Advances in Photovoltaics Part 4

by Weber, Eicke R.; Willeke, Gerhard P

Advances in Photovoltaics Part 4 Advances in Photovoltaics Part Four provides valuable information on the challenges faced during the transformation of our energy supply system to more efficient renewable energies The volume discusses the topic from a global perspective presenting the latest information on photovoltaics a cornerstone technology It covers all aspects of this important semiconductor technology reflecting on the tremendous and dynamic advances that have been made on this topic since 1975 when the first boo

Publisher : Academic Press

Author : Weber, Eicke R.; Willeke, Gerhard P

ISBN : 9780128010211

Year : 2015

Language: en

File Size : 9.67 MB

Category : Engineering Transportation

VOLUME NINETY TWO

SEMICONDUCTORS AND
SEMIMETALS
Advances in Photovoltaics: Part 4

SERIES EDITORS
EICKE R. WEBER
Director
Fraunhofer-Institut
f€
ur Solare Energiesysteme ISE
Vorsitzender, Fraunhofer-Allianz Energie
Heidenhofstr. 2, 79110
Freiburg, Germany

CHENNUPATI JAGADISH
Australian Laureate Fellow
and Distinguished Professor
Department of Electronic
Materials Engineering
Research School of Physics
and Engineering
Australian National University
Canberra, ACT 0200
Australia

VOLUME NINETY TWO

SEMICONDUCTORS AND
SEMIMETALS
Advances in Photovoltaics: Part 4
Edited by

GERHARD P. WILLEKE

Fraunhofer-Institut fu€r Solare
Heidenhofstr. 2, 79110
Freiburg, Germany

EICKE R. WEBER
Fraunhofer-Institut
fu€r Solare Energiesysteme ISE
Vorsitzender, Fraunhofer-Allianz Energie
Heidenhofstr. 2, 79110
Freiburg, Germany

AMSTERDAM • BOSTON • HEIDELBERG • LONDON
NEW YORK • OXFORD • PARIS • SAN DIEGO
SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier
225 Wyman Street, Waltham, MA 02451, USA
525 B Street, Suite 1800, San Diego, CA 92101-4495, USA
125 London Wall, London, EC2Y 5AS, UK
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK
First edition 2015
© 2015 Elsevier Inc. All rights reserved
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage and
retrieval system, without permission in writing from the publisher. Details on how to seek
permission, further information about the Publisher’s permissions policies and our
arrangements with organizations such as the Copyright Clearance Center and the Copyright
Licensing Agency, can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright by
the Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and
experience broaden our understanding, changes in research methods, professional practices,
or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described
herein. In using such information or methods they should be mindful of their own safety and
the safety of others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability for any injury and/or damage to persons or property as a matter of
products liability, negligence or otherwise, or from any use or operation of any methods,
products, instructions, or ideas contained in the material herein.
ISBN: 978-0-12-801021-1
ISSN: 0080-8784
For information on all Academic Press publications
visit our website at store.elsevier.com

CONTENTS
Contributors
Preface

vii
ix

1. Silicon Crystallization Technologies

1

Peter Dold
1. Silicon Feedstock
2. Fundamental Parameters for Silicon Crystallization
3. Crystallization Technologies
4. Summary and Final Remarks
References

2. Wafering of Silicon

1
12
19
54
56

63

€ller
Hans Joachim Mo
1. Introduction
2. Multiwire Sawing Technology
3. Basic Sawing Mechanisms
4. Alternative Wafering Technologies
References

3. Reliability Issues of CIGS-Based Thin Film Solar Cells

63
65
90
102
105

111

Thomas Walter
1. Reliability
2. Metastabilities
3. Partial Shading and Hotspots
4. Potential-Induced Degradation
5. Back Contact
References
Index
Contents of Volumes in this Series

111
115
117
132
138
148
151
155

v

This page intentionally left blank

CONTRIBUTORS
Peter Dold
Fraunhofer CSP, Halle, Germany. (ch1)
Hans Joachim M€
oller
Fraunhofer Technology Center for Semiconductor Materials, Freiberg, Germany. (ch2)
Thomas Walter
Faculty of Mechatronics and Medical Engineering, University of Applied Sciences Ulm,
Ulm, Germany. (ch3)

vii

This page intentionally left blank

PREFACE
The rapid transformation of our energy supply system to the more efficient
use of increasingly renewable energies is one of the biggest challenges and
opportunities of the present century. Harvesting solar energy by photovoltaics is considered to be a cornerstone technology for this truly global transformation process, and it is well on its way. The speed of progress is
illustrated by looking at some figures of the cumulative installed PV peak
power capacity. In Part 1 of this series of “Advances of Photovoltaics,”
published in 2012, the introduction mentioned 70 GWp installed at the
end of 2011. As we write this preface of Part 4 in the spring of 2015, 1%
of the world electricity generation is now already supplied by PV, and in
the coming months the global PV installation figure will have tripled
compared with 2011! But this is just the beginning of the thousands of
GWp that are likely to be installed in the decades to come.
Key for this extraordinary development was the rapid decrease of PV
prices and thus the cost of solar electricity. This was fueled by a rapid
technology development with soaring efficiencies at reduced production
cost, coupled with an effective market introduction policy, especially the
well-designed German feed-in tariff. Today, we can harvest solar electricity
even in Germany—with insolation comparable to Alaska!—for about
10 $ct/kWh, and in sun-rich areas for half of this amount, far below the cost,
e.g., electricity obtained from Diesel generators.
As already mentioned above, this book presents the fourth volume in the
ongoing series “Advances in Photovoltaics” within Semiconductors and
Semimetals. This series has been designed to provide a thorough overview
of the underlying physics, the important materials aspects, the prevailing and
future solar cell design issues, production technologies, as well as energy system integration and characterization issues. The present volume deals with
three important issues, of crystallizing silicon, the dominating PV material,
the ways of how to transform it into wafers for solar cells, as well as the issue
of reliability of CIGS-based thin film solar cells and modules. Following the
tradition of this series, all chapters are written by world-leading experts in
their respective field.
As we write this text, the German PV market is likely to collapse from a
7.5 GWp/a market as recently as 2012 to a 1 GWp/a level in 2015, a market
size that we last had in 2007. Fortunately, other markets in China, Japan, and
ix

x

Preface

the USA are now taking over by currently developing into 10 GWp per year
and more markets.
The solar PV revolution has started irreversibly, it is now fueled by
economics in addition to the concern for reducing climate gas emissions,
and it takes rapid foothold beyond Europe in Asia and the Americas, the
other parts of our planet will follow in a few year’s time!
GERHARD P. WILLEKE AND EICKE R. WEBER
Fraunhofer ISE, Freiburg, Germany

CHAPTER ONE

Silicon Crystallization
Technologies
Peter Dold1
Fraunhofer CSP, Halle, Germany
1
Corresponding author: e-mail address: [email protected]

Contents
1. Silicon Feedstock
1.1 Polysilicon: The Base Material for over 90% of All Solar Cells
1.2 The Chemical Path
1.3 Fluidized Bed Reactor
1.4 The Metallurgical Path: UMG-Si
1.5 Different Poly for Different Crystallization Techniques
2. Fundamental Parameters for Silicon Crystallization
2.1 Material Properties, Material Utilization, and Chemical Reactivity
2.2 Numerical Simulation
3. Crystallization Technologies
3.1 Pulling from the Melt: The Cz Technique
3.2 Directional Solidification: Growth of Multicrystalline Silicon
3.3 FZ Growth
4. Summary and Final Remarks
References

1
1
3
6
9
11
12
12
18
19
20
36
45
54
56

1. SILICON FEEDSTOCK
1.1 Polysilicon: The Base Material for over 90% of All
Solar Cells
The roller coaster ride of the polysilicon industry during the last 10 years was
quite extraordinary—even compared with the ups and downs of the semiconductor business over the last half century. The golden age of polysilicon
in the years 2007–2010, when companies could make billions of dollars if
they were able to deliver polysilicon at all, was followed by the severe crush
in the years 2011–2012, when most of the newcomers marched into bankruptcy and disappeared. And, even some of the old ones had to fight heavily
Semiconductors and Semimetals, Volume 92
ISSN 0080-8784
http://dx.doi.org/10.1016/bs.semsem.2015.04.001

#

2015 Elsevier Inc.
All rights reserved.

1

2

Peter Dold

to survive. During the golden years, spot market prices had reached highs of
200–300 or even 400 US$/kg polysilicon, simply because the market was
swept and the order books of the cell and module manufacturers were full.
The polysilicon industry was not prepared for such a fast ramp-up, investment is high,1 and equipment could not readily be ordered. The longestablished companies either have an exclusive partnership with a specific
equipment manufacturer, or they make the equipment in-house. Production capacity could not easily be ramped up, but once the train was running,
it also could not be stopped so easily and could not be adjusted to the then
changed market situation, partly because typical polysilicon projects take
several years from the financing phase all the way up to full production,
and partly because the players did not want to believe that the silicon
bonanza was over. The huge shortage was followed by a tremendous over
supply with spot market prices as low as 14–16 US$/kg in 2013—which was
below the actual production costs. Today, spot market prices leveled off
around 17–18 US$/kg and no significant changes are expected for the near
future.
As a consequence, all (or at least as good as all) of the new and innovative
approaches for polysilicon refinement, for upgrading metallurgical silicon
(an excellent review was given by Heuer, 2013), or for alternative production methods (compare Bernreuter and Haugwitz, 2010) could not find a
market share and disappeared again. The traditional Chemical Vapor Deposition (CVD)-based Siemens process (Fabry and Hesse, 2012), probably not
the most sophisticated technology for solar-grade-silicon production—but
for sure the most matured technique, was the match winner. A good overview of the market situation and an in-depth analysis of the trends are given
by Bernreuter every first or second year (Bernreuter, 2014).
Basically, two main routes might be distinguished for the refinement of
polysilicon: (I) the chemical path: bringing silicon into the gas phase and
purifying it by distillation, followed by thermal pyrolysis of the gaseous species; and (II) the metallurgical path, where impurities are removed from silicon by mixing it with another metal or with a slag, then let the impurities
segregate into the second phase, separate the different phases somehow
mechanically, and clean the surface of the silicon crystallites by chemical
etching.
1

Back in 2008, a polysilicon plant with a capacity of 10,000 t/a required an investment of at least 1 billion US$. Today, it might be something in the range of 400–600 M$, depending on the location.

Silicon Crystallization Technologies

3

1.2 The Chemical Path
The Siemens process (or modified Siemens process, as many manufacturers
like to call their variation) allows to produce ultrapure polysilicon, with
metallic bulk impurity levels as low as a few tens of ppt (parts per trillion)
or an equivalent of 10–11N. Electrically active elements (donors, acceptors)
are in the ppt range and only carbon and oxygen show up in higher concentrations, where lower single-digit parts per million levels are found. For
semiconductor applications, there is no alternative so far to the polysilicon
produced by the Siemens process.
The Siemens process itself goes back to a patent in the late 1950s filed by
the German electronics company Siemens (Reuschel, 1963; Schweickert
et al., 1961), which stepped out of the polysilicon business long ago. It
can be described by the following process steps:
I. Milling of the metallurgical silicon (purity: 98–99%) into millimeter/
submillimeter particles.
II. Reaction between the fine silicon particles and gaseous HCl at temperatures around 300–350 °C in a fluidized-bed reactor (FBR). The reactor might be heated from the outside, but the chemical reaction is also
strongly exothermic. Mainly copper is used as a catalyst. The main
product is TCS (trichlorosilane, SiHCl3).
III. Fractional distillation of the TCS and the by-products, like metal chlorides, boron, and phosphorus components, and so on. The result will
be ultrapure TCS.
IV. Pyrolytic decomposition of TCS in a bell-jar reactor (Fig. 1) at increased
pressure (normally 6 bar) and temperatures of 1000–1150 °C (Fig. 2).
High-purity polysilicon will be obtained (Fig. 3).
Steps I–III are relatively straightforward, although the installation of the
hardware reaches easily the size and complexity of a huge chemical plant
for typical production capacities of around 10,000 t/a. Step IV is more
difficult:
– The high temperature required for the silicon deposition is rather energy
intensive. The silicon rods on which the deposition takes place are
directly heated by an electrical current.
– Deposition rates on these U-shaped rods are on the order of 0.5–1 mm/h
(layer growth); beyond this rate, the rod morphology becomes unstable
and so-called “popcorn” or “broccoli” growth takes place.
– Only part of the TCS decomposes to silicon, and a significant part reacts
with the HCl formed during the deposition to STC (silicon tetrachloride,

4

Peter Dold

Figure 1 Schematic drawing of a Siemens bell-jar reactor for polysilicon deposition
from the gas phase. The U-shaped silicon rods are heated up to a temperature of
1000–1150 °C by direct current. The process gas enters and leaves the reactor chamber
through the base plate. By courtesy of Wacker Chemie AG.

Figure 2 Silicon deposition from TCS in a research reactor. Left: beginning of the deposition, right: after 30 h process time. In particular, in the elbow area, current and temperature distribution might be nonuniform.

SiCl4). Decomposition of STC is too low at the typical rod temperatures
in the bell-jar; therefore, it has to be removed from the reactor and has to
be back-converted into TCS.
In former times, back-conversion of STC to TCS was carried out mainly
in thermal STC converters (Paetzold et al., 2007; Sirtl et al., 1974), and the
process is also referred as “hydrogenation.” At high temperature in a hot
carbon rod reactor (>1200 °C), STC reacts with hydrogen back to TCS
(and other by-products), an another energy-intensive process step. Nowadays,

© 2018-2019 uberlabel.com. All rights reserved