µm to 2 mm Metal foils (%) 2 mm) decreased while the fractions of 2 mm. Table 1 shows the main component in the fractions of 2 mm after attrition milling. The results demonstrated that attrition milling of shredded LIB packs enriched active materials, metal foils, and low-value components into three size fractions.

Conclusion This study investigated the primary mechanical treatment of EOL LIB packs and the subsequent attrition milling as a secondary treatment. A significant portion of active electrode materials was observed to be unliberated after the primary shredding and sieving, attributing to layered composite chunks in the large fractions. Attrition milling of shredded LIB packs effectively separated and liberated LIB components and enriched active electrode materials, metal foils, and low-value materials into the fractions of 2 mm. A combination of shredding, sieving, and attrition milling for the LIB packs with pouch cells would increase the

Mechanical Beneficiation of End-of-Life …

267

yield of lithium metal oxide in 99.999%) gases were used in this experiment to carbonize the powder.

Experimental Procedure The 43% CaO-24% TiO2 -33% SiO2 ternary slag was designed for the research. The raw materials were weighed and mixed after roasting at 1173 K (900 °C) for 12 h. 50 g mixed powders were pelletized to a cylinder with a diameter of 35 mm under 10 MPa for 2 min and then put into a molybdenum crucible (45 mm in inner diameter and 50 mm in height). The sample was melted at 1873 K (1600 °C) and held for 2 h to homogenize, and then cooled to 1373 K (1100 °C) at a rate of 3 K min−1 . The whole experimental process was protected by argon gas (the purity >99.999%) in a high-temperature furnace, as shown in Fig. 1. Then, around 2 g slag powder was weighed and loaded into a corundum crucible (30 mm in inner diameter and 30 mm in height). The crucible with the sample was placed inside the vertical tube, and high-purity argon was introduced to flush air out

Crystallization and Carbonization of TiO2 –CaO–SiO2 Ternary Slag

337

Fig. 1 Schematic diagram of the experimental apparatus (1 furnace shell, 2 crucible, 3 beaker flask, 4 temperature controller, 5 computer, 6 gas flow controller)

of the furnace. The flow rate of gas was controlled by gas flow controllers (Alicat, Model MC-500SCCM-D and MC-1SLPM-D). The furnace was first heated from room temperature up to a desired reduction temperature at a rate of 5 K/min while maintaining the argon atmosphere. Once the thermal balance stabilized, argon gas was switched to reaction gas. After reacting for a certain time, the reaction gas was switched to argon again, and samples were cooled down to room temperature. The schematic diagram of the experimental apparatus and the temperature controlling program are shown in Fig. 1.

Characterization After the crystallization and carbonization experiments, the samples were crushed, polished, and ground for the subsequent detection. X-Ray diffraction (XRD) (PANalytical AERIS) measurements were conducted on the reacted samples to find the phases existing in the samples. The thermodynamics calculations were performed by using FactSage software.

Results and Discussion Crystallization The theoretical compositions of the TiO2 –CaO–SiO2 ternary system slag containing 24% TiO2 during cooling were analyzed by the Equilibrium module of FactSage 6.2, and the results are shown in Fig. 2. It can be found that three phases, perovskite (CaTiO3 ), sphene (CaTiSiO5 ), and wollastonite (CaSiO3 ), can be formed during the cooling process from 1873 to 1373 K (1600–1100 °C). With increasing the basicity

338

G. Fan et al.

Fig. 2 Theoretical phase compositions of 24% TiO2 –CaO–SiO2 ternary system slag during cooling

(expressed as R), the amount of perovskite and wollastonite increases, while that of sphene decreases. In the cooling process, with the basicity increasing, the crystallization temperature of perovskite increases from 1670 to 1811 K (1397–1538 °C), and that of wollastonite also increases, but the sphene crystallizes at a constant temperature of 1631 K (1358 °C), as shown in Fig. 3a. Figure 3b shows the crystallization mass percent of the solid phases as a function of basicity. It can be found that the mass percent of both perovskite and wollastonite phases increase, but that of sphene phase decreases with increasing the basicity. Namely, when the basicity increases, Ti element transfers into the perovskite phase from the sphene phase. The theoretical calculation results show that titanium can be enriched in perovskite phase by increasing the basicity of slag.

Fig. 3 a Crystallization temperature and b mass percent of crystallized phases as a function of basicity

Crystallization and Carbonization of TiO2 –CaO–SiO2 Ternary Slag

339

Carbonization In the crystallization process, the titanium-containing phases are perovskite and sphene, and the perovskite is the first crystallization and the main titanium-containing phase. Therefore, the carbonization of perovskite was analyzed by the FactSage software based on the database of Cao et al. [15]. The reaction was set in open system for 1000 steps, with the molar ratio among the CH4 , N2, and H2 keeping at a constant, and the results are shown in Figs. 4 and 5. In the 8 vol.% CH4 -92% vol.% H2 gas mixture system, it can be found that with the increase of CH4 , CaTiO3 transfers to Ca2 Ti2 O5 , Ca3 Ti2 O7 , Ca3 Ti2 O6 , and Ti(C,O). Ca2 Ti2 O5 can be thought of as a combination of two CaO and one Ti2 O3 , and Ca3 Ti2 O6 also can be thought of as a combination of three CaO and one Ti2 O3 . It is worth noting that Ca3 Ti2 O7 , Ca3 Ti2 O6 , and Ti(C,O) are generated simultaneously. With the reaction temperature increasing, the CH4 /CaTiO3 molar ratio point, that the three phases generated, is advanced from the 2.18 to 0.335. More importantly, the increase in temperature leads to an increase of the amount of Ti(C,O).

Fig. 4 Calculated equilibrium phases for reaction of perovskite with 8 vol.% CH4 -92 vol.% H2 gas mixture at different temperatures

340

G. Fan et al.

Fig. 5 Calculated equilibrium phases for reaction of perovskite with 8 vol.% CH4 -8 vol.% H2 -84 vol.% H2 gas mixture at different temperatures

In the 8 vol.% CH4 -8 vol.% N2 -84% vol.% H2 gas mixture system, with the reaction process going on, CaTiO3 transfers to Ca2 Ti2 O5 , Ca3 Ti2 O7 , Ca3 Ti2 O6 , TiN, and Ti(C,O). At 1100 and 1200 °C, there does not exist Ti(C,O) in the product phases. At 1300 °C, the Ti(C, O) exits as an intermediate product, and with the reaction progress, Ti(C,O) transfers to TiN. While the temperature increasing to 1400 °C, the final product is Ti(C,O). It shows that the increase in temperature is beneficial to the formation of Ti(C,O), namely final titanium-bearing product transfers from TiN to Ti(C,O), in the CH4 –N2 –H2 gas mixture system. In low temperature, compared with the CH4 –H2 gas mixture system, the addition of nitrogen is beneficial to the formation TiN, namely the amount of TiN in the final product is more than that of Ti(C,O) at the same temperature. The 43% CaO-24% TiO2 -33% SiO2 ternary slag was crystallized as the experimental procedure. Figure 6 shows the XRD result of the cooling slag, and it can be found that the phases in the slag are consistent with the above calculation result. The slag powder was reacted with the 8 vol.% CH4 -8 vol.% N2 -84% vol.% H2 gas mixture for 5 h at 1200 °C, and the XRD result was shown in Fig. 7a. It can be found that Ti(C,O) or TiN cannot be detected in the product, and the phases in the reacted

Crystallization and Carbonization of TiO2 –CaO–SiO2 Ternary Slag

341

Fig. 6 XRD pattern of slag after cooling

Fig. 7 XRD pattern of slag samples after reacted for 5 h with different additions: a No addition; b Fe2 O3 ; and c CH4 N2 O

342

G. Fan et al.

slag were consistent with the crystallized slag, indicating that the slag did not react with the gas mixture. The probable reason may be the poor gas permeability and reactivity of powder. In order to improve the reaction condition, CH4 N2 O and Fe2 O3 were introduced into the reaction. Fe2 O3 can catalyze the reaction, and CH4 N2 O can improve the gas permeability. Figure 7b and c shows the XRD result of samples with Fe2 O3 and CH4 N2 O, respectively. However, Ti(C,O) or TiN cannot be detected in the product, either. So, CH4 N2 O and Fe2 O3 were introduced into the reaction simultaneously. Figure 8 shows the XRD results of samples reacted for 5 h with the addition of Fe2 O3 and CH4 N2 O at different temperatures from 1200 to 1400 °C. It can be found that Ti(C, O) or TiN can be detected in all samples. Because TiC, TiO, and TiN can form a solid solution Ti(N,C,O), the peaks in the XRD patterns were labeled as Ti(N,C,O). The amount of TiN, TiC, and TiO can be quantified by other detection methods. With the temperature increasing from 1200 to 1400 °C, the content of

Fig. 8 XRD pattern of slag samples after reacted for 5 h with the addition of Fe2 O3 and CH4 N2 O at different temperatures: a 1200 °C; b 1300 °C; and c 1400 °C

Crystallization and Carbonization of TiO2 –CaO–SiO2 Ternary Slag

343

Ti(N,C,O) decreased, meaning that the increase in temperature was not conducive to the reaction. The probable reason may be that the powder was sintered to form large particles at high temperature, and it hider the flow of gas and the reaction between slag powder and gas. At 1400 °C, the slag melted, and the gas cannot pass through the material, so the reaction was blocked and very little Ti(N,C,O) formed as shown in Fig. 8c. It showed that the ternary slag can be carbonized and nitride by the CH4 –N2 –H2 gas mixture at low temperature. In the reaction process, the following reactions may exist: CaTiO3 + CH4 = TiO + CaO + CO + 2H2

(1)

CaTiO3 + 3CH4 = TiC + CaO + 2CO + 6H2

(2)

CaTiO3 + 2CH4 + 1/2N2 = TiN + CaO + 2CO + H2

(3)

TiO + TiC + TiN → Ti(N, C, O)

(4)

CaO + CaSiO3 = Ca2 SiO4

(5)

CaO + 2CaSiO3 = Ca3 SiO7

(6)

2CaSiO3 + Al2 O3 = Ca2 Al2 SiO7

(7)

CaSiO3 + Al2 O3 + SiO2 = CaAl2 Si2 O8

(8)

CaSiO3 + Al2 O3 + SiO2 + Fe2 O3 → Ca(Fe, Al)(Si, Al, Fe)O6

(9)

Equations (7–9) are the formation of gehlenite, anorthite, and hedenbergite aluminian.

Conclusion The crystallization, reduction, and carbonitridation of TiO2 –CaO–SiO2 ternary slag were studied in this work, and the selective precipitation and following carbonitridation in the CH4 –N2 –H2 gas mixture have been proved to be feasible. The following conclusions could be drawn: (1) The perovskite could be used as the titanium-rich phase in the selective precipitation process, and with the increase in basicity, the content of perovskite increased.

344

G. Fan et al.

(2) The perovskite could be reduced to Ti(C,O) in CH4 –H2 gas mixture and Ti(C,N) in CH4 –N2 –H2 gas mixture, and the addition of N2 could lower the reaction temperature. (3) The addition of CH4 N2 O and Fe2 O3 could improve the reaction condition, which was beneficial for the reaction. Acknowledgements Thanks are given to the financial supports from the National Natural Science Foundation of China (51674053, 51604046), National Key R&D Program of China (2018YFC1900500), Fundamental and Frontier Research Project of Chongqing (cstc2017jcyjAX0322), Graduate Scientific Research and Innovation Foundation of Chongqing (CYB19003), Undergraduate Scientific Research and Innovation Foundation of Chongqing (S201910611297), and Young Elite Scientists Sponsorship Program by CAST (2018QNRC001).

References 1. Fan G, Jie D, Lv X, Hu M (2018) Effect of basicity on the crystallization behavior of TiO2 – CaO–SiO2 ternary system slag. Cryst Eng Comm 20:5422–5431 2. Shi H, Feng K, Wang H, Chen C, Zhou H (2016) Influence of aluminium nitride as a foaming agent on the preparation of foam glass-ceramics from high-titanium blast furnace slag. Int J Miner Metall Mater 23:595–600 3. Yang H, Ma ML, Gao ML, Xue XX, Tang YQ (2009) Research on heat treatment process of foam glass prepared by titania-bearing blast furnace slag. Adv Mater Res 79–82:1587–1590 4. Kong XW, Ren LL, Ai X, Zhang J (2013) Preparation of a new unburned brick from Ti-bearing blast furnace slag and PVA modified by Epikote. Adv Mater Res 785–786:328–331 5. R Huang, C Bai, X Lv, S Liu (2012) Preparation of Titanium alloy from titania-bearing blast furnace slag. In: 3rd international symposium on high-temperature metallurgical processing, pp 75–83 6. Jiao H, Tian D, Wang S, Zhu J, Jiao S (2017) Direct preparation of Titanium alloys from Ti-bearing blast furnace slag. J Electrochem Soc 164:511–516 7. Gao J, Zhong Y, Guo Z (2016) Selective precipitation and concentrating of perovskite crystals from Titanium-bearing slag melt in supergravity field. Metall Mater Trans B 47:2459–2467 8. Gao J, Zhong Y, Guo Z (2016) Selective separation of Perovskite (CaTiO3 ) from Titanium bearing slag melt by super gravity. ISIJ Int 56:1352–1357 9. He S, Sun H, Tan DG, Peng T (2016) Recovery of Titanium compounds from Ti-enriched product of alkali melting Ti-bearing blast furnace slag by dilute sulfuric acid leaching. Procedia Environ Sci 31:977–984 10. R Zhang, D Liu, BS, G Fan, H Sun (2019) Thermodynamic and experimental study on the reduction and carbonization of TiO2 through gas-solid reaction. Int J Energy Res 43:1782–1801 11. R Zhang, J Dang, D Liu, Z Lv, G Fan, L Hu (2019) Reduction of perovskite-geikielite by methane–hydrogen gas mixture: thermodynamic analysis and experimental results. Sci Total Environ 12. Dang J, Fatollahi-Fard F, Pistorius PC, Chou KC (2017) Synthesis of Titanium Oxycarbide from concentrates of natural ilmenite (weathered and unweathered) and natural rutile, using a methane-hydrogen gas mixture. Metall Mater Trans B 48:2440–2446 13. Dang J, Fatollahi-Fard F, Pistorius PC, Chou KC (2018) Synthesis of Titanium Oxycarbide from Titanium slag by methane-containing gas. Metall Mater Trans B 49:123–131

Crystallization and Carbonization of TiO2 –CaO–SiO2 Ternary Slag

345

14. Rezan SA, Zhang G, Ostrovski O (2012) Effect of gas atmosphere on carbothermal reduction and nitridation of Titanium Dioxide. Metall Mater Trans B 43:73–81 15. Cao Z, Wei X, Jung IH, Du G, Qiao Z (2015) Critical evaluation and thermodynamic optimization of the Ti–C–O system and its applications to Carbothermic TiO2 reduction process. Metall Mater Trans B 46:1782–1801

Gravity Separation of Zinc Mine Tailing Using Wilfley Shaking Table to Concentrate Hematite Jonathan Tenório Vinhal, Raquel Húngaro Costa, Amilton Barbosa Botelho Junior, Denise Crocce Romano Espinosa and Jorge Alberto Soares Tenório

Abstract The flotation process is used to concentrate the zinc ore known as willemite. The process generates a residue consisted of hematite and dolomite. The residue used in this work presents desegregated hematite and dolomite particles, previously identified by MEV. In such a way, this work aims through gravity separation to concentrate the mineral phase hematite. The Wilfley shaking table was used to process 500 g of the tailing. Three different materials were obtained after the separation process: hematite concentrate, middling, and gravity separation tailing (GST). The characterization was carried out in SEM-EDS, XRD, and ICP-OES. SEM-EDS presented for concentrate, an iron content increasing from 7 to 44.5%. In concentrate, the XRD showed hematite peak intensity greater than the initial sample, different from other fractions. Analyses in ICP-OES showed that the concentrate material had 40.4% of iron, while in raw material, it was 7%. Therefore, gravity separation using Wilfley shaking table is seen as an interesting alternative since it reduces the zinc mine tailing, requires less energy and makes possible the material return to process in the steel industry. Keywords Gravity separation · Wilfley shaking table · Hematite · And extractive processes

Introduction As reported by International Lead and Zinc Study Group (ILZSG) [1], in 2018 the mine production, reached over 12 million tonnes of zinc, maintaining the noticed high demand in last year. According to Abkhoshk et al. [2], much of the zinc consumption has increased due to construction industries and will keep such behavior for the next years. Besides it, growth in the consumer durables and automobile industries will

J. T. Vinhal (B) · R. Húngaro Costa · A. B. Botelho Junior · D. C. R. Espinosa · J. A. S. Tenório Department of Chemical Engineering, Polytechnic School of University of Sao Paulo, São Paulo, SP, Brazil e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_33

347

348

J. T. Vinhal et al.

also be responsible for remaining the demand high. Mining or beneficiation process is the stage to concentrate zinc ore before sending it to the refining process. In the mining industry after mechanical and chemical extraction processes is obtained besides the product, mining tailings. That material is sent to tailing ponds, retained behind dams and embankments. For mining industry that is an issue due to the material volume, which can cause accidents and contamination of watercourses considering the metals. For instance, Tara mine in Ireland is one of the largest zinclead mines, producing every year nearly 2.7 million tonnes of zinc and lead concentrates. The tailings composed of high content calcium and magnesium coming from crushing, grinding, froth flotation, drying, and cycloning are placed on tailing ponds sized about 160 ha and 12.5 m high earthfill dam [3]. The residues produced on beneficiation processes have stimulated some studies. Liu et al. [4] proposed to produce glass ceramic foam from lead-zinc mine tailing, using a residue coming from Chenzhou, China, consisted of 25% Fe2 O3 and 28% CaO. On the other hand, Wang et al. [5] conducted a study to produce Ultra-High Performance Concrete replacing cement by different proportions of lead-zinc mine tailing. Not just zinc, but iron demand has increased over the years as a result of population growth. During 2017, the steel demand reached 587.4 million tonnes. Commonwealth of Independent States region, Europe, Asia, North America, and South America together were responsible for 73% of all global demand steel [6]. Brazil is the tenth largest iron exporter, maintaining a commercial relationship with more than 110 other countries. The three largest importers of Brazilian ores are the USA, Turkey, and Germany. Brazilian iron ore exportation in 2017 represented 3% of all ore exported worldwide. In 2018, Brazil has exported 6.9 million tons of iron [7]. Considering the iron demand all over the world and consequently an increase in iron production, recovering the metal even from low iron content materials has become attractive. Gravity separation and magnetic separation can be applied in order to concentrate iron oxide. According to Mackay et al. [8], the gravity separation using Wilfley shaking table separates materials based on particle densities. Since hematite and dolomite particles have different densities, 5.25 and 2.87 g/cm3 , respectively, and that method requires less energy, this paper aims to concentrate hematite from a zinc mining tailing composed of 3% SiO2 , 7% Fe2 O3 , and 90% CaMg(CO3 )2 through gravity separation using Wilfley shaking table.

Materials and Methods Gravity Separation The MEV-EDS of zinc mine tailing is presented desegregated hematite and dolomite particles which makes possible the physical separation of the two compounds. The

Gravity Separation of Zinc Mine Tailing Using Wilfley …

349

residue is composed of Fe 7%, Mg 10.4%, Ca 19%, Si 1.75%, and Zn 1.7%, previously characterized by inductively coupled plasma optical emission spectrometry, Agilent 710 series ICP-OES equipment. 50% of zinc mine tailing presents grain size up to 48.74 μm and 90% up to 250 μm. The gravity separation test was conducted with 500 g of residue in a Wilfley shaking table. Three water flows were evaluated assuming different positions on the table. The flows placed from feeding to the deposit vessels were 975, 700, and 70 L/h.

Materials Characterization X-ray diffraction (XRD) was carried out on a Rigaku equipment, Miniflex 300 model, with Kα copper radiation (λ = 1.5418 Å), 30 kV and 10 mA power. The analyses were performed between 2° and 80°, steps of 0.02° and a speed of 5°/min. The chemical concentration of products was provided by ICP-OES. The samples were digested on microwave CEM, MARS6 model, using 8 mL of HCl, 5 mL of HNO3 and 1 mL of HF for hematite concentrate; 5 mL of HCl, 3 mL of HNO3 , and 2 mL of HF for middling and tailing, and neutralized with 1 mL of H3 BO3 for each 1 mL used in the digestion. Scanning electron microscopy (SEM) was conducted on Phenom, ProX model with coupled energy dispersive spectrometer (EDS).

Results and Discussion Gravity Separation Figure 1a shows the setting up shaking table with the outputs, tailing, middling, and concentrate. It is also seen the separation of zinc mine tailing, which creates a black line composed of higher density particles. Figure 1b shows the material feeding and the lower density particles being drag down by water flows. At last, Fig. 1c is the concentrate material composed of greater density particles. Since the material is constituted of hematite and dolomite, which have different colors, black and pink, respectively, the separation can be improved adjusting the settings to make the black line wider. Thus, the launder trays can be moved to increase the collection efficiency of concentrate, middling, and tailings. Figure 2 is the flowchart of zinc mine tailing gravity separation with three output streams. At the end of the process, after drying, it was obtained 34.4 g of concentrate, 207.6 g of middling and 258 g of gravity separation tailing (GST). It is possible to notice in Fig. 3 that the separation has occurred. The three output materials have different aspects when compared to the initial sample (a). The concentrate (d) is the darkest when compared to middling and GST.

350

J. T. Vinhal et al.

Fig. 1 Gravity separation in Wilfley shaking table. a Setting up of shaking table and material outputs. b Lower density particles being dragged down and c greater density particles being concentrated

500g

Gravity separa on

Gravity separa on tailing (GST)

Middling

Concentrate

Fig. 2 Flowchart of zinc mine tailing gravity separation with three material outputs: concentrate, middling, and gravity separation tailing (GST)

Materials Characterization • Scanning Electron Microscopy with Energy-Dispersive Spectroscopy (SEMEDS)

Gravity Separation of Zinc Mine Tailing Using Wilfley …

351

Fig. 3 Images of zinc mine tailing gravity separation: a Zinc mine tailing, 0 b GST, c middling, and d concentrate

Figure 4 shows the backscattered electron image obtained by SEM and EDS spectrum for (a) zinc mine tailing, (b) GST, (c) middling, and (d) concentrate. SEM-EDS of the zinc mine tailing and the output materials refer to a microregion. Fractions (c) and (d) are composed of larger particles when compared to (a) and (b). Knowing the hardness of dolomite (3.5–4) and hematite (5–6), it can be inferred that the material with higher grain size may have greater hematite contents since iron oxide is more abrasive than dolomite. In (c), it can be found particle sizes up to 500 μm, and in (d) up to 400 μm, whereas in (a) and (b), the largest particles present diameters of 200 μm and 150 μm, respectively. Roy and Das [9], in their study, separated hematite and goethite from kaolinite and quartz through dense media separation. They noticed that as smaller is the material granulometry, lower will be the contents of dense minerals, hematite, and goethite. The authors identified 68.47% of hematite and goethite for grain size 75–150 μm. The chemical composition of the zinc mine tailing and output materials provided by EDS are shown in Table 1. Comparing to (a), it is noted in (b) and (c) that Mg and Ca contents increases while Fe decreases. For the concentrate (d), there is the inverse behavior, reducing Mg and

352

J. T. Vinhal et al.

Fig. 4 General image obtained from the scanning electron microscopy and EDS spectrum of a Zinc mine tailing, b GST, c middling, and d concentrate

Gravity Separation of Zinc Mine Tailing Using Wilfley …

353

Table 1 EDS–elementary concentration of zinc mine tailing and output materials from gravity separation Mg (%) Zinc mine tailing (a)

Ca (%)

Fe (%)

Zn (%)

Si (%)

8.56

12.98

7.02

0.78

2.44

GST (b)

10.38

18.16

6.52

1.28

2.65

Middling (c)

12.36

17.67

4

–

1.53

4.98

6.34

44.52

–

1.97

Concentrate (d)

Ca contents, and increasing Fe. The residue composed of Fe 7.02% is concentrated to Fe 44.52%, equivalent to 64% of Fe2 O3 . Mackay et al. [8], in their study, propose to concentrate minerals containing Nb and rare earth element (REE), one of these minerals is columbite-(Fe) [Fe2 + Nb2 O6 ], density 5.3–7.3 g/cm3 . It is shown in their work how gravity separation using Wilfley shaking table is able to concentrate indicator minerals. Performing a recovery study, they identified effective concentrations of Nb, REE, Y, and Fe, which can be as hematite, magnetite, and columbite-(Fe) phases. According to the authors, they concentrated samples of 5% Fe in materials up to 37% Fe. • X-Ray Diffraction (XRD) and Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-OES) Figure 5 shows the XRD of the zinc mine tailing and the materials obtained in the separation process. GST fraction (b) presents an increase in dolomite phase peaks when compared to the zinc tailing (a). The chemical analysis in (b) by ICP-OES indicates the concentration of Mg and Ca, and the decreasing of Fe content, since the levels of Mg, Ca, and Fe in (b) are, respectively, 11.6%, 20.5%, and 4.5%, while in zinc tailing (a) are 10.4, 18.6, and 7%. The middling (c) is composed of 10% Mg, 18% Ca, and 4.4% Fe. Fig. 5 X-ray diffraction of zinc mine tailing gravity separation. a Zinc mine tailing, b GST, c middling, and d concentrate

354

J. T. Vinhal et al.

The XRD of concentrate (d) presents higher hematite peaks. The chemical composition by ICP-OES shows 3.5% Mg, 9% Ca, and 40.4% Fe. According to AbakaWood et al. [10], the gravity separation using Wilfley shaking table of iron oxide silicate tailing with 26.2% of Fe provides a concentrate fraction of 49.3% Fe. The Abaka-Wood study recovered 42% of Fe present in the tailing. The results obtained in this paper showed 39.3% of Fe recovery. These results confirm the concentration of iron oxide in (d), as previously proposed in SEM-EDS results. For a future study, it is intended to perform the reduction kinetic investigation of self-reduction briquettes made of hematite concentrate and coal.

Conclusion • The study presented output materials with different aspects, in which the concentrate showed a darker color compared to middling and GS tailing. • 34.4 g of concentrate, corresponding about 7% of the initial mass sample, was obtained at the end of gravity separation. • Middling and concentrate showed particle greater than GST. In other words, it was noticed particle diameters up to 500 μm in middling, in concentrate up to 400 μm, and in GST up to 150 μm. • Higher hematite peaks on XRD were identified in concentrate, and the chemical analysis by ICP provided iron content equal to 40%. The Fe recovery in that study was 39.3%. • The concentrate is composed basically of hematite and dolomite. Besides iron recovery from concentrate, it is known that the material can act as a fluxing agent in steel industry processes, due to its dolomite fraction. Acknowledgements This technical effort was performed in support of CNPq (306936/2016-0) and Ore and Industrial Waste Treatment Laboratory, Department of Mining and Petroleum Engineering, USP Polytechnic School.

References 1. International Lead and Zinc Study Group (2019). Available at http://www.ilzsg.org/static/ statistics.aspx?from=1. Accessed 30 Aug 2019 2. Abkhoshk E, Jorjani E, Al-Harahsheh MS, Rashchi F, Naazeri M (2014) Review of the hydrometallurgical processing of non-Sulfide Zinc ores. Hydrometallurgy 149(October):153– 167. https://doi.org/10.1016/J.HYDROMET.2014.08.001 3. Quille M.E, O’Kelly BC (2010) Geotechnical properties of Zinc/Lead mine tailings from tara mines, Ireland. Geoenvironmental Engineering and Geotechnics, Reston, VA, May 2010, pp 111–117. https://doi.org/10.1061/41105(378)16

Gravity Separation of Zinc Mine Tailing Using Wilfley …

355

4. Liu T, Lin C, Liu J, Han L, Gui H, Li C, Zhou X, Tang H, Yang Q, Lu A (2018) Phase evolution, pore morphology and microstructure of glass ceramic foams derived from tailings wastes. Ceram Int 44:14393 5. Wang Xinpeng, Rui Yu, Shui Zhonghe, Zhao Zemin, Song Qiulei, Yang Bo, Fan Dingqiang (2018) Development of a novel cleaner construction product: ultra-high performance concrete incorporating Lead-Zinc tailings. J Clean Prod 196(September):172–182. https://doi.org/10. 1016/J.JCLEPRO.2018.06.058 6. Mercier F, Mabashi D, Steidl C (2019) Steel market developments – Q4 2018. Sci Technol Innov (STI) 7. International Trade Administration (2019) Steel export report. Brazil 8. Mackay DAR, Simandl GJ, Luck P, Grcic B, Li C, Redfearn M, Gravel J (2015) British Columbia ministry of energy and mines. British Columbia geological survey paper 2015, vol 1, pp 189–195 9. Roy S, Das A (2008) Characterization and processing of low-grade iron ore slime from the jilling area of India. Miner Process Extr Metall Rev. https://doi.org/10.1080/08827500801997886 10. Abaka-Wood G, Quast K, Zanin M, Addai-Mensah J, Skinner W (2019) A study of the feasibility of upgrading rare earth elements minerals from iron-oxide-silicate rich tailings using Knelson concentrator and Wilfley shaking table. Powder Technol 344(Feb-2019):897–913. https://doi.org/10.1016/j.powtec.2018.12.005

Minimization of Copper Contamination in Steel Scrap Hyunsoo Jin and Brajendra Mishra

Abstract Copper and tin, as tramp elements in the steel scrap, cause some harmful effects, such as hot shortness caused by a loss of ductility and surface defects. It is also difficult to maintain the quality of the product because the amount of the residual constituents in steel scrap is not consistent. The secondary steel products require consistent copper content and standard due to the limited use of the product for the various applications. Thermodynamically, removing copper from scrap is a viable option, but in reality, the impurities and the included copper in the melt of steel scrap are difficult to remove by conventional methods. Besides, the research of the recycled steel scrap regarding iron and impurities is limited, and it needs to be conducted, in terms of physical and chemical techniques, as the preliminary study to find the efficient separation method. Keywords Secondary steel · Tramp elements · Steel recycling

Introduction Issue of Recycling Secondary Used Steel Scrap The recycling of steel scrap is mainly constrained by its concentration of the ‘tramp element.’ Copper and Tin are two of the most common tramp elements. Those tramp elements not only cause hot shortness in steel product but also limit the range of grades steel shop can produce. Table 1 shows some examples of steel grades with a maximum Cu residuals [1].

H. Jin · B. Mishra (B) Worcester Polytechnic Institute, Worcester, USA e-mail: [email protected] H. Jin e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_34

357

358

H. Jin and B. Mishra

Table 1 Maximum tolerable Cu content in common steel grades (wt%) Interstitial free

Deep drawing quality

Drawing

Commercial

Structural

Fine wire

Rebar

0.03

0.04

0.06

0.1

0.12

0.07

0.4

Theoretical Aspects Theoretically, controlling the copper impurities to under 0.2 wt% is proven by aspects of thermodynamics. To be specific, Nakajima et al. [2] investigated the distribution of elements between the metal, gas, and slag phases in current steel refining practice. The results provide that the copper and iron are not alloying with each (see Fig. 1). Besides, Zaitesv et al. [3] checked the activities of copper and iron at high temperature and this represents copper and iron are not strongly interacted with each other (see Fig. 2). However, in the field, the problem is these theories did not work in the real field.

Fig. 1 Traditional recycling process [4]

Minimization of Copper Contamination in Steel Scrap

359

Fig. 2 Cu level in steel scrap that used traditional method

Recycling Process Traditional Recycling Process In the steel scrap field, the recycling process depends on the target materials what the company wants to recover or recycle. Although the process is not the same, the steps are almost similar and the difference is the order or some additional process for the target materials. Figure 1 is the traditional recycling process; the main step is feeding, shredding, sort and extracting pollutants, magnetic separation, and plastic sorting. Specifically, for the separation, magnetic, air, Eddy-current, and sink-float/fluidized bed density are used. The challenging part of the shredding and sorting process is maintaining the level of content. For example, in tramp elements, it is hard to maintain the impurities level because the contents of these impurities depend on the feeding materials and these feeding materials are not consistency. Figure 2 shows the range of copper that is included in steel scraps. This data is obtained from Company A and this data is based on the conventional method.

Gamma-Ray Process Company A used a gamma-ray process that is almost similar to the traditional method. It also faced the problem that Cu contaminant is fluctuation and it is too high to use widely (Fig. 2). Company A developed a new strategy called ‘The Cross-Belt Recycled Metal Analyzer.’ Briefly, the principle is emitting the radioactive source (Cf 252) and when the scrap nucleus is excited, they emit the gamma-ray than the detector collects the gamma-ray and show the results promptly. When they receive the results, they can

360

H. Jin and B. Mishra

remove the high Cu contaminant part. It makes not only a lower copper contaminant but also maintains the range of copper. After using this equipment, Fig. 3 is the composition of Cu in steel scrap. This range shows the equipment can improve the process and reduce the tramp elements; however, in the future, if the feedstock Cu contaminant has increased, this solution will be faced on new problems to overcome. According to Company A, the three key factors of ‘Prompt Gamma Neutron Activation Analysis’ that makes it uniquely fit for scrap applications are 1. Inbound neutrons and outbound gamma-rays are very penetration in nature so there is interaction with all of the material as it passes through the tunnel 2. The reaction occurs at the speed of light and therefore, it is a real-time analysis • Scrap processors or steel mills can use it to control the process/chemistry of shredded scrap • The analyzer is capable of handling production rates of over 400tons/hour) 3. The reaction does not activate the material • There is no measurable residual radioactivity as the material exits the analyzer This strategy is efficient and easy to adjust to the existing field. However, these fields need to prepare future high contaminant Cu in waste steel. Several teams try to overcome this problem in the past. Thermodynamically, there are a lot of possible methods [5] existing but most of that is not efficient to use in the field or not work. This research is testing the solvent extracting method and electrochemical method to check the probability and develop efficient ways (Fig. 4) [5].

Fig. 3 Cu level in steel scrap that used gamma-ray method

Minimization of Copper Contamination in Steel Scrap

361

Fig. 4 The energy consumption for reducing the copper concentration [5]

Separating Methods To separate the tramp elements from the steel scrap, there are several strategies to overcome this problem. Figures 5 and 6 classify the separating methods by the shape of the beginning state [5]. In this project, the solvent extraction method and the slag melt method are conducted for different beginning state and composition. Efficiency and possibility are also one factor to consider.

Fig. 5 Methods to separate tramp elements from solid scrap, showing the separating agent applied in each method and the property difference exploited [5]

362

H. Jin and B. Mishra

Fig. 6 Methods to separate tramp elements from the melt, showing the separating agent applied in each method and the property difference exploited [5]

Copper Extraction Method Nonferrous Metal Bath The basic concept of this method is the different solubility of copper and steel. In the lead or aluminum bath, the copper solubility is higher than steel. According to the literature review, 1000 kg of leads can reduce the copper content in 1t of obsolete from 0.3 to 0.1%. The problem can cause human disease and not environmentally friendly. In contrast, the aluminum bath is attractive and one good possible way to conduct because the amounts of aluminum are unclear and aluminum–copper can bring valuable byproducts that are Al–Cu alloy. According to the reviewed paper, the solubility of copper in the aluminum bath is 65% at 730 °C and it can remove 80% of copper impurity. It needs only 20 min to remove [6, 7].

Slag Melts This method is using sulphur-containing slag to extract the impurity. The basic concept of this is that copper sulfide is more stable than iron sulfide at above 600 °C. 2[Cu] + FeS ↔ Cu2 S + [Fe]

Minimization of Copper Contamination in Steel Scrap

363

Table 2 Initial composition of ferrous bulk materials conducted by XRF Al

Cr

Ni

Si

Mn

Fe

Cu

1

ND

ND

ND

0.92

0.19

98.52

0.04

2

ND

ND

ND

1.34

0.27

97.82

0.04

3

ND

ND

ND

1.23

0.23

98.23

0.03

4

ND

ND

ND

1.94

0.49

94.19

0.03

Avg.

ND

ND

ND

1.36

0.30

97.19

0.035

According to the papers, 100 kg of slag needs to reduce 1t steel treatment. This method needs fewer amounts of slags than the nonferrous lead bath but it needs high temperature [8–11].

Electrochemical Electrochemical is based on the difference reduction potential of copper and iron. The required condition for this process is dissolving agents that need to make the elements dissolved into the liquid. If it is dissolved in the liquid, electro-winning is available. Copper and iron can be selectively recovered by using a suitable reduction potential.

Sample Information The sample was obtained from Company B and randomly selects four samples. To check the initial compositions, XRF analysis was conducted. Table 2 is the results by using XRF. The average of Cu contaminant is 0.035 wt%. This Cu amount is allowable at the secondary steel field. If it is hard to find a suitable sample, this project will make artificial Cu–Fe samples to see the feasibility.

Conclusion and Future Work In the secondary steel field, the ultimate goal is to reduce or remove the Cu impurities from steel scraps and previous research suggests several methods based on the thermodynamic aspect. In this study, three routes will be considering that is an electrochemical method, a nonferrous metal bath, and a slag extraction method. The samples accepted at Company A and B recycling fields. Before they send the samples, the samples were conducted basic separation processes in the field. The sample is roughly three, nonferrous Al–Cu samples, bulk steel samples, and bulk wired steel

364

H. Jin and B. Mishra

sample. By XRF and ICP–OES analysis, the initial composition of Cu–Al samples and bulk steel samples are checked. In the future, the Al bath extraction and slag extraction methods will be conducted and check the possibility and efficiency. For the electrochemical method, sulfuric acid will use as a separation agent and the mole value of sulfuric acid and dissolving time and the surface area will use as a variable.

References 1. Effect of tramp elements in flat and long products (1995) Final report, Contract no. 7210ZZ/555+ZZ/564, EGKS, Brussels 2. Nakajima K, Takeda O, Miki T, Matsubae K, Nakamura S, Nagasaga T (2010) Thermodynamic analysis of contamination by alloying elements in aluminum recycling. Envirion Sci Technol 44:5594–5600 3. Zaitsev AI, Shelkova NE, Litivina AD, Shakhpazov EK, Mogutnov BM (2001) An investigation of evaporation of liquid alloys of iron with copper. High Temp 39(3):388–394 4. Example of a Process of SHA Treatment at SITA. Available. http://eco3e.eu/en/base/sha/ 5. Katrin E, Serrenho AC, Allwood J (2019) Finding the most efficient way to remove residual copper from steel scrap. Metal Mater Trans B 50(3):1–16 6. Iwase M, Tokinori K, Ohshita H (1993) Iron Steelmaker 20(7):61–66 7. Iwase H, Ohshita H (1994) Steel Res 65(9):362–367 8. Cohen A, Blander M (1998) Metall Trans B 29b(2):493–495 9. Lee J (1997) Kupferproblematik beim Schrottschmelzen. Shaker, Aachen. ISBN 3-8265-23202 10. Jimbo I, Sulsky MS, Fruehan RJ (1988) Iron steelmaker 15(8):20–23 11. Savov L et al (2003) Copper and tin in steel scrap recycling. RMZ Mater Geoenvironment 50(3):627–640

Recycling of Blast Furnace Flue Dust with In-flight Reduction Technology: Reduction Behavior and Kinetic Analysis Jin Xu, Nan Wang, Min Chen and Haiyang Yu

Abstract Blast furnace (BF) flue dust is a by-product and collected from the gas cleaning systems during the blast furnace ironmaking process, which can be recycled as one of the secondary sources due to the valuable contents of iron and carbon. A novel in-flight reduction technology is considered to allow utilizing the large quantities of fine iron-bearing metallurgical dust directly to bypass the sintering/pelletization and conventional coke-making steps. In this work, the reduction behavior and kinetic mechanism of the blast furnace dust during in-flight process in hydrogen atmosphere are studied with lab-scale high-temperature drop tube furnace. The effects of temperature and gas composition on the reduction degree are examined. With the morphological observation, it is found that the unreacted shrinking core model can describe the in-flight reduction process. According to the kinetic analysis, the rate-controlling step is determined as the chemical reaction at the particle surface. The activation energy E a is determined to be 224.8 kJ/mol and the pre-exponent factor A as 7.2 × 106 m/s. Keywords Blast furnace flue dust · In-flight reduction process · Hydrogen · Reduction kinetics · Unreacted core model

Introduction During the blast furnace (BF) ironmaking process, a variety of residues including dust, sludge, and slag are generated as by-products [1]. As one of which, blast furnace (BF) flue dust is the solid component that collected from blast furnace gas cleaning system. The output of BF flue dust from off-gas cleaning facilities was estimated to be 70–110 kg per ton of steel production [2, 3]. Generally, BF flue dust contains many high contents of Fe (>30%) and C (>15%) with small amount of Ca, Si, Al, Zn, and Pb oxides [4, 5], which can be recognized as a secondary resource for recycling. However, most of the dust is stored as solid waste in landfills and cause serious J. Xu · N. Wang (B) · M. Chen · H. Yu School of Metallurgy, Northeastern University, Shenyang 110819, Liaoning, China e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_35

365

366

J. Xu et al.

environmental pollution. Therefore, it is significant to explore efficient treatment for the dust recycling. Currently, the prevalent treatment of the dust is fed by factories back to sintering. However, with the characteristics of fine particle size and zinc content, the utilization of BF flue dust in sintering process has long been difficult, because of the low gas permeability and the low removal efficiency for Zn [6, 7]. Several positive efforts have been also made toward the application of RHF process that makes it possible to recycle dust combining with de-zinc process [8, 9]. Rotary hearth furnace (RHF) process is considered as one of the effective technologies for treating metallurgical wastes based on the overall evaluation of treatment capacity, profitability, and operational flexibility. The metallurgical dust, sludge, scale, and pulverized coal are mixed in a given ratio and then pelletized into green pellets to produce direct reduction iron (DRI) after heated and reduced under temperature ~1573 K. During the heating process, the zinc oxide is reduced to metal zinc and gasified into the off-gas and collected as secondary dust shipped to zinc refining plant. However, the produced DRI has comparatively higher sulfur contents and lower physical strength, and at the same time, the heat transfer efficiency between the high-temperature gas and pellet layers is thought to be lower in RHF. Recent years, a novel in-flight reduction technology has been developed, in which fine iron ore concentrate is flash pre-reduced by gaseous reductants (H2 , CO, natural gas, et al.) at high temperature during a very short fly time, and the final reduction is completed in the molten bath. The in-flight reduction technology is expected to significantly reduce energy consumption and carbon dioxide emissions and eliminate the problematic sintering, pelletization, and coke-making steps of the blast furnace process. Considering the fewer restrictions of raw material used, the in-flight reduction technology can be considered to allow efficient recycling of the fine iron-bearing wastes, in which the iron contained in iron oxides is expected to be pre-reduced with flash process and extracted to hot metal finally while the non-volatile impurities are removed to the molten slag. Besides, other easy volatilization valuable elements such as Zn and alkalis can be separated and recycled in the exhausted gas simultaneously. Several experimental studies relevant to this process have been performed in many research efforts. Sohn and coworkers [10, 11] investigated the H2 and CO reduction of fine magnetite and hematite concentrate particles relevant flash process under various experimental conditions by using a drop tube reactor in the temperature range from 1423 to 1673 K and developed rate expressions of reduction kinetics based on the nucleation and growth theory. Qu et al. [12, 13] conducted a series of experiments on the in-flight reduction kinetics of fine hematite ore at the temperature range from 1550 to 1750 K for different reaction time (210–2020 ms) in a high-temperature drop tube furnace. Shimizu et al. [14] investigated the rapid reduction of fine ore particle transported with CO, H2 , and/or CH4 gas and analyzed the mechanism and kinetic of the reduction by CH4 gas with spherical wustite fine particles. However, no specific experimental study has been reported on the reduction of various metallurgical dusts with in-flight reduction technology at high temperature. As the novel in-flight reduction technology is still in the development stage, the reduction mechanism of the iron-bearing dust under various complex conditions is not clear. This paper aims

Recycling of Blast Furnace Flue Dust with In-flight Reduction …

367

to study the reduction behavior of BF flue dust during in-flight process. The effects of temperature, reaction time, and gas concentrations on the reduction behavior were investigated. The morphology and microstructure changes of the reduced BF flue dust were observed. A global kinetic model for the reduction of BF flue dust was further established based on the experimental results.

Experiment Experimental Apparatus The schematic diagram of the experimental apparatus is shown in Fig. 1. The hightemperature drop tube furnace mainly consists of a vertical tubular furnace housing an alumina tube (10 cm inner diameter, 130 cm long), a syringe pump powder feeding system, a gas delivery system, a water-cooled powder injection probe, a sample cooling, collecting system, and an off-gas outlet system. The vertical tubular furnace

Fig. 1 Schematic diagram of the experimental set-up

368

J. Xu et al.

was heated by six MoSi2 heating elements. The injection probe could be moved freely in vertical direction so as to change the length of the reaction zone and adjust the reduction time of BF flue dust particle simultaneously. The experimental temperature of the reaction zone was controlled by a PtRh30-PtRh6 thermocouple located at the center of reactor tube. In the sample collecting tube at the bottom of the furnace, the N2 was used to cool and avoid re-oxidation of the reduced sample.

Raw Material The BF flue dust used in this work was provided by Baosteel, China. The main chemical compositions of original BF flue dust are given in Table 1, and the total iron and carbon contents in BF flue dust are about 43.6% and 20.5%, respectively. Owing to the strict controlling of raw materials in Baosteel, it can be noted that the amount of zinc oxide in original BF flue dust is less than 0.5%. The XRD result is shown in Fig. 2a suggests that iron exists in hematite, wustite, metallic iron, and calcium ferrite phases. The true density of BF flue dust is about 3462 kg/m3 measured by BelPycno-MicrotracBEL densimeter. The size distribution of the BF flue dust is determined by the method of laser diffraction on a volume basis, as shown in Fig. 2b. In this work, the size range of 50–74 µm was used for the experiment, and the volume median diameter of 62 µm was employed to calculate the residence time. Morphological observation shows that iron-bearing phase and carbon coexist in the original BF flue dust as shown in Fig. 3a. Figure 3b shows the iron-bearing phase with an irregular shape and smooth surface. Hydrogen of 99.99% purity was used as Table 1 Main chemical composition of original BF flue dust used in the present experiment Composition

TFe

FeO

SiO2

CaO

Al2 O3

MgO

ZnO

C

Mass (%)

43.6

15.5

6.5

5.5

2.9

0.7

1573 K) and greater H2 content (>50% in reductive gas mixtures) are recommended to obtain a higher reduction degree for BF flue dust during in-flight process. (2) The surface morphology observation of the reduced dust samples by SEM found that most of the particle keeps irregular shape while a little of the particle changes to spherical shape. The morphology of the enlarged detailed of single reduced particle presented that the particle surface became coarser and some pores generated on the surface. An unreacted core and a product layer could be identified by colors on the cross section of the reduced BF flue dust particle through the cross sections observation of the reduced BF flue dust particle samples. (3) Based on the unreacted core model, the reduction kinetics of BF flue dust particles during the in-flight process were analyzed. The kinetic results showed that the rate-controlling step was the chemical reaction at the particle surface. The chemical reaction constant k r at different conditions was obtained and the activation energy E a is determined to be 224.8 kJ/mol and A as 7.2 × 106 m/s. Acknowledgements This research was financially supported by the National Natural Science Foundation of China (Grant Nos. 51574065, 51574066, 51774072, 51774073).

References 1. Xu G (2015) Applied basic research on recycling of blast furnace dust. Ph.D. thesis, University of Science and Technology, Beijing 2. Deng YC, Jia SQ, Wu SL, Jiang YJ (2015) Removal of the harmful elements and iron recovery from the blast furnace gas ash by chloridizing roasting. Iron Steel Vanadium Titanium 36:51–56 3. Lanzerstorfer C, Kroppl M (2014) Air classification of blast furnace dust collected in a fabric filter for recycling to the sinter process. Resour Conserv Recycl 86:132–137 4. Zhao D, Zhang JL, Wang GW, Conejo AN, Xu RS, Wang HY, Zhong JB (2016) Structure characteristics and combustibility of carbonaceous materials from blast furnace flue dust. Appl Therm Eng 108:1168–1177 5. Leimalm U, Lundgren M, Okvist LS (2010) Off-gas dust in an experimental blast furnace part 1: characterization of flue dust, sludge and shift fines. ISIJ Int 50(11):1560–1569 6. Stecko J, Stachura R, Nieslar M, Bernasowski M, Klimczyk A (2018) Utilisation of metallurgical sludge by multi-layer sintering. Ironmaking Steelmak 45(9):779–786

376

J. Xu et al.

7. Lanzerstorfer C, Bamberger-strassmayr B, Pilz K (2015) Recycling of blast furnace dust in the iron ore sintering process: investigation of coke breeze substitution and the influence on off-gas emissions. ISIJ Int 55(4):758–764 8. Hu T, Lv XW, Bai CG (2016) Enhanced reduction of coal-containing titanomagnetite concentrates briquette with multiple layers in rotary hearth furnace. Steel Res Int 87(4):494–500 9. Chung SH, Kim KH, Shon II (2015) DRI from recycled iron bearing wastes for lower carbon in the blast furnace. ISIJ Int 55(6):1157–1164 10. Wang HT, Sohn HY (2012) Hydrogen reduction kinetics of magnetite concentrate particles relevant to a novel flash ironmaking process. Metall Mater Trans B 44(B):133–145 11. Chen F, Mohassab Y, Zhang SQ, Sohn HY (2015) Kinetics of the reduction of hematite concentrate particles by carbon monoxide relevant to a novel flash inromaking process. Metall Mater Trans B 46(B):1716–1728 12. Qu Y, Yang Y, Zou Z, Zeilstra C, Meijer K, Boom R (2015) Reduction kinetics of fine hematite ore particles with a high temperature drop tube furnace. ISIJ Int 55(B):952–960 13. Qu Y, Yang Y, Zou Z, Zeilstra C, Meijer K, Boom R (2015) Melting and reduction behavior of fine hematite ore particles. ISIJ Int 55(1):149–157 14. Takeuchi N, Nomura Y, Ohno K, Maeda T, Nishioka K, Shimizu M (2007) Kinetic analysis of spherical wustite reduction transported with CH4 gas. ISIJ Int 47(3):386–391 15. Srinivasan NS, Lahiri AK (1977) Studies on the reduction of hematite by carbon. Metall Mater Trans B 8(b):175–178 16. Tiwari P, Bandyopadhyay D, Ghosh A (1992) Kinetics of gasification of carbon and carbothermic reduction of iron oxide. Ironmaking Steelmak 19:464–468 17. Piotrowski K, Mondal K, Wiltowski T, Dydo P, Rizeg G (2007) Topochemical approach of kinetics of the reduction of hematite to wustite. Chem Eng J 131(1–3):73–82 18. Li B, Wang H, Wei YG (2012) Kinetic analysis for non-isothermal solid state reduction of nickel laterite ore by carbon monoxide. Trans Inst Min Metall C 121(C):178–184

Recycling Technologies of Zn–C Batteries: Review and Challenges for a Circular Economy in Colombia Natalia Cardona-Vivas, Mauricio A. Correa and Henry A. Colorado

Abstract In this research, a comprehensive literature review was conducted over the valorization, utilization, and circular economy of Zn–C batteries after their useful life. The status of this battery waste in Colombia a Latin America is also presented, as well as some case studies conducted in Medellin involving small business enterprises and research at the university level. Therefore, some characterization including scanning electron microscopy and x-ray diffraction was included. Results show the high potential of this waste for using in construction materials, electronic materials, and formulations tailored for agriculture fertilizers. Keywords Batteries · Waste · Recycling · Circular economy · Characterization

Introduction A battery is an electrochemical device that can convert chemical energy into electrical energy. It consists of an anode (negative electrode), a cathode (positive electrode), an electrolyte (ionic conductor), separators, and an external housing. What differentiates the various batteries are the materials used as electrodes and electrolytes, which give each one special characteristic [1]. The anode selection criteria depend on its efficiency as a reducing agent, high performance Ah/g, good conductivity, stability, ease of manufacture, and low cost. Zinc has been a predominant anode because it has these properties. In the case N. Cardona-Vivas · H. A. Colorado CCComposites Laboratory, Universidad de Antioquia UdeA, Calle 70 N°. 52-21, Medellín, Colombia M. A. Correa Grupo de Investigación en Ingeniería y Gestión Ambiental, Universidad de Antioquia UdeA, Calle 70 N°. 52-21, Medellín, Colombia H. A. Colorado (B) Universidad de Antioquia, Facultad de Ingeniería, Bloque 20, Calle 67 No. 53-108, Medellin, Colombia e-mail: [email protected] © The Minerals, Metals & Materials Society 2020 X. Chen et al. (Eds.), Energy Technology 2020: Recycling, Carbon Dioxide Management, and Other Technologies, The Minerals, Metals & Materials Series, https://doi.org/10.1007/978-3-030-36830-2_36

377

378

N. Cardona-Vivas et al.

of the cathode, it is required to be a good oxidizing agent, stable when in contact with the electrolyte, and must have a useful working voltage. Most of the common cathode materials are metal oxides. For the electrolyte, a material with good ionic conductivity, but not electrically conductive, is required to avoid an internal short circuit. In addition, it is important the high chemical stability the electrode materials, good resistance to temperature changes, safety for handling, and low cost. Most electrolytes are aqueous solutions, except in thermal and lithium anode batteries, where molten salts and other non-aqueous electrolytes are used to prevent the reaction of the anode with the electrolyte [2]. Although the term battery is frequently used elsewhere, the electrochemical unit that carries out the reaction is a cell. A battery is composed of one or more of these cells, connected in series or parallel, depending on the desired voltage and output capacity [2]. There are two types of household batteries: single-use or primary and rechargeable cells or secondary cells (Table 1). Among the primary batteries, the most commonly used are zinc-carbon and alkaline-manganese. These batteries usually come in sizes AAA, AA, C, D, and 9 V. Zn–C batteries have been studied for hundreds of years. The two types of Zn–C batteries most used are those of Leclanché and those of the zinc chloride system. These batteries are characterized by low cost, availability, and wide performance in a large number of applications [2]. Alkaline batteries occupy most of the primary battery market: approximately 60%. These are the most economical option that provides great performance at low cost for the operation of various devices [3]. Between 2002 and 2008, about 14,000 tons of zinc, 13,000 tons of manganese, 60 tons of cadmium, 15 tons of chromium, 100 tons of nickel, 30 tons of lead, 350 kg of mercury, and 350 kg of lithium have been discharged into the environment in Colombia, mainly in landfills and open dumps [4]. This represents a serious threat to the public health safety and therefore is required a coordinated action to prevent it.

Materials and Processes Zn–C Battery A zinc–carbon battery is a type of dry cell that offers a potential of 1.5 volts between a zinc metal electrode and a carbon rod through an electrochemical reaction, between zinc and an electrolyte-mediated manganese dioxide. Zinc components work as anode with a negative potential, while the inert carbon rod is the positive cathode. General-purpose batteries can use an aqueous paste of ammonium chloride as an electrolyte, possibly mixed with some solution of zinc chloride [2].

Recycling Technologies of Zn–C Batteries: Review and Challenges …

379

Table 1 Classification and use of primary batteries Battery type

Components

Characteristics

Primary (Non-rechar(Zn/C) or Leclanché dry batteries)

Manganese dioxide, graphite, carbon (cathode) Zinc sheet metal (anode) Ammonium chloride, zinc chloride (electrolyte)

Is a low-cost battery. For light applications like flashlights and radios, all types of electrical and electronic equipment simple and low consumption. Called “common batteries”

Magnesium and aluminum batteries

Manganese dioxide with acetylene (cathode) Magnesium alloy, aluminium (anode) Magnesium perchlorate with barium and lithium, chromium chloride (electrolyte)

This battery has two main advantages over the zinc-carbon battery, namely twice the service life or capacity of the zinc battery of equivalent size and the ability to retain this capacity, during storage, even at elevated temperatures. They are little used for their “voltage delay” and the corrosion of Mg

Zinc/manganese dioxide (Zn/MnO2 ) or alkaline

Manganese Dioxide (cathode) Zinc powder (anode) Concentrated caustic, usually potassium hydroxide (electrolyte)

They can be used on the same devices as common batteries, but they perform better

Mercury oxide (Zn/HgO)

Mercury oxide (cathode) Zinc (anode) Potassium hydroxide, sodium hydroxide (electrolyte)

The zinc/mercuric oxide battery is noted for its high capacity per unit volume, constant voltage output, and good storage characteristics. For hearing aids and medical equipment. Usually button type. They contain about 30% mercury

Zinc/air (Zn/O2 )

Oxygen (cathode) Zinc (anode) Alkaline electrolyte

For portable applications. They have a large number of tiny holes in their surface. High capacity. They contain more than 1% of mercury

Silver oxide

Silver oxide (cathode) Zinc amalgam (anode) Potassium hydroxide (electrolyte)

Usually small button type, they contain about 1% of mercury. It is an alternative to mercury oxide

Adapted from [2]

380

N. Cardona-Vivas et al.

Waste Disposal There are different alternatives for the final disposal of the batteries: • Landfill: so far, most of the batteries, especially primary batteries, dispositions in municipal waste and sanitary materials. • Stabilization: it represents a previous treatment of the batteries to avoid the contact of the metals with the environment in a landfill. The process is not widely used since it involves high costs. • Incineration: it is used when batteries are disposed in a municipal waste facility and then taken to combustion sites. The incineration of this waste can generate emissions of polluting gases such as mercury, cadmium, lead, and dioxins. • Recycling: the metals present in the batteries can be recycled through hydrometallurgical and pyrometallurgical processes. These recycling processes are being studied in different parts of the world [1].

Waste Recovery Processes In recent years, many processes have been developed for battery recycling due to new environmental regulations [5]. Most battery contents can be technically recovered through mechanical and chemical treatments. The recovery of these materials serves to be used in other applications.

Physical Processes Physical processes are a pre-treatment that should be done to battery waste, prior to any other recycling process. The physical processes consist of a sequence of steps: classification, magnetic separation, dismantling, and crushing. The classification is related to a manual process in which Zn–C batteries are separated from the rest of the batteries. In the dismantling stage, it is sought to separate the battery components into their different component materials. Magnetic separation is used to separate the metal from the other materials. Finally, crushing is generally performed using a ball mill, in order to reduce particle size and to obtain powders that improve the efficiency of leaching process. However, crushing is an expensive step due to the energy demand of the process [5] (Fig. 1).

Chemical Processes • Pyrometallurgical: they are mainly associated with the production of iron, ferromanganese alloys, among others. Pyrolysis furnaces with a controlled atmosphere

Recycling Technologies of Zn–C Batteries: Review and Challenges …

381

Fig. 1 Mechanical processing [6]

are also used in pre-treatment to remove mercury and organic material such as plastic and paper. These processes have selective volatilization of metals at high temperatures, followed by their condensation [7]. To obtain zinc and manganese, an oxidation or a smelting furnace is used, where metals are obtained with high purity after certain treatments [8]. The pyrometallurgical route does not require the dismantling stage and is a relatively simple process. Also, through this route, zinc and mercury can be completely re-generated [6]. The disadvantages of this process are that high energy consumption is required and the emissions of some gases such as dioxins, chlorinated compounds, and dusts that require other coupled systems are required to avoid further contamination [9]. There are two possible pyrometallurgical treatments: secondary metallurgy processes that use batteries as raw material and processes created specifically for batteries (such as pyrolysis, reduction, and incineration) [1]. • Hydrometallurgical: these are processes connected to leaching steps in acidic or alkaline media followed by purification, this in order to dissolve the metal fractions, so the metal solutions that can be used in the chemical industry can be recovered [1]. • Hydrometallurgical processes are the most studied to recover metals from battery waste in an efficient way. It is chosen as an extraction process and also as an environmental control, since the extracted metal will prevent the production of waste. Hydrometallurgical processes consist of the pre-treatment steps mentioned above,

382

N. Cardona-Vivas et al.

followed by leaching and separation of metals. Compared to pyrometallurgical processes, hydrometallurgical agents have the advantage of being less expensive and cleaner. Some pre-treatment steps are used to improve dissolution rates of the aqueous phase metal [7]. The procedure is long, and the use of different chemicals is required. Wastewater from hydrometallurgical processes can also be a hazardous waste, therefore requiring a separate disposal process. Hydrometallurgical processes have been the most studied for the recycling of Zn–C batteries, for this reason; there are patented processes since 1984 where zinc oxides and manganese oxides are obtained from battery residues [9].

Available Technologies for Battery Recycling See Table 2. Table 2 Current technologies of batteries recycling Name

Process type

Recover materials

Reference

BATREC

Pyrometallurgical

Ferromanganese, Hg, and Zn

[10]

U.S. Patent No 5,242,482

Pyrometallurgical

Mercury

[11]

U.S. Patent No 6,009,817

Pyrometallurgical

Products for metallurgical processes

[12]

BATENUS

Hydrometallurgical

MnO2

[13]

PLACID

Hydrometallurgical

Mercury

[14]

RECYTEC

Hydrometallurgical

Zn, MnO2

[15, 16]

HYDROMETAL SPA

Hydrometallurgical

Pb-acid

[17]

REVABAT/

Hydrometallurgical

Mn and Zn oxides and salts

[18]

RECUPYL

Hydrometallurgical

Mn and Zn carbonate

[19]

Acid leaching using sulfuric acid

Hydrometallurgical

Metallic Zn and MnO2

[20]

U.S. Patent No 5,411,643

Hydrometallurgical

ZnO and MnO

[21]

Integrated process using hydrochloric acid

Hydrometallurgical

Metal mix for use in the steel industry

[22]

Neutral leaching with KOH

Hydrometallurgical

Zn and Mn oxides

[5]

Acidic-reductive leaching

Hydrometallurgical

Zn y Mn sulfates

[7]

REVATECH

Solvent extraction

Liquid–liquid extraction

Manganese

[23]

Chemical precipitation

Chemical precipitation

Zn, Mn

[7]

Recycling Technologies of Zn–C Batteries: Review and Challenges …

383

Characterization of Battery Waste Scanning electron microscopy was used to analyze the particle size in a JEOL JSM 6700R operated in high vacuum mode. Zn–C battery waste was ground to obtain powder of batteries. Then, samples were mounted on carbon tape and gold sputtered to obtain a thin gold film. X-ray diffraction (XRD) characterization was conducted in an X’Pert PRO diffractometer (Cu Kα radiation, λ = 1.5406 Å), using a 45 kV voltage and scanning with 2θ between 10 and 90º. Figure 2 shows the process followed in Colombia [25] after a mechanical process summarized in Fig. 2a. The typical materials that appeared in these primary batteries are presented in Fig. 2b. The recycling issue is focused in the graphite and oxides, as metals, plastic, and papers are recycled easily. Figure 3 summarizes SEM images for the powders from Fig. 2b: graphite, zinc oxide, and carbon dioxide. From Fig. 3 is obvious the microstructure is very irregular,

Fig. 2 Zn–C battery grinding process implemented in Colombia [25]

Fig. 3 a SEM of battery waste 500X. b SEM of battery waste 2000X

384

N. Cardona-Vivas et al.

Fig. 4 XRD for the primary battery waste from Colombia used in this research

amorphous grains, and the particle size ranges from nano- to microparticles (about 200 nm to 50 μm). A more detailed analysis of these powders has been conducted via X-ray diffraction (XRD), as shown in Fig. 4. Crystalline phases such as zincite, franklinite, magnetite, and ramsdelita have been identified.

Circular Economy: Challenges and Advances in Its Implementation with Zn–C Batteries With the growth of the economy and industrial development, the circular economy (CE) has taken a special interest today due to the need to close production cycles, turning products that are at the end of their useful life, into resources to produce some new parts. The CE approach refers to a system that focuses on being restorative and generative. The system works in such a way that the value of the products, materials, and resources is maintained, as long as possible, so it shows the amount of waste throughout the life cycle of the goods produced. It also allows the increase in resource efficiency and economic opportunities concerning the use of waste as raw material

Recycling Technologies of Zn–C Batteries: Review and Challenges …

385

for new goods [24]. For CE initiatives to work, there is a need for a strong relationship between government entities, universities with their research groups, and private enterprise. When these three actors are integrated, it is possible to generate successful initiatives not only economically, but produce an environmental impact. China has several success stories in CE reported in the literature, driven by the implementation of government policies, with important contribution to the development of the country’s economy [24]. In Colombia, battery waste recycling is just started to being implemented, see Fig. 2, only Zn–C batteries. Before grinding, the batteries are sorted manually to ensure the same type [25]. The legislation that exists regarding batteries is aimed at the producer and importer, but not the final consumer. One of the initiatives is a campaign supported by the private industry known as “Pilas con el Ambiente” created in 2010. The main objective of this is the collection of all types of used batteries, to avoid that their waste reaches sanitary landfills. Although there are post-consumer campaigns for batteries, not all people properly separate these wastes. The batteries used in the landfills generate a leachate, due to factors such as temperature and humidity that can produce a high environmental impact in the medium term. Battery recycling has a big challenge regarding the correct disposal of the waste. Also, one of the main problems is the mechanical process that must always be carried out, regardless of the route to be applied for the recovery of the different elements. After the classification and grinding of the waste, a new challenge of finding potential applications for this waste appears: the use to all the processed waste in useful products, from which the country is still far.

References 1. Bernardes AM, Espinosa DCR, Tenório JAS (2004) Recycling of batteries: a review of current processes and technologies. J Power Sources 130(1–2):291–298 2. Linden D, Reddy TB (2013) Handbook of batteries. 33(04) 3. GIR (Global Info Research) (2018) Global Zinc-carbon battery market 2019 by manufacturers, regions, type and application, forecast to 2024 4. Martínez Ayala CA (2017) Propuesta Metodológica Para la Recuperación de las Pilas Alcalinas y Zinc-Carbono. UNIVERSIDAD PEDAGÓGICA Y TECNOLÓGICA DE COLOMBIA 5. Tanong K, Coudert L, Mercier G, Blais JF (2016) Recovery of metals from a mixture of various spent batteries by a hydrometallurgical process. J Environ Manage 181:95–107 6. Nan J, Han D, Cui M, Yang M, Pan L (2006) Recycling spent zinc manganese dioxide batteries through synthesizing Zn–Mn ferrite magnetic materials. J Hazard Mater 133(1–3):257–261 7. De Michelis I, Ferella F, Karakaya E, Beolchini F, Vegliò F (2007) Recovery of zinc and manganese from alkaline and zinc–carbon spent batteries. J Power Sources 172(2):975–983 8. Belardi G, Lavecchia R, Medici F, Piga L (2012) Thermal treatment for recovery of manganese and zinc from zinc–carbon and alkaline spent batteries. Waste Manag 32(10):1945–1951 9. Sayilgan E et al (2009) A review of technologies for the recovery of metals from spent alkaline and zinc-carbon batteries. Hydrometallurgy 97(3–4):158–166 10. Zenger T, Krebs A, Huibert J, Van Deutekom H (2003) Method of and apparatus for dismantling and storage of objects comprising alkali metal, such as alkali metal containing batteries. EP1333523B1

386

N. Cardona-Vivas et al.

11. Cangini G, Figari L, Moglie L, Pescetelli A (1992) Process for treating spent batteries. US 5,242,482 12. Hanulik J (1995) Process for the recycling of batteries, especially dry batteries. US 6,009,817 13. Fröhlich S, Sewing D (1995) The BATENUS process for recycling mixed battery waste. J Power Sources 57(1–2):27–30 14. Diaz G, Andrews D (1996) Placid—a clean process for recycling lead from batteries. Journal Miner Met Mater Soc (JOM) 48(1):29–31 15. Jordi H (1995) A financing system for battery recycling in Switzerland. J Power Sources 57(1–2):51–53 16. Nguyen TT (1990) Process for the simultaneous recovery of manganese dioxide and zinc. US4992149A 17. Ducati U (1981) Hydrometallurgical method for recovering metal materials from spent leadacid storage batteries. US4460442A 18. Serstevens A (2000) Method for recycling and treating of salt and alkaline batteries. EP1148571B1 19. Poinsignon CJL, Tedjar F (1993) Method for electrolytical processing of used batteries. EP0620607B1 20. Toro L, Veglio’ F, Beolchini F, Pagnanelli F, Zanetti M, Furlani G (2004) Process and plant for the treatment of run-down batteries. EP1684369B1 21. Cawlfield DW, Ward LR (1994) Integrated process of using chloric acid to separate zinc oxide and manganese oxide. US 5,411,643 22. Elliott KW (1994) Process for battery recycling. US 5,456,992 23. Martin D, Garcia MA, Diaz G, Falgueras J (2001) A new zinc solvent recovery application: spent domestic batteries treatment plant. In: Proceedings of the international solvent recovery conference, vol 1, pp 201–206 24. Ghisellini P, Cialani C, Ulgiati S (2016) A review on circular economy: the expected transition to a balanced interplay of environmental and economic systems. J Clean Prod 114:11–32 25. Colorado HA, Colorado SA (2016) Portland cement with battery waste contents. In: REWAS 2016: towards materials resource sustainability

Selective Recovery of Lithium from Ternary Spent Lithium-Ion Batteries Using Sulfate Roasting-Water Leaching Process Chang Di, Chen Yongming, Xi Yan, Chang Cong, Jie Yafei and Hu Fang

Abstract A novel sulfate roasting-water leaching method was proposed to selectively extract lithium from ternary spent lithium-ion batteries (LIBs). Ternary spent LIBs were first pretreated by discharge, thermal treatment, crushing and sieving to separate active powder from Al foil and Cu foil. The mixture of active powder and Na2 SO4 was roasted, and final leaching of the roasted product was done at 80 °C for 120 min with hot water. Detailed operating parameters were systematically investigated. The results showed that more than 85% of Li was selectively leached, while efficiencies of Ni, Co and Mn were