Photovoltaic Manufacturing E-Book

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Serie: Solar Cell Manufacturing
• Sprache: Englisch

Seitenzahl: 326

Veröffentlichungsjahr: 2021

Beschreibung

PHOTOVOLTAIC MANUFACTURING

This book covers the state-of-the-art and the fundamentals of silicon wafer solar cells manufacturing, written by world-class researchers and experts in the field.

High quality and economic photovoltaic manufacturing is central to realizing reliable photovoltaic power supplies at reasonable cost. While photovoltaic silicon wafer manufacturing is at a mature, industrial and mass production stage, knowing and applying the fundamentals in solar manufacturing is essential to anyone working in this field.

This is the first book on photovoltaic wet processing for silicon wafers, both mono- and multi-crystalline. The comprehensive book provides information for process, equipment, and device engineers and researchers in the solar manufacturing field. The authors of the chapters are world-class researchers and experts in their field of endeavor. The fundamentals of wet processing chemistry are introduced, covering etching, texturing, cleaning and metrology. New developments, innovative approaches, as well as current challenges are presented. Benefits of reading the book include:

• The book includes a detailed discussion of the important new development of black silicon, which is considered to have started a new wave in photovoltaics and become the new standard while substantially lowering the cost.
• Photovoltaics are central to any country’s “New Green Deal” and this book shows how to manufacture competitively.
• The book’s central goal is to show photovoltaic manufacturing can be done with enhanced quality and lowering costs.

Audience

Engineers, chemists, physicists, process technologists, in both academia and industry, that work with photovoltaics and their manufacture.

Leseprobe

Table of Contents

Cover

Title Page

Copyright

Preface

1 Metal-Assisted Chemical Etching of Silicon: Origin, Mechanism, and Black Silicon Solar Cell Applications

1.1 Introduction

1.2 History and Mechanism of Metal-Assisted Chemical Etching of Silicon

1.3 Fabrication and Optical Properties of Black Silicon by MacEtch of Silicon

1.4 Photovoltaic Solar Cell Applications of MacEtch Black Silicon

1.5 Concluding Remarks

Acknowledgements

References

2 Alkaline Texturing

2.1 Introduction to Alkaline Texturing

2.2 Pyramid Formation

2.3 Chemical Mixtures Used in the Alkaline Texturing

2.4 Mechanisms of Alkaline Texturing, Important Parameters Involved in Alkaline Texturing

2.5 Surface Conditioning Prior to Alkaline Texturing

2.6 Problems Associated to Alkaline Texturing

References

3 Advanced Texturing

3.1 Introduction to Advanced Texturing

3.2 History and Definition of Metal-Assisted Chemical Etching

3.3 Mechanisms of Metal-Assisted Chemical Etching

3.4 Methods of Metal-Assisted Chemical Etching

3.5 Copper-Assisted Chemical Etching

3.6 Conclusion

References

4 Wet Chemical Cleaning for Industrial Application

4.1 Introduction

4.2 Status of Production Technology in Solar Cell Manufacturing

4.3 Wet Chemical Process Technology

4.4 Contamination Management

4.5 Cost Considerations

4.6 Conclusion

Acknowledgments

References

5 Analytical Techniques for Wet Processing

5.1 Introduction

5.2 Metal Analysis by ICP-MS

5.3 Determination of Organic Contaminations

Acknowledgments

References

Index

Also of Interest

End User License Agreement

List of Tables

Chapter 2

Table 2.1. Micrographs of texturing for different concentrations (100%, 50%, and...

Table 2.2. Requirements and advantages of the candidates to substitute IPA.

Chapter 4

Table 4.1. Detection and quantification limits calculated according to [66] of t...

Table 4.2. DOE for the cleaning of textured wafers (slurry/diamond wire sawed).

Chapter 5

Table 5.1. Semi-quantitative results of a TD-GC-MS analysis of organic surface c...

Guide

Cover

Table of Contents

Title Page

Copyright

Preface

Begin Reading

Index

Also of Interest

End User License Agreement

Pages

v

ii

iii

iv

ix

x

xi

xii

xiii

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

185

186

187

Scrivener Publishing

100 Cummings Center, Suite 541J

Beverly, MA 01915-6106

Publishers at Scrivener

Martin Scrivener ([email protected])

Phillip Carmical ([email protected])

Photovoltaic Manufacturing

Etching, Texturing, and Cleaning

Edited by

Monika Freunek Müller

This edition first published 2021 by John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA and Scrivener Publishing LLC, 100 Cummings Center, Suite 541J, Beverly, MA 01915, USA

© 2021 Scrivener Publishing LLC

For more information about Scrivener publications please visit www.scrivenerpublishing.com.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.

Wiley Global Headquarters

111 River Street, Hoboken, NJ 07030, USA

For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.

Limit of Liability/Disclaimer of Warranty

While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials, or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read.

Library of Congress Cataloging-in-Publication Data

ISBN 978-1-119-24189-8

Cover image: Top figures: RENA Technologies GmbH – BatchTex machine for solar wafer texturing Bottom row left: Technological Institute of Microelectronics (TiM), University of the Basque Country (UPV/EHU), Bilbao, Spain

Bottom row right: Fraunhofer Institute for Microstructure of Materials and Systems IMWS, Research Unit Center for Silicon Photovoltaics CSP, Halle (Saale), Germany

Cover design by Russell Richardson

Set in size of 11pt and Minion Pro by Manila Typesetting Company, Makati, Philippines

Printed in the USA

10 9 8 7 6 5 4 3 2 1

Preface

The last two decades were groundbreaking for photovoltaic (PV) technology. Countless researchers, engineers, technicians, politicians, and individuals all over the world contributed with their work and enthusiasm to the progress of this field. In this time, silicon PV cells increased their efficiency to 26.1% [1], being close to their theoretical limit for real cells of 29.8% [2]. PV technologies such as multijunction solar cells achieved a maximum of 39.2% efficiency in nonconcentrated applications [1], and new emerging technologies such as perovskites evolved. Figures 1 and 2 visualize the impressive progress in photovoltaics, depicting the best research cell efficiencies (Figure 1) and the champion module efficiencies (Figure 2). Both figures start with a few technologies, remarkable achievements, and, especially in the case of modules, a somewhat steady progress. The first cell type ever recorded in these data, an a:Si:H cell, evolved from 2.4% efficiency in 1976 to 14.1%. However, shortly around the year 2000, 10 new photovoltaic technologies evolved, increasing the record PV efficiency from 31.9% in 2000 by more than 50% to 47.1%. Not only were 40% of the technologies of today developed during the last two decades but also their efficiency trends are improving much steeper than ever before in the PV history.

The same is true for the record PV module efficiencies depicted in Figure 2. The efficiencies demonstrated with small-scale modules in the year 2000 increases with some module technologies doubling or even tripling their module record efficiencies within a decade and less. This is especially remarkable as some the technologies were scaled up to large module productions. Again, more than 40% of today’s module technologies depicted in Figure 2 were not evolved at module or even research scale just 20 years ago.

The global solar module production reached the Gigawatt range soon after 2000 and ramped up to 165 GW in 2020 [3].

The use of PV technology has changed from test and research sites and fist operational installations to being an essential part of national energy strategies worldwide. In many regions, PV is part of the landscape in both rural and urban areas. The levelized cost of energy (LCOE) is a common measure to compare the average net cost of different electric energy production technologies over their lifetime. The LCOE of PV per MWh has taken a breathtaking journey from about 400 USD/W in 2010 to 50 USD/MWh and lower globally with some projects reaching values as low as 23 USD/MWh [4, 5]. This trend is expected to continue with a predicted LCOE of 20 USD/MWh for 2030.

Figure 1 Conversion efficiencies of best research solar cells worldwide from 1976 through 2020 for various photovoltaic technologies. Efficiencies are determined by certified agencies/laboratories. Image by Nikos Kopidakis, National Renewable Energy Laboratory (NREL), Golden, CO, 2021, under public domain.

Figure 2 NREL chart of the highest confirmed conversion efficiencies for champion modules for a range of photovoltaic technologies, plotted from 1988 to the present. Image by Nikos Kopidakis, National Renewable Energy Laboratory (NREL), Golden, CO, 2021, under public domain.

Despite this impressive progress, the processing of PV still is far from having reached its limits, and new challenges have to be addressed. Emerging developments, such as black silicon, provide a huge potential to make PV even more competitive in the field of energy conversion. Production efficiency requires a minimization of material and process losses, reproducible results, and economic scaling of the technology. The long-term nature of most PV applications and their large-scale implementation increases the importance of recycling. Ideally, this is included as part of the processing and manufacturing strategy.

The processing of PV today follows well-established standards, but as anyone involved knows, the detailed result will be highly dependent on the local machines and processing steps. Any difference in the settings might make a critical difference in the PV product performance and might distinguish the market leader from its competitors. This book introduces the readers to the theory and practical aspects of solar processing.

Metal-assisted chemical etching (MacEtch) for black silicon (b-Si) is expected to be the leading solar manufacturing technology in the future. Chapter 1 introduces this micro-/nanofabrication approach as one of the most promising prospects to further reduce the costs of photovoltaic devices while increasing their efficiency. The origin of MacEtch and the underlying mechanism are explained with a special focus on b-Si. The history, the state of the art, and an outlook toward the large-scale deployment in silicon photovoltaic industry are given.

Chapter 2 introduces the reader to alkaline texturing for the reduction of optical losses in monocrystalline silicon solar cells. The underlying process and the most important factors, parameters, and issues are explained. In addition, the texturing process is located in the whole manufacturing process of the solar cell, highlighting the importance of the previous steps for a high-quality result.

Chapter 3 provides a detailed introduction to advanced texturing with metal-assisted chemical etching in silicon solar wafers in general. The underlying electrochemical mechanisms are explained. Common methods, typical process steps, and structure characteristics obtained by metal-assisted chemical etching methods are introduced. Examples of the characteristics of topography and anti-reflection of the structures obtained using different metal catalysts and different etchant ratio are discussed.

Wet chemical cleaning of wafer surfaces and the most common cleaning technologies are outlined and discussed regarding their potential in the solar manufacturing process in Chapter 4. The reader is introduced to types and impact of contamination and to the concept of “contamination management.” Examples for this innovative approach are given. The chapter closes with an economic perspective on the topic.

Reliable quality control, reproducibility, and the development of processing technologies all rely on analytics. Chapter 5 covers impurity analytics for the manufacturing of photovoltaic solar cells. With a special focus on the chemical analysis of silicon wafer surfaces, a detailed description of the analysis of trace metals is given. Current developments in analytical techniques for organic contamination are reviewed, and an overview on recent analytical techniques with application examples is provided.

This book is a comprehensive review on the most important steps in processing a high-quality solar wafer while keeping track of economic key values. The essential knowledge is explained by recognized experts in their field of endeavor. Outlooks on future topics are given, and recent challenges and innovations are presented. This book provides you with an efficient and solid start to this important field of photovoltaics. The technological limits of photovoltaic are still to be reached—may this compilation help you in exploring them.

The editor would like to thank Karen Reinhardt for the initial idea and the beginning of this project. I also thank sincerely all the authors for their willingness to share their expertise, their efforts to make their knowledge understandable for a larger audience, and for staying patient and focused during the publishing process of a book in the midst of a pandemic.

Monika Freunek

Lighthouse Science Consulting and Technologies, Canada

June 2021

References

1. F. Haase et al., Laser contact openings for local poly-Si-metal contacts enabling 26.1%-efficient POLO-IBC solar cells, Sol. En. Mat. Sol. Cells, 186, November 2018, 184–193.

2. T. Tiedje et al., Limiting Efficiency of Silicon Solar Cells, IEEE Trans. El. Dev., 31(5), May 194, 711–716.

3. IEA, https://www.iea.org/data-and-statistics/charts/solar-pv-module-manufacturingand-demand-2014-2020, accessed 21-02-2021.

4. https://en.wikipedia.org/wiki/Cost_of_electricity_by_source, accessed 21-02-2021.

5. https://www.pv-magazine.com/2020/04/30/lcoe-from-large-scale-pv-fell-4-to50-per-megawatt-hour-in-six-months/, accessed 21-02-2021.

1Metal-Assisted Chemical Etching of Silicon: Origin, Mechanism, and Black Silicon Solar Cell Applications

Chenliang Huo, Jiang Wang, Haoxin Fu and Kui-Qing Peng*

Beijing Key Laboratory of Energy Conversion and Storage Materials, Department of Physics, Beijing Normal University, Beijing, China

Abstract

Metal-assisted chemical etching (MacEtch) of silicon in hydrofluoric acid (HF) aqueous solutions is a widely used top-down approach for silicon micro/nanofabrication due to its cost-effectiveness, simplicity, versatility, and scalability. This method has recently emerged as a powerful surface micro/nanostructuring technique for low-cost and scalable production of black silicon (b-Si) with excellent light trapping properties, which might lead to both efficiency increase and cost reduction of solar cells. This review of MacEtch of silicon provides a critical description of its origin and the understanding of underlying mechanism highlights the story of MacEtch b-Si from initial discovery, through engineering improvements, toward the large-scale deployment in silicon photovoltaic industry.

Keywords: Metal-assisted chemical etching (MacEtch), solar cells, black silicon, photovoltaic

1.1 Introduction

Widespread deployment of solar photovoltaics (PVs) is critical to meeting the world’s growing energy demand, tackling fossil fuel shortage and mitigating climate change in future, but solar PV remains uncompetitive (expensive and unreliable) relative to other technologies such as current fossil fuels-based electricity generation [1–3]. It is acknowledged that lowcost high-efficiency PV cells could capture more of the sun’s energy and hence make sunlight-generated electricity economically competitive with fossil fuels-generated electricity [2–7]. Over the last decades, whereas crystalline silicon (c-Si) has been the most dominant semiconducting material for commercial PV cells due to its low-cost, earth-abundance and reliability, poor infrared absorption resulting from its indirect band gap, as well as relatively high reflectivity resulting from conventional surface texturing technologies greatly hurts the cell efficiency and poses a major challenge to the large-scale deployment of silicon PV modules [8, 9]. Hence, new ideas that strive for efficient utilization of much sunlight were under development [10–21]. One especially promising highefficiency silicon solar PV technology is the black silicon (b-Si) solar cell, which is based on the c-Si wafer with micro/nanostructured surface that appears black to human’s naked eye since it could capture sunlight across a broad range of wavelengths and angles of incidence very efficiently [2, 15–22]. In addition, excellent light trapping capability of b-Si allows reduction of wafer thickness and even no specific application of antireflection coating (ARC), thereby is an ideal material candidate for low-cost high-efficiency PV applications.

Low-cost and scalable production of b-Si is critical for industrial silicon solar cells but remains a significant challenge. Over the last decades, reactive ion etching [23–27], electrochemical/electroless etching, [28–30], femtosecond laser microstructing [15, 32–34], and metal-assisted chemical etching [16, 17, 35, 36] have been developed to texture silicon surfaces on the microand nano-scale for b-Si production. Currently, one of the biggest trends in b-Si manufacturing is the adoption of metalassisted chemical etching technology as it has many advantages: it is low-cost, rapid, no requirement for expensive facilities, and scalable for industrial application. Recent advances in the scalable production of b-Si by metal-assisted chemical etching have pushed forward its practical applications in high efficiency silicon solar cells. The success of metalassisted chemical etching b-Si technology is closely linked to its simplicity and well compatibility with existing industrial silicon solar cell production facilities. This review covers the history, mechanism, methods and recent achievements on b-Si by metal-assisted chemical etching for PV application. This article provides the reader with critical understanding of metal-assisted chemical etching, the b-Si fabrication processes, and the roadmap for large-scale deployment of b-Si solar cells.

1.2 History and Mechanism of Metal-Assisted Chemical Etching of Silicon

Metal-assisted chemical etching (MacEtch or MACE) of silicon generally refers to wet etching of silicon in the presence of noble metal particles or their film with openings, which are physically or chemically introduced onto silicon surface to enhance silicon etching at open-circuit potential (OCP) in oxidizing hydrofluoric acid (HF) aqueous solutions [35–40]. Within the last couple of years, MacEtch of silicon in HF aqueous solutions has aroused great interest and is presently one of the most popular topdown approach for micro/nanofabrication due to its cost-effectiveness, simplicity, versatility, and scalability [39–56]; it is also becoming increasingly important in various applications ranging from industrial solar cells, thermoelectric devices, batteries to medical biosensor and drug delivery [16, 17, 57–76]. Generally, for etching at OCP one can distinguish two mechanisms: electrochemical and chemical. The difference between both mechanisms is whether free holes are involved. The etching process at OCP involving free holes and depending on the redox potential generally is referred to as electroless etching. The MacEtch of silicon at OCP in oxidizing HF aqueous solutions is a noble metal catalyst-mediated electrochemical charge transfer process in which free holes are involved and spontaneously galvanic currents flow between local silicon anode and local metal cathode sites [37–40, 77, 78]. Therefore, the MacEtch of silicon is considered electrochemical reaction in nature. Besides MacEtch [38], the terms metal-assisted etching, metal-catalyzed chemical etching [35] or metal-catalyzed electroless etching (MCEE) [79–81] could be observed in literature. Among these terms, the acronym MCEE more clearly reveals the fundamental aspect of silicon etching electrochemistry in the presence of metal. Although we prefer MCEE or metal-catalyzed wet etching, considering the MacEtch has become common as Prof. Li suggested in the 2014 Spring Meeting of the Materials Research Society, we used it throughout this article.

1.2.1 History of Metal-Assisted Chemical Etching of Silicon

The discoveries and studies of metal-enhanced silicon etching actually began in the early 1990s [37, 82–90]. It is critically important to monitor metallic contamination on silicon wafer surfaces in order to achieve high-performance ultra-large-scale integration (ULSI) devices. Metallic contaminants such as Au, Cu, Ag, Pt, and Pd generally have fatal effects on device characteristics and must be suppressed to below 1010 atom/cm2. Therefore, the behaviors of metallic contaminants, especially noble metal ions and their metallic particles on silicon wafer surfaces in wet chemical cleaning solutions have been extensively investigated to understand the underlying mechanism of metal deposition and then remove them from the silicon surface efficiently. Ohmi and other researchers found that noble metal cations such as cupric ions deposit on silicon surface in the metallic state through charge transfer from silicon to metal cations at the silicon/solution interface, simultaneously induces silicon oxidation nearby metallic deposits and then induces surface pits and microroughness in dilute HF (DHF) solution cleaning [37, 83–89]. The noble metal cations reduction takes places by either withdrawing conduction band electrons of silicon or injecting holes into silicon valence band holes. Moreover, Morinaga et al. found that silicon surface becomes rougher when an oxidizing DHF-H2O2 cleaning is used remove noble metallic particles [83]. Figures 1.1a–c show the scanning electron microscope (SEM) images of silicon surface with ultrafine Au particles after DHF-H2O2 cleaning. Porous silicon with small holes were clearly observed on the silicon surface. No obvious differences could be observed between the initial state and after DHF-H2O2 cleaning for silicon surface without particles. Morinaga et al. suggested that Au features higher electronegativity than silicon and hence attracts electrons from silicon to induce silicon oxidation nearby it. Since the etch rate of SiO2 by HF is always higher than its formation rate by H2O2, local excessive silicon oxidation and dissolution occur, resulting in silicon surface micro roughness or pitting due to irregular etching, as schematically shown in Figure 1.1d.

Figure 1.1 SEM images of the silicon wafers with Au particles after DHF-H2O2 cleaning. The silicon surface became rougher after cleaning. (a) initial. (b) 1 min. (c) 5 min. (d) Schematics of surface microroughness mechanism caused by Au particles with DHF-H2. Reprinted with permission from H. Morinaga et al., J. Electrochem. Soc. 142, 966 (1995). Copyright 1995 The Electrochemical Society.

The pioneering fabrication of luminescent porous silicon by metal enhanced silicon etching was first demonstrated by Zhang and co-authors in 1993 [91]. They produced porous silicon layer on n-type silicon polished surface when the n-silicon back in contact with a noble gold metal was illuminated in a HF solution containing oxygen without an externally applied potential. In 1997, Dimova-Malinovska et al. reported the fabrication of porous silicon by etching an aluminium coated silicon substrate in HF-HNO3 aqueous solution [82]. They claimed that the incubation time necessary for the porous silicon formation was dramatically decreased due to the presence of the Al film. In 1999, inspired by the discovery of Zhang et al., Kelly and co-authors demonstrated that luminescent porous silicon layer can be made on p-type silicon surface in a similar way without illumination [92]. In their experimental setup as schematically shown in Figure 1.2a, a silicon wafer was short-circuited to an inert metal by evaporating inert Au/Cr film onto the back side of the wafer, thus a galvanic cell was formed. The resulting porous silicon layer with a thickness of 7 µm is shown in Figure 1.2b. They proposed that the porous silicon is achieved by galvanic etching due to the formation of metal/silicon galvanic cell in which the silicon acting as anode and metal as cathode in HF aqueous solution containing oxidants. The galvanic etch rate can be controlled by the metal/silicon area ratio and oxidant concentration of in HF solution.

Simultaneous platinum deposition and formation of a visible light-emitting porous silicon layer on silicon surface in fluoride solution containing platinum ions were reported by Gorostiza et al. in 1999 [93, 94]. They carried out electrochemical measurements and demonstrated that the hole injection from the platinum ions into the silicon results in the formation of porous silicon layer on silicon without requiring an electrochemical cell and voltage source. The SEM image depicted in Figure 1.2c shows the strongly etched silicon surface during Pt deposition in HF-K2PtCl6 solution. In 2000, Li and Bohn reported rapid preparation of light-emitting porous silicon by introducing metal nanoparticles (Au, Pt, or Au/Pd) to Si(100) surfaces prior to immersion in HF-H2O2 aqueous solution [38]. They demonstrated the metal coatings enhance silicon etching and result in a simple and effective way of producing porous silicon on the time scale of seconds. They termed the method “H2O2-metal-HF (HOME-HF) etching” in the main text and “metal-assisted chemical etching (MacEtch)” in the title. Figures 1.2d, e show the SEM images of Au-coated p+ silicon etched in HF-EtOHH2O2 aqueous solution for 30 seconds. Large interconnected pores are like the morphology observed on anodically etched porous silicon could be observed on the Au-coated areas. They found that strongly luminescent porous silicon can also be produced away from the metal-coated areas, implying lateral transport of charge carriers and chemical species during etching. Figure 1.2f shows the top-view SEM image of Pt-coated Si (100) surface after etching in HF/H2O2 for 30 s. A localized electrochemical process with the metal nanoparticle acting as a local cathode and silicon acting as an anode was proposed. Bohn and coauthors subsequently extended the MacEtch method to other semiconductors [95, 96].

Figure 1.2 (a) Schematical view of the experimental setup with an Au electrode on the back side of the silicon wafer. (b) SEM image showing a cross-section of a porous Si layer formed using a gold-backed silicon wafer in an air-saturated HF solution. Reprinted with permission from C.M.A. Ashruf et al., Sensor Actuat or A-Phys. 74 118 (1999). Copyright 1999 Elsevier. (c) SEM image of porous silicon on silicon surface produced after Pt deposition in HF-K2PtCl6 solution. Reprinted with permission from P. Gorostiza et al., J. Electrochem. Soc. 144, 4119 (1997). Copyright 1997 The Electrochemical Society. (d, e) SEM images of Au-coated area on p+-Si(100) and off the Au-coated area on p+-Si(100) after etching in HF/H2O2 for 30s. (f) SEM image of Pt-coated area on p−-Si(100) after etching in HF/H2O2 for 30s. Reproduced with permission from Li XL et al., Appl. Phys. Lett. 77, 2572 (2000). Copyright 2000 American Institute of Physics.

In 1960, Turner has proposed that the wet etching of silicon in HF aqueous solution containing oxidant such as nitric acid (HNO3) at OCP is an electrochemical process [97], in which silicon oxidation and dissolution takes place at local anode areas while the oxidizing agent is reduced at local cathode areas. Turner suggested that an etch pit will form at one site when it is anodic much more than it is cathodic while hillock will be produced at area that is cathodic more than it is anodic. However, the etching process is stochastic and non-preferential since any given area on silicon surface continually alternate between being anode and cathode. Inspired by the Turner’s idea on the galvanic etching of silicon in HF solutions, Peng put forward and elaborate on a silicon micro/nanofabrication idea where [39, 98], locking micro/nanoscale anode or cathode sites leads to selective etching of silicon without mask in oxidizing HF aqueous solutions. They predicted that large-area silicon micro/nanostructures such as high-aspect ratio nanowires and nanoholes could be produced on silicon surface if something makes the idea come true. After many efforts, large-area aligned silicon nanowire (SiNW) arrays were firstly prepared by single-step silver-catalyzed etching of silicon wafer in HF-AgNO3 aqueous solution in 2001, and they initially named the process selfassembling nanoelectrochemistry [39]. They ascribed the selective etching of silicon to the numerous microscopic localized silver/metal galvanic cells that are produced on silicon surface during etching in HF-AgNO3 solution. Their first proposed etching process for SiNW formation in 2002 is incorrect, but subsequently were corrected based on compelling experimental results. Figure 1.3 shows the representative SEM images of the silicon substrates after etching in in HF-AgNO3 aqueous solutions. Besides the silver dendritic film on the top of SiNW array, many silver nanoparticles can be observed at the interface between SiNW arrays and the intact Si substrate. The cross-sectional SEM observations confimed that the silver particles induced selective etching of silicon, as shown in Figure 1.3 [99]. Aligned SiNWs also could be produced on silicon surface after selective etching in HF-KAuCl4 aqueous solution [100, 101].

1.2.2 Mechanism of Metal-Assisted Chemical Etching of Silicon

So far, great efforts have been made to reveal the underlying mechanism of metal (e.g., Ag, Au, Pt, Pd, and Cu) assisted etching of silicon wafer in HF aqueous solutions containing strong oxidants (e.g., noble metal ions, nitrate, nitric acid, H2O2, and dissolved O2) [38–126]. It has been revealed that the behavior of MacEtch varies with the etchant compositions and doping types of silicon, silicon at the metal-silicon interface undergone fast etching and the morphologies of resulting silicon micro/nanostructures depend on the shape or pattern of metal catalysts on silicon surfaces. Despite the fact that the MacEtch continues to dazzle us with its promising applications and unusual behaviors, the underlying physics and chemistry of the etching system are still not well understood. For example, recent work clearly demonstrated that metal/silicon galvanic cell formation depends on the type and concentration of oxidants in HF aqueous solution [78]. Nonetheless, the mechanisms involving metal particles mediated charge transfer, metal particles catalyzed galvanic silicon etching, and the movement of metal particles into bulk silicon in unison have been widely accepted.

Figure 1.3 (a) SEM image of aligned SiNW array prepared by one-step etching in HF-AgNO3 solution. Reprinted with permission from K.Q. Peng et al., Adv. Mater. 14, 1164 (2002). Copyright 2002 Wiley-VCH. (b) SEM image showing aligned SiNW array on p-type (111) Si wafer etched in HF/AgNO3 solution for 5 min. (c, d) SEM images showing aligned SiNW array on p-type (111) Si wafer etched in HF/AgNO3 solution for 30 min. Reprinted with permission from K.Q. Peng et al., Chem. Eur. J. 12, 7942 (2006). Copyright 2006 Wiley-VCH.

Since the galvanic cell-based silicon micro/nanofabrication idea came true [39], Peng and coauthors have systematically investigated the characteristics of MacEtch of silicon in HF aqueous solution containing oxidants [35, 40, 57, 77–81, 98–102]. They found that metal particles or metal film with openings induce fast dissolution of silicon underneath them and move into bulk silicon in the same direction during etching, and the motion direction of metal particles may suddenly change in unison due to some unknown reasons in some cases; the catalytic activity of the metal influences the oxidant reduction rate and thereby the etching rate. They suggested that the electrically coupled metal-silicon spontaneously constructs a microscopic short-circuited galvanic cell in which metal acts as local cathode for oxidant reduction while the silicon beneath metal acts as local anode and is subjected to oxidation and dissolution in HF solution [77], as shown in Figure 1.4a. For the sake of simplicity, the most widely used MacEtch system Ag-Si-HF-H2O2 is discussed here. The sustained silicon oxidation/dissolution and in unison movement of metal particles inward bulk silicon eventually leads to the formation of silicon micro/nanostructures such as nanowire and nanoholes (Figures 1.4b, c).

On the basis of experimental results [78], silicon dissolution during MacEtch of silicon in HF solutions consisting of divalent and tetravalent dissolution processes was proposed, and the two half-cell reactions that include cathodic and anodic reactions are expressed in the following equations (Equations 1.1–1.4). Note that the additional cathodic reaction expected to occur on metal surface or silicon surface is the reduction of noble metal ions since their metallic form may dissolve in the oxidizing HF solution, while the strong oxidizing solution environment does not favor the cathodic reduction of hydrogen ions. Clearly, besides the highly localized silicon dissolution at the metal/silicon interface, the overall electrochemical reactions in MacEtch of silicon are identical to those in electrochemical etching or stain etching of silicon in HF aqueous solution containing strong oxidants [27, 82, 128].

Cathodic reduction of oxidant at metal surface:

(1.1)