solar cells

Advanced Rear-Side Contact Schemes on i-PERC (industrial Passivated Emitter and Rear Cell) solar cells Angel Uruena Chair: Prof. Herman Neuckermans Promotors: Prof. Robert P. Mertens Prof. Jef Poortmans Members of the Examination Committee:

Dissertation presented in

Prof. Johan Driesen

partial fulfilment of the

Prof. Andre Stesmans

requirements for the degree of

Prof. Cor Claeys

Doctor in Engineering

Dr. Daniel Biro Dr. Joachim John In collaboration with

September 2013

© 2013 KU Leuven, Science, Engineering & Technology v.u. Leen Cuypers, Arenberg Doctoraatsschool, W. de Croylaan 6, 3001 Heverlee Alle rechten voorbehouden. Niets uit deze uitgave mag worden vermenigvuldigd en/of openbaar gemaakt worden door middel van druk, fotokopie, microfilm, elektronisch of op welke andere wijze ook zonder voorafgaandelijke schriftelijke toestemming van de uitgever. All rights reserved. No part of the publication may be reproduced in any form by print, photoprint, microfilm, electronic or any other means without written permission from the publisher. ISBN 978-94-6018-711-7 WD/2013/7515/97

A mis padres, que siempre me lo han dado todo

To my parents, who have always given me everything

ABSTRACT

For many years, the photovoltaic industry has been using Al to form the back contact of crystalline silicon solar cells. The objective of this PhD is to reach a better understanding of the contact formation at the rear side of the Passivated Emitter and Rear Cell (PERC) type solar cell. This type of cell features at the rear side a stack of dielectrics as a passivation layer which is pre-opened in order to allow the subsequent metallization step. The way this dielectric stack is opened, by laser ablation, the deposition of the metal, with various techniques, and the high temperature step required for the formation of the Al-Si alloy which serves as a contact, will be investigated in this work. It will be demonstrated that Si dissolution in Al plays an important role in the formation of the contacts, giving on one hand the needed dopants to create an effective Back Surface Field (BSF) layer, and on the other hand forming a deep pyramidal shape in the contact region which might affect the reflection of the light inside in the cell, and increase the surface recombination velocity of the device. A novel method to characterize the local Al-Si alloy formation is introduced, allowing the in-situ observation of the high temperature step for the contact formation. By doing this, details of the process which cannot be taken into account once the device is finished, are studied and explained through hypotheses involving the phase diagram between both elements. The performance at cell level of the different combination of parameters affecting the formation of the BSF region has been evaluated. For this,

iii

iv

ABSTRACT

changes in contact pitch, firing temperature and profile, Al thickness and composition are studied for the PERC cells. At the end, the rear reflectance loss mechanisms are also investigated by altering the dielectric layers used for passivation, together with the study of the effect of Si incorporation during the firing step, showing that this Si presence at the back side is the main responsible for the loss observed.

S A M E N VAT T I N G

In de fotovoltaïsche industrie wordt al vele jaren aluminium gebruikt als materiaal voor de achterzijde contactering van kristallijne silicium zonnecellen. Het doel van dit proefschrift is inzicht te verkrijgen in de vorming van lokale contactpunten van zogenaamde PERCcellen (Passivated Emitter and Rear Cell), een type van hoog efficiënte industriële silicium-zonnecellen. Dit type zonnecel heeft aan de achterzijde verschillende diëlektrische passivatie lagen, waarin een opening gemaakt moet worden voor de formatie van contact punten. Deze studie beschrijft het onderzoek naar het openen van de diëlektrische passivatie lagen met behulp van een laser, naar verschillende methodes voor de depositie van het metaal en naar het sinteren van de contacten, nodig voor de formatie van de aluminium-silicium contacteringslegering. In dit werk zal er aangetoond worden dat het oplossen van silicium in het aluminium een belangrijke rol speelt in het vormen van de achterzijde contactpunten. Dit is nodig omdat op deze manier een effectief elektrisch veld aan de achterzijde gecreëerd wordt door de geïntroduceerde dopering. Anderzijds ontstaan op deze manier diepe contactpunten in de vorm van een piramide, welke de interne reflectie kunnen beïnvloeden en de recombinatiesnelheid verhogen. Verder wordt een nieuwe methode geïntroduceerd om in-situ de formatie van de aluminium-silicium legering gedurende het sinteren te karakteriseren. Hierdoor kunnen verschillende processen onderzocht en verklaard worden wat niet mogelijk is wanneer de zonnecel volledig

v

vi

SAMENVATTING

afgewerkt is. De observaties worden vervolgens met behulp van het fase diagram van beide materialen (Al en Si) verklaard. In dit proefschrift zijn de parameters, die de formatie van het elektrische veld aan de achterzijde beïnvloeden, gevarieerd om te onderzoeken wat het effect is op de prestaties van de zonnecel. Het effect van de contacten, sinterprofiel en -temperatuur, aluminium dikte en samenstelling op hoog efficiënte industriële silicium zonnecellen is onderzocht. Tenslotte is, door het variëren van de diëlektrische passivatie lagen, het reflectie verlies mechanisme aan de achterzijde bestudeerd. De resultaten van dit onderzoek, samen met de studie naar het effect van silicium toevoegingen tijdens de sinterstap, tonen aan dat reflectie aan de achterzijde voornamelijk wordt verminderd door de aanwezigheid van opgelost silicium in de Al-matrix.

ACKNOWLEDGMENTS

Because this is the most commonly read section in every PhD thesis1 , for me it’s a big pleasure to write it. It basically means to look back on the last 5+ years of my life, and when doing so I can only remember kind and helpful people, who I’d like to sincerely thank. I’d like to begin with the man who initiated everything, Joachim. He welcomed me to imec as an internship student, and later opened the door to this PhD. Throughout this time he has consistently helped me in a tireless manner, whenever I popped up in his office. Thanks!. Getting started wasn’t easy. I am extremely grateful to my promoters Professors Jef Poortmans and Robert Mertens, who together initially with Guy Beaucarne, and subsequently Emmanuel Van Kerschaver, established and continued with follow-up meetings, in which they gave suggestions and the motivation for me to continue. In addition, they provided me with answers to all my questions, and gave guidance for this work in parallel. Towards the end of this chapter in my life, Francis was the balanced counterpart for the non-technical knowledge, a definite help against ambivalence. Thanks for that. This manuscript has been carefully revised and corrected by the members of my examination committee. For that, I would like to thank Professors Johan Driesen, Andre Stesmans and Cor Claeys, and also the external member Dr. Daniel Biro. 1 E. Robert Schulman and C. Virginia Cox. “How to Write a Ph.D. Dissertation”, Annals of Improbable Research, Vol. 3, No. 5, pg. 8.

vii

viii

ACKNOWLEDGMENTS

Here, I would also like to acknowledge the chair of this committee, Professor Herman Neuckermans, for his excellent direction in defenses. An important contribution of this work has come from all the suggestions, tips, work and enormous help from many people at imec, too numerous to mention. A big thanks to you all! It is obvious that the biggest portion of that has come from the closest collaborators in my team and group. Particular thanks to Mónica, Yue, Xavier, Christophe, Anne, Jo Das, Patrick, Izabela, Riet, German, Jose Luis, María, Richard, Aude, Sukhvinder, Erik, Sara and Brett, the last of which helped with the corrections in this section. Not to forget, there wouldn’t have been such a thing called a group, without the work of Chantal and Pauline. Thanks a lot for the help throughout the years, as well as to Inge. Knowing that a full support and operations team ready for action when requested is just a phone call away, times have been a lot easier when trying to complete experiments. Therefore, many thanks to Kristof, Andre, Didier, Stefan, Dirk, and later also Reinoud. Giovanni took good care of us guys as he did on my many questions to understand the tricks of the not-that-obvious tool called Duplo, together with Johan. This help has been really appreciated. Since I started in 2008, I had a massive support for characterization from the MCA group at imec. Especially from Danielle, Pierre, Thomas and Wilfried with all of their students (Thilo, Felix, Martin), who did an extensive and excellent job in measuring and analyzing our structures with (several hundreds) SSRM images. The TEM images are the result of the great analysis work of Olivier. Obviously, I can’t forget the members of the “PhD Dream Team”, Bart, Emanuele, Loic and Victor, with who I went through all the ups and

ACKNOWLEDGMENTS

downs of “the roadmap to a PhD”, which, not being the title of a movie, took us as main characters with a winding plot in which, paradoxically, we were the screenwriters as well. Luckily enough, there were moments to enjoy, like memorable conferences and the “occasional” drink together. Lately, the Total team at imec has been rapidly expanding. Starting with Périne, Julien and Michel, to later incorporate Christophe, Matthieu, Amada, Jara, Jens and last but not least Caroline. All of you have been encouraging me, on an earlier or later stage, which I’m very grateful for. One person who devoted long late afternoons-evenings to assist me greatly at the end of the PhD was Jörg Horzel. His commitment to improve my work and strive towards perfection was easily transmitted and helped me in the final push to achieve my goals. Resting assured that the rest of the PhD students will graduate sooner or later in the PV group and will achieve their goals as well, I would like to say thanks for the here-and-there assistance throughout these years. Ending the imec people co-workers group, I want to now highlight two influential imec colleagues, who arrived near the end of my student time but still left a large footprint on me and this work. Filip and Josef, there couldn’t have been a better moment to meet again, in spite of everything. Great to have you here once more!. Nederlands is the Dutch word for the spoken language in Leuven (besides English), and because of this I want to thank Hans for the translation of the abstract into a “samenvatting” and Filip for its proof-reading. A person is not the same without his/her family. My whole family, who, although being thousands of kilometers away, gave me their entire support for this adventure, and I want to acknowledge you for that. &

ix

x

ACKNOWLEDGMENTS

Leuven has given me another family too: Alicia, Dr. Mirko, Dr. Silvia, Jan, Lorena and Dr. Luis, plus Dr. Dani and Marta, who, although they two moved away from here, it was only in distance, and not in my mind. They’ve all been taking relentless care of me in the good times and especially in the bad. Undoubtedly, I’ve found a place in which I could find some many nice people from around the world. Because I’ve met you here, I really enjoy Leuven. Thanks to Mónica, Antoine, Anna, Lorenzo, Jara, Amada, Elena, Mikel, Dongping, Alex, Rufi, Jenny, Denis, Sylwia, Paweł, Arno, Elena, Alessandro, Francesca, Gabriella, Marilú, Stefano, Roberto, Jan, Christos, Hans, Chi, Izabela, Ounsi, Riet, Ivan, Carolina, Steve, Olalla, Friederike, Ruth, Ainhoa, David, Arantxa, Dennis, Ainhoa, Nitin and probably many others that I might forget, unwillingly. In those bad moments and away for almost one year, many friends were encouraging me to recover and come back to Leuven, my second home. Still, my home town, Valladolid, has of course given me very special friends as well, and because of the warm welcomes when coming back to Spain and the far-from-home chats from time to time, I would like to thank especially David, Beni and the Javis :). : Finally, I can hardly imagine how this adventure would have been without you, being of a CAPITAL importance to me, as it has been reflected in this text. Y cómo no, no podía terminar estos agradecimientos sin usar mi lengua materna, y de nuevo decir Gracias a mi familia por estar siempre conmigo, esté yo donde esté.

L I S T O F A B B R E V I AT I O N S A N D S Y M B O L S

Abbreviations ADC Analog to Digital Converter AFM Atomic Force Microscopy ARC Anti Reflection Coating BSF

Back Surface Field

CAD Computer Aided Design CVD Chemical Vapour Deposition DC

Direct Current

DF-STEM Dark Field Scanning Transmission Electron Microscopy EDS/EDX Energy-dispersive X-ray spectroscopy FTC

Fire Through Contact

HAADF-STEM High Angle Annular Dark Field Scanning Transmission Electron Microscopy IR

Infrared

xi

xii

LIST OF ABBREVIATIONS AND SYMBOLS

IV

Current-Voltage

Laser Light amplification by stimulated emission of radiation LFC

Laser Fired Contact

LMJ

Laser Micro Jet

LTC

Laser Transferred Contact

PECVD Plasma Enhanced Chemical Vapour Deposition PERC Passivated Emitter and Rear Cell PERL Passivated Emitter and Rear Locally diffused PID

Proportional–Integral–Derivative

PL

Photoluminiscence

PSG

Phosphoric Silicate Glass

PVD Physical Vapour Deposition SEM Secondary Electron Microscopy SRP

Spreading Resistance Profiling

SSRM Scanning Spreading Resistance Microscopy TEM Transmission Electron Microscopy UV

Ultraviolet


Symbols %Sialloy Si percentage in alloy (%) η

Efficiency (%)

µp

Hole mobility (cm2 /V.s)

ρ

Resistivity (Ω.cm)

ρSi

Si density (g/cm3 )

Alalloy Al weight in alloy (g) Ec

Conduction band energy (eV )

EFn

Electron quasi-Fermi level energy (eV )

EFp

Hole quasi-Fermi level energy (eV )

EF

Fermi level energy (eV )

Ev

Valence band energy (eV )

FF

Fill factor (%)

hν

Light energy (J)

Jsc

Short circuit current density (mA/cm2 )

Lp

Pitch distance (µm)

pAl

Al weight (g)

xiii

xiv


Sidis,bulk Si weight dissolved in alloy (g) Voc

Open circuit voltage (mV )

WBSF Back surface field thickness (µm) Wp

Watt peak (W )

A

Device area (cm2 )

a

Electrical radius of the SSRM tip (nm)

D

Minority carrier diffusion coefficient (m2 /s)

E

Al-Si eutectic composition weight percentage (%)

F

Si weight percentage in liquid phase (%)

L

Diffusion length (µm)

n

Density of free electrons (cm−3 )

p

Density of free holes (cm−3 )

q

Elementary charge (C)

R

Resistance (Ω)

T

Temperature (○ C)

t

Time (s)

Z

Elemental atomic number

CONTENTS Abstract Samenvatting Acknowledgments List of abbreviations and symbols 1 Introduction 1.1 Working principle of a solar cell . . . . . . . . . 1.2 Objective . . . . . . . . . . . . . . . . . . . . . . 1.3 Recombination mechanisms in solar cells . . . . 1.3.1 Radiative recombination . . . . . . . . . 1.3.2 Auger recombination . . . . . . . . . . . 1.3.3 Bulk recombination through defects . . 1.3.4 Surface recombination through defects . 1.3.5 Emitter recombination . . . . . . . . . . 1.4 PERC cells and back surface field principle . . 1.4.1 Local Al-BSF (PERC) . . . . . . . . . . 1.4.2 Full Al-BSF . . . . . . . . . . . . . . . . . 1.5 Thesis outline . . . . . . . . . . . . . . . . . . . . 2 Al-Si alloys and BSF formation 2.1 Introduction to Al-Si alloys . . . . . . . . . . . . 2.2 Phase diagrams . . . . . . . . . . . . . . . . . . . 2.3 Laser ablation process . . . . . . . . . . . . . . . 2.4 Alloy process formation . . . . . . . . . . . . . . 2.4.1 Full Al-BSF . . . . . . . . . . . . . . . . . 2.4.2 Local Al-BSF . . . . . . . . . . . . . . . . 2.5 Si distribution in Al after firing . . . . . . . . . 2.6 Conclusions . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

iii v vii xi 1 4 4 7 8 9 9 11 12 13 13 17 19 21 21 24 27 30 31 35 43 45

xv

xvi

CONTENTS

3 Local Al-BSF formation by in-situ observation 3.1 In-situ observation of the contact formation . . . . . . 3.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 Experimental details . . . . . . . . . . . . . . . . . . . . 3.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 4 Impact of the BSF on the performance of PERC 4.1 Metallized contact fraction . . . . . . . . . . . . . . . . 4.1.1 Pyramidal shaped contacts . . . . . . . . . . . 4.1.2 Irregular-shape contacts . . . . . . . . . . . . . 4.2 The effect of the temperature . . . . . . . . . . . . . . 4.3 Behavior upon cooling . . . . . . . . . . . . . . . . . . . 4.4 Impact of the metallization material . . . . . . . . . . 4.4.1 Al-Si eutectic PVD . . . . . . . . . . . . . . . . 4.4.2 Screen printed Al . . . . . . . . . . . . . . . . . 4.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 5 Dielectric degradation. Impact on reflectance 5.1 Reflectance . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Effect of Al on the reflectance . . . . . . . . . . . . . . 5.3 Transmission electron microscopy (TEM) . . . . . . . 5.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 6 Conclusions and Outlook 6.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix A Appendix B List of Publications Bibliography

47 . . . 47 . . . 48 . . . 50 . . . 50 . . . 56 cells 59 . . . 59 . . . 59 . . . 70 . . . 74 . . . 79 . . . 82 . . . 82 . . . 85 . . . 85 89 . . . 89 . . . 91 . . . 94 . . . 97 99 . . . 99 . . . 101 103 107 109 119

1

INTRODUCTION

We are lucky. We have the cleanest and most powerful nuclear reactor delivering energy for free at a walking distance at light speed and it is not on earth. The humankind has been using the solar energy from the beginning of our existence. Sending us more energy than we consume even nowadays, it is a valuable source for power generation. This actual demand of energy is believed to increase over the coming years globally, as it is reflected in figure 1.1.

200

Real data

World energy consumption [TWh]

175

Prediction

150 125 100 75 50 25 0

1990

1995

2000

2005

2010

2015

2020

2025

2030

Time (Year)

Figure 1.1: World energy consumption and prediction from 2015 till 2030. Source: Enerdata and BP’s Energy outlook 2030 [1][2]

1

2

INTRODUCTION

Figure 1.2: Record efficiencies over the years and their authors for different technologies. The figure is published in the NREL website [3]

Because of this, and the limitation in resources, price volatility of oil supplies and climate change risks for fossil fuels that nowadays are being used to satisfy this demand, the use of renewable energies is key to a sustainable future. Solar energy is one of the multiple choices for renewable energy that we have available to ensure this sustainability. One can say that the history of the solar cells started in 1839 when Becquerel discovered the photovoltaic effect. It was later, in 1883, when Charles Fritts fabricated the first photovoltaic cell using selenium as semiconductor material and gold to form the junctions [4]. Then, in 1888, Aleksandr Stoletov, a Russian physicist, built the first photoelectric cell1 based on the photoelectric effect which was discovered in 1887 by Heinrich Hertz 1 http://en.wikipedia.org/wiki/Solar_cell

INTRODUCTION

2.0

1.8

mc-Si based module price 1.6

p

Price ($/W )

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0 29/04/2010

29/01/2011

29/10/2011

29/07/2012

29/04/2013

Date

Figure 1.3: Weekly average multicrystalline Si module price evolution since April 2010; After [7]

[5]. Chapin et al. in 1954 made the first usable solar cell with a 6% efficiency [6]. These first discoveries and developments were crucial for the expansion of the photovoltaic research and those were the enablers for the most recent records achieved by several institutes and companies throughout the years, as it is shown in figure 1.2. Lately, solar incentives have been cut all over the world, and Si prices have been going down in the last two years quite dramatically, decreasing by a factor of 3. The price per watt peak at module level has also dropped in the last years, as reflected in figure 1.3. This shows that the actual drivers for the market are the efficiency and the cost of the system, in order to be competitive.

3

4

INTRODUCTION

Front contact Emitter n-type

Light

+-

Base p-type

External load

Back contact

Figure 1.4: Schematic cross-section representation of the working principle of a solar cell

1.1.

Working principle of a solar cell

As already introduced at the beginning of this chapter, the solar cell working principle is the photovoltaic effect. The word “photovoltaic” comes from, first, the Greek term “ρωτ ” (phot-, meaning light) and second, after the Italian physicist Volt, who invented the voltaic pile (meaning electricity). Then, a solar cell is basically a device that converts light into electricity. This light that falls onto the device generates an electron-hole pair, which, due to the difference in the energy potential created by the p-n junction present in the device, is separated to both sides of the cell and extracted to the external circuit through the metal contacts, thus providing power. Figure 1.4 depicts a schematic representation of the solar cell.

1.2.

Objective

In this work, the focus will be the research on the PERC type of solar cells. This thesis aims first to understand the formation of the contacts

1.2 OBJECTIVE

Screen printed Ag n+ Al-BSF

Screen printed Ag

SiN y Si p

Al paste

n+ Al-Si alloy

Al-BSF

SiO2/SiNy Si p SiO2/SiOx/SiNy Al-Si alloy

Figure 1.5: Full Al-BSF (left) vs. PERC or local Al-BSF (right)

and then to find a way to improve those contacts for the mentioned PERC type of solar cells. A schematic representation of this type of cells is displayed in figure 1.5 (right side) together with the standard in industry, the full Al-BSF (left side), on which the comparison for the formation of the contact on the back side of the cells will be presented in section 1.4. The main results achieved in this work are based on Al PVD metallization for the rear side of the cells using this PERC concept. As a comparison, cells metallized with Al screen printing technology are also covered in this study. The first definition of this type of cells was given by the University of South Wales in Australia [8–10]. It basically overcomes the three mayor limitations present in the full Al-BSF cells: • High minority carrier recombination at the rear surface, due to the lack of any passivation layer on the back side. • Poor optical enhancement, as light gets absorbed in the Al-Si alloy, not acting as a reflector. • Visible bowing of cells, complicating the cell to module integration process.

5

6

INTRODUCTION finger

(a)

n+

n

"inverted" pyramids

p- silicon

rear contact

oxide finger

"inverted" pyramids

oxide

n+ p+

p+ n

p- silicon p+

rear contact

oxide

(b)

p+

oxide

Figure 1.6: PERC (a) /PERL (b) cell concepts as defined in [8, 9] and [10]

The incorporation of a dielectric layer in between the bulk Si and the metallization layer is the key of the improvements listed above. It increases the passivation properties, allowing a higher Voc , increases the rear reflectance, improving the Jsc , and acts as a buffer that prevents stress coming from the difference in thermal expansion coefficients of Al and Si, lowering the bowing of the cell [11, 12], without the need of an extra cooling as reported in [13, 14]. As Si solar cells with well passivated front and rear surfaces lead to high conversion efficiencies, the photovoltaic industry is currently on the way to apply local rear contacts with locally doped regions underneath the rear contacts instead of using full area rear contacts of Al alloys. The PERC cell features a passivated front on top of the n-type emitter and passivated rear with local contacts connecting the p-type base to the external circuit, whereas the full Al-BSF has full contact area between Al and Si, providing limited passivation for the back side. It was originally presented by Blakers et al. without a BSF, as it can be observed in figure

1.3 RECOMBINATION MECHANISMS IN SOLAR CELLS

1.6, just using highly doped base Si in order to create an ohmic contact [8], but in this thesis an alloy between an Al layer and the bulk Si will be formed, giving, upon cooling down, a highly doped region formed by the re-crystallization of Si expelled by Al from the melt [15, 16]. This Al will be deposited either by means of PVD or screen printing. Particularly, the study of the formation of this back contact with PVD Al will be the main part of this investigation, both in terms of material exchange characterization and electrical performance. The PERC structure is also called local Al-BSF, due to the limited contacted area in comparison with the full Al-BSF cell in which the whole back side of the device is acting as a contact. In summary, the objective of this thesis is to investigate the formation of the back contact in PERC structures for crystalline silicon solar cells. In industry, the use of Al screen printing is still dominating. However, the use of PVD Al to substitute this screen printing technology when using the PERC concept is getting into the market [17]. Along with this investigation, the modification of the properties of the dielectric layers at the rear side of the cells influenced by this contact formation will be also analyzed. A novel way to improve the contact formation is also developed and demonstrated.

1.3.

Recombination mechanisms in solar cells

In this section, the recombination mechanisms are presented. This theoretical approach has been described in detail by Kerr [18]. Any electron which exists in the conduction band is in a meta-stable state and will eventually stabilize to a lower energy position, filling an empty valence band state. Therefore, when the electron stabilizes back down

7

8

INTRODUCTION

into the valence band, it annihilates a hole. There are different types of recombination in semiconductors. These are explained in the following subsections.

1.3.1

Radiative recombination

Radiative recombination is simply the direct annihilation of an electronhole pair as depicted in figure 1.7. It is the inverse process to optical generation, the excess energy being released mainly as a photon with an energy close to that of the bandgap. It involves a conduction band electron falling from an allowed conduction band state into a vacant valence band state (a hole). The radiative recombination rate, U rad , therefore depends on the concentration of free electrons, n, and free holes, p:

Urad = Bnp,

(1.1)

where B is the coefficient of radiative recombination. From detailed balance considerations, the value of B for silicon has been calculated to be 2.10−15 cm3 s−1 [19, 20]. Experimental determinations for the radiative recombination coefficient in silicon are significantly higher however, with B = 9, 5.10−15 cm3 s−1 [21]. The reasons for the discrepancy are not clear, although it is believed that excitonic effects and the assumed value for the intrinsic carrier concentration in silicon are responsible [22].


eEc Ev

h+

Photon

Figure 1.7: Radiative recombination process

1.3.2

Auger recombination

Traditionally, Auger recombination is viewed as a three-particle interaction where a conduction band electron and a valence band hole recombine, with the excess energy being transferred to a third free electron or hole, as depicted in figure 1.8 [23]. The charge carriers involved are assumed to be non-interacting quasi-free particles [24]. The eeh process denotes when the excess energy is transferred to another electron, with the recombination rate given by Ueeh = Cn n2 p. Similarly, the ehh process denotes when the excess energy is transferred to another hole, with recombination rate Uehh = Cp np2 , where Cn and Cp are Auger coefficients. The total Auger recombination rate, UAuger , is then given by UAuger = Cn n2 p + Cp np2

1.3.3

(1.2)

Bulk recombination through defects

The presence of defects within a semiconductor crystal, be they from impurities or crystallographic imperfections such as dislocations,

9

10

INTRODUCTION

b)

eEc Ev

h+ a) Figure 1.8: Auger recombination process

produces discrete energy levels within the bandgap. As shown in figure 1.9, these defect levels, also known as traps, greatly facilitate recombination through a two-step process whereby a free electron from the conduction band first relaxes to the defect level and then relaxes to the valence band where it annihilates a hole [25]. The dynamics of this recombination process where first analyzed by Shockley and Read [26] and Hall [27], with the recombination rate, USRH , for a single defect level given by

USRH =

np − n2i , τpo (n + n1 ) + τno (p + p1 )

(1.3)

where τpo and τno are the fundamental hole and electron lifetimes which are related to the thermal velocity of charge carriers, nth , the density of recombination defects, Nt , and the capture cross-sections, svp and svn , for the specific defect: τpo =

1

svp nth Nt

and τno =

1

svn nth Nt

(1.4)


Finally, n1 and p1 are statistical factors defined as n1 ≡ NC exp (

Et − EC EC − EG − Et ) and p1 ≡ NV exp ( ) kT kT

(1.5)

where NC and NV are the effective density of states at the conduction and valence band edges, EC and EG are the conduction band and bandgap energies and Et is the energy level of the defect.

1.3.4

Surface recombination through defects

The surfaces or interfaces of a silicon substrate represent a severe discontinuity in its crystalline structure [22]. The large number of partially bonded silicon atoms give rise to many dangling bonds, and therefore a large density of defect levels are found within the bandgap near the semiconductor surface as illustrated in figure 1.9. Even if the silicon surface is not covered by a surface layer, say due to a native oxide, the presence of silicon-oxygen bonds can stress the crystal structure at the surface, which again introduces many defect states. The SRH analysis of subsection 1.3.3 again applies, although it has to be reformulated in terms of recombination events per unit surface area, rather than per unit volume. For a single defect at the surface, the rate of surface recombination, US , is given by

US =

ns ps − n2i ps +p1 , ns +n1 Spo + Sno

(1.6)

11

12

INTRODUCTION

eEc Ev

h

+

Themal energy

Figure 1.9: Shockley Read Hall (SRH) recombination process

where ns and ps are the concentrations of electrons and holes at the surface, and Sno and Spo are related to the density of surface states per unit area, Nts , and the capture cross-sections, σp and σn .

1.3.5

Emitter recombination

Modeling the recombination occurring within an emitter region from first principles is relatively complicated. Doping profiles are normally not uniform and so the spatial variation in the dopant concentration needs to be considered, as does the possibility of a dead layer from the diffusion process. Heavy doping effects including the degeneracy of the semiconductor, band-gap narrowing and free-carrier absorption need to be included, along with intrinsic recombination processes and the normal SRH recombination processes related to defects, both within the emitter and at the emitter’s surface. Because the emitter regions are heavily doped, two simplifications follow. Firstly, the minority carrier concentration in the emitter region normally remains low and secondly, Auger recombination is likely to be the dominant bulk recombination mechanism. It follows that the recombination lifetime in the emitter region is constant with injection level, and the recombination current into the emitter (Jrec ) can be expressed as in equation 1.7, where J0E is defined to be the emitter saturation current density, and n and p refer to

1.4 PERC CELLS AND BACK SURFACE FIELD PRINCIPLE

the electron and hole concentrations on the base side of the space-charge region. Jrec = J0E

np n2i

(1.7)

Emitter recombination can be viewed as a special case of surface recombination. In figure 1.10, a device simulation is plotted showing how much these types of recombination account for the total recombination in the cell for a given wafer resistivity [28]. Both cell concepts considered in this thesis, the full Al-BSF and the local Al-BSF, are compared (continuous lines and dotted lines, respectively). There is a lower recombination current in the BSF region when using a local Al-BSF, almost an order of magnitude. This motivates working on the back side of this type of cells, apart from the benefits explained in section 1.2.

1.4.

PERC cells and back surface field principle

In this section, the process flow for the solar cells used in this thesis is presented. Furthermore, the basic working principle of the back surface field in a solar cell is shown. At the end of the section, the process flow for the reference type of solar cells, the so called full Al-BSF will be shown.

1.4.1

Local Al-BSF (PERC)

Figure 1.11 shows the different processes and steps realized for the standard process. These steps are:

13

INTRODUCTION

Afterbdegradation

Recombinationbcurrentbdensityb[mA/cm2]

14

Waferbresistivity:b3bOhm.cm

FullbAlEBSF LocalbAlEBSF

10

Total 1

Emitter BSF Base Vmpp

0.1 0.40

0.45

0.50

0.55

Voc 0.60

0.65

Voltageb[V] Figure 1.10: Simulated recombination losses in various device regions of an optimized full-area BSF cell (continuous lines) and in a cell with local Al-BSFs (dashed lines); After [28]

1.4 PERC CELLS AND BACK SURFACE FIELD PRINCIPLE

1

7

2

8

3

9

4

10

5

11

6

12

ScreenxprintedxAgx n+

Al-BSF

SiO2/SiNy Sixp SiO2/SiOx/SiNy Al-Sixalloy

Figure 1.11: Process flow for the i-PERC type of cells used in this work

15

16

INTRODUCTION

1. The base substrate chosen is mono-crystalline p-type Si 100oriented, 170 µm thick, resistivity 0.5-3 Ω.cm. 2. Two-sided alkaline texturing in order to remove the saw damage present on the wafers and provide the surface conditioning to absorb the light more efficiently into Si. 3. Rear side polishing for roughness removal to achieve better passivation and provide the right surface conditioning for the laser ablation step. 4. Emitter formation. Phosphorus is the chosen dopant to create the p-n junction in p-type silicon material. In this work, a two step diffusion process is applied, first a deposition of a PSG glass by means of POCl3 followed by a drive in step at moderate temperature (850 ○ C). 5. Emitter removal on the rear side. In order to electrically decouple the front side from the rear, the diffused rear side has to be removed. This is done in an inline tool by means of acidic etching of Si, removing a few hundred nanometers thick layer. 6. Low temperature ( 8.5 s

400 300

< 2.5s

200

Fast cooling (885 °C)

100

Fast cooling (935 °C)

0

Slow cooling (935 °C)

0

5

10

15

20

25

30

35

40

Time (s)

Figure 4.20: Comparison between two types of firing profiles, averaged over four thermocouples (fast and slow cooling).

Standard cool down 10 µm

Slow cool down 10 µm

Figure 4.21: BSF thickness comparison between fast (left) and slow cooling (right)

4.3 BEHAVIOR UPON COOLING 1500

1500 1300

Liquid

1100 900

TemperatureMºC

TemperatureMºC

1300

MaximumMtemperature

700 577MºC

500 300

10

20

30

40

1100 900

MaximumMtemperature

700 577MºC

500

StandardM coolingMdown

0 Al

Liquid

50

WeightMwMSi

60

70

80

90

100 Si

300

SlowM coolingMdown

0 Al

10

20

30

40

50

WeightMwMSi

60

70

80

90

100 Si

Figure 4.22: Possibles scenarios at phase level for the standard (left) and slow cool down (right)

re-crystallization as the Al amount and the firing peak temperature were identical for both cases. This result suggests that the cooling down process does not follow the thermodynamic equilibrium for standard conditions when the cooling is forced after the heating zones in the furnace, differentiating it from a natural cooling, which is closer to the slow cooling profile. Although not a purely natural cooling condition, this process it is much closer to it than the standard reference process. This result does not contradict the hypothesis detailed in chapter 2 related to the process formation of this BSF region as the phase diagram does not consider time restrictions. Figure 4.22 sketches a possible sequence for the two cases. By cooling fast, the thermodynamic equilibrium could be altered and part of the melt, instead of following the liquidus line, can also solidify in hypereutectic form, i.e. Si content higher than 12.7%. This has been studied by SEM and EDX, as pictured in chapter 2, where Si rich island were found around the contacts. Figure 4.23 depicts one of this Si island with its correspondent EDX map, with Si colored in green and Al in red. The thicker BSF observed when cooling down slowly, as opposed to the

81

82

BSF ON THE PERC PERFORMANCE

20 µm

20 µm

Figure 4.23: Detail of one of the Si-rich islands found for the standard firing conditions after Al etching. SEM is on the left and EDX on the right (Al in red and Si in green)

one with fast cooling, can be due to the enhanced re-crystallization, as the melt should follow the liquidus line over longer time during cooling down in the first case. No Si islands were found in the slow cooling sample. In this section, the contact formation has been studied from the point of view of time which is not taken into account in the phase diagram (infinite time supposed) to understand the effect of a different cooling rate in the BSF formation, resulting in a thicker region due to improved re-crystallization.

4.4.

4.4.1

Impact of the metallization material

Al-Si eutectic PVD

As seen in section 4.1, the use of PVD Al-Si with the eutectic composition (12.7% Si in Al) has advantages over pure Al metallization in terms of Voc mainly due to enhanced re-crystallization of the BSF layer. But also

4.4 IMPACT OF THE METALLIZATION MATERIAL

the series resistance increases as the conductivity of Si rich alloys is lower than Al, which might affect the integration of this technology at module level. As a case study, in this subsection, we evaluate the possible benefits of the addition of an extra Al layer just on top of the mentioned Al-Si 12.7 %. This should help to increase the conductivity of the back metallization. In order to analyze this, a similar experiment as the one done in section 4.1 was performed, but the laser pitch and the metal thickness in this case were kept fixed, and the front grid optimized. As first group and reference, a 2 µm Al metallization was used. In the second group, an Al-Si metal layer of 1 µm was deposited, and finally in a third group, a combination of a first layer of Al-Si of 1 µm, plus an additional 1 µm of pure Al was deposited in order to compensate the higher series resistance, responsible for the FF loss, observed for the Al-Si metallization. All the results are collected in figure 4.24. For the Jsc there is an improvement for the last group, Al-Si +Al, which is due to the improved reflectance caused by the extra Al layer. For the Voc , the group with highest average is the one with only Al-Si, due to the smallest pyramid size in the contacts. The FF benefits in the case of the extra layer of Al on top of Al-Si, due to the increase in conductivity and thus the decrease in contact resistance, having the highest values in average comparing to the other two groups. Finally, looking at the efficiency, the third group with this combination of Al-Si and Al represents an improvement of 0.1 % absolute in average over the Al-Si case. These results have demonstrated that it is possible to compensate at cell level the higher series resistances associated to the Al-Si sputtering by adding an extra Al layer, which also results in a higher rear reflectance improving the Jsc of the cells. By lowering the series resistance, the FF values increase as well, overcompensating a small loss in Voc coming from a higher amount of Al available during the firing step.

83

38.8 38.7

665

38.6

660

Voc [mV]

Jsc [mA/cm²]


38.5 38.4 38.3 38.2 38.1

655 650

Al 2 µm

AlSi 1µm AlSi 1µm + Al 1µm Metal split

Each Pair Student's t 0.05

80.0

Al 2 µm


Al 2 µm



20.4 Efficiency [%]

79.5 FF [%]

84

79 78.5 78 77.5 Al 2 µm



20.2 20 19.8 19.6 19.4


Figure 4.24: IV results for Al-Si metallization with different layers

4.5 CONCLUSIONS

4.4.2

Screen printed Al

At the beginning of this work, several references have been made about the use of screen printed pastes for the metallization of the cells on the back side. The nature of these pastes makes necessary the addition of other components as frits, binders, and other metals as lead, for example. All this non-pure Al, in contrast with the PVD Al referred in the previous sections which has a high purity (5N), impacts the process explained in chapter 2, as Urrejola suggested in [96]. Due to this intrinsic nature with many elements present in the pastes, a full scientific study is very complex. Here a simple comparison between the PVD route and the screen printed paste one is reported. Samples were fabricated following the process flow used in this work as explained in chapter 1 up to the metallization step. Then, two groups were differentiated, one using Al PVD and the other an Al containing paste. The cell results are contained in figure 4.25. The most significant difference is in the Voc values. This is mainly due to the different imprint size for the two splits. Although the laser opening parameters are kept constant, the different amount of Al applied, which is bigger for the paste case (~30-40 µm thick versus 2 µm PVD), makes the pyramid size different between them, as more Al is available to dissolve Si. Figure 4.26 collects SEM images from both, showing pyramid sizes of 83 and 100 µm as base sides, for the same firing conditions.

4.5.

Conclusions

In this chapter 4, the different parameters influencing the formation of the BSF region and the contact itself are investigated. These parameters

85

650

38.1 38 37.9

Voc [mV]

Jsc [mA/cm²]


37.8 37.7 37.6 37.5 Al Paste

Al PVD

Paste split

640 635 Al Paste

Al PVD

Al Paste

Al PVD

Paste split


Efficiency [%]

19.6

79.8 79.6 79.4 79.2

645

630


80

FF [%]

86

Al Paste

Al PVD

Paste split


19.4 19.2 19 18.8

Paste split


Figure 4.25: IV result for the comparison between Al PVD and Al paste

50 µm

50 µm

Figure 4.26: SEM images for PVD case (left) and screen printed case (right)

4.5 CONCLUSIONS

are metal contact fraction, firing peak temperature, firing profile shape and composition of the metal. All these have been tested and explained over the different sections. It has been demonstrated that the addition of Si to Al improves the voltage of the cells. A modified sputtered layer structure has been implemented to compensate the loss in FF caused by an increased series resistance due to the presence of Si. In terms of peak firing temperature, 850 ○ C has resulted in the best efficiency for these type of cells. The use of a modified firing profile has led to a thicker BSF, which enhances the field effect to repel minority carriers from the back surface. At the end of the chapter, a comparison between PVD and screen printing metallization is presented. This experimental work has revealed the open circuit voltage to be the most differentiating parameter, mainly due to the higher amount of Al applied in the screen printed case (~30-40 µm) versus 2 µm PVD, which produces bigger pyramids for the first case.

87

5

D I E L E C T R I C D E G R A D AT I O N . I M PA C T O N R E F L E C TA N C E

In chapter 3, the in-situ observation of the alloying process has been explained. There, it has been shown how far and fast the Si dissolution from the bulk takes place during the alloying process. In this chapter, we will investigate how the presence of this Si has an impact on the reflectance at the back side of the cells. First, the rear reflectance will be studied for different dielectric stacks, showing the effect of depositing pure Al or Al-Si12.7% on this reflectance. The impact of having or not laser openings in the layers will be also taken into account. After that, transmission electron microscopy images will illustrate what is happening at the interface between the dielectric layers and the metal layer. Finally, some of these configurations will be used to fabricate solar cells to evaluate the short circuit current values of these configurations.

5.1.

Reflectance

Typically, the performance of the solar cell is dependent on its total reflectance, and different types of cells have shown different rear reflectance [121]. This reflectance contributes in the long wavelength range to return the light back into the cell instead of letting it transmitting through, generating in this way more carriers. For the evaluation of this reflectance at the back side, the samples were measured

89

DIELECTRIC DEGRADATION. IMPACT ON REFLECTANCE

70 60

Non fired 685 ºC

50 Reflectance (%)

90

735 ºC 785 ºC

40

835 ºC 885 ºC

30

935 ºC

20 10 0 800

900

1000

1100

1200

1300

Wavelength (nm)

Figure 5.1: Total reflectance for different peak firing temperatures.

from the front side, and then looking at the long wavelengths, following the methods given by Rand [122]. From previous experiments, the reflectance loss due to the absorption in the emitter, which affects the free carrier absorption (FCA) [123], was analyzed. The firing temperature also has a big impact on this reflectance. Figure 5.1 visualizes the total reflectance of a series of samples, from non-fired, serving as a reference (in dotted line), to samples fired at different temperatures, from 685 to 935 ○ C, in which a trend can be observed, having lower reflectance when the temperature increases. This can be caused by several individual mechanisms, all of which occur during this firing step: • large imprint size with high parasitic absorption, blocking subsequent light bounces • free carrier absorption in the BSF (not discussed here)

5.2 EFFECT OF Al ON THE REFLECTANCE

• degradation of Al reflectance by incorporation of Si • degradation of dielectric reflectance by reaction with Al or Si • reduced Al coverage by out-gassing of H contained in rear dielectrics The investigation in this study will be focused on determining which of these mechanisms is dominant on the reflectance loss. Because of the result presented in figure 5.1, the peak firing temperature was kept constant for this analysis. After observing the strong loss in reflectance when firing the cells, the goal of this experiment was to study the dielectric degradation, as highlighted in chapter 3, and its possible causes on this rear reflectance. Together with that, the addition of Si to Al for metallization will be studied as well. In order to do that, the dielectric layers were modified to understand the possible influence and the use of the Al target containing the eutectic composition of Si will be used.

5.2.

Effect of Al on the reflectance

Test wafers were prepared for two separate experiments. For both experiments no emitter diffusion was performed to prevent free carrier absorption in that region which might affect the reflectance measurements in the 900-1200(1300) nm range. In a first experiment, the rear passivation stack was varied (A = SiO2 /SiOx /SiNy /Al, B = SiO2 /SiNy /Al) while keeping the front ARC identical (SiO2 /SiNy ). SiO2 refers to a thin dry thermal oxide, while SiOx /SiNy refers to PECVD oxide/nitride respectively. Here, no contact hole was opened with the laser. Reflectance and transmission measurements were performed prior to 2 µm PVD Al deposition, after Al deposition, and after Al firing (885°C set peak temperature). This process flow is outlined in figure 5.2.

91

92

DIELECTRIC DEGRADATION. IMPACT ON REFLECTANCE StackyAy=ySiO2BSiOxBSiNyBAly

StackyB =ySiO2BSiNyBAl

Inlineyalkalineytexturization Inlineyrearysideypolishing Cleaning Thermalyoxidationy8bbyC ARCyfrontySiNx PECVDyrearySiOx passivation RearySiNy ReflectanceBTransmissionymeasurements Rearylaseryablation 2yµmyAlyPVD

2yµmyAlSi12%7Py PVD

Rearylaseryablation 2yµmyAlyPVD

ReflectanceBTransmissionymeasurements Firing ReflectanceBTransmissionyymeasurements

Figure 5.2: Process flow for the test structures used in this section

Results are shown in figure 5.3. The reflectance measurements as a function of the wavelength are on the left and the transmission measurements on the right for the two mentioned stacks, A (in red) and B (in blue), after the three steps mentioned in figure 5.2, which means the dielectric layers as deposited (represented by squares), after the Al deposition (circles) and after firing (shown with triangles). Based on these results, it can be seen that when no contact holes are present, firing (and melting) pure Al only results in a ~2 % drop in reflectance at 1200 nm which corresponds to light being transmitted through the wafer. This is likely caused by H out-gassing during firing reducing Al coverage. If a simplified stack (B) is used instead, initially, without Al, the reflectance is at the same exact level than the full stack. With Al present and also after firing, stack B clearly has a lower reflectance than stack A so SiOx plays an important role (~15 % reflectance loss without it).

5.2 EFFECT OF Al ON THE REFLECTANCE 80 70

50

30

B_no laser_no Al B_no laser_Al B_no laser_Al fired

20

r

40

Reflectancer(%)

60

rB_norlaser_norAl rB_norlaser_Al rB_norlaser_Alrfired

40

A_no laser_no Al A_no laser_Al A_no laser_Al fired

30 20

10

Transmission (%)

rA_norlaser_norAl rA_norlaser_Al rA_norlaser_Alrfired

10 900

950

1000

1050

1100

Wavelengthr(nm)

1150

0 1200

900

950

1000

1050

1100

1150

0 1200

Wavelength (nm)

Figure 5.3: Reflectance (left) and transmission (right) with no laser openings

In a second experiment, the rear stack A was used and each wafer was divided into 3 zones: i) no rear ablation, ii) picosecond ablation with an opening diameter of 20 µm and pitch Lp = 600 µm (ps-laser) and iii) nanosecond ablation with an opening diameter of 50 µm and Lp = 600 µm (ns-laser). Al PVD or Al-Si12.7% PVD 2 µm thick were deposited and reflectance measurements were performed before and after firing. Figure 5.4 collects these results, on the left with pure Al and in the right with Al-Si12.7% . From the results, it is found with Al-Si12.7% that the reflectance drops significantly after firing even without laser openings, contrary to the pure Al firing, and that no further drop is observed if local Al-BSF contacts are formed. This demonstrates that Si present in Al is the main cause for the rear reflectance loss upon firing (~20%). In the case of pure Al layers, the reflectance loss is caused by Si coming from the wafer which is affected by the laser opening diameter. Reducing the diameter opening reduces the rear reflectance loss (ps-laser) although this leads to increased series resistance.

93

94

DIELECTRIC DEGRADATION. IMPACT ON REFLECTANCE 80

80 70

50 40 30 20

Reflectancer(%)

60

rA_norlaser_AlSi 12.7rA_norlaser_AlSirrrrrrfired 12.7rA_ns-laser_AlSi 12.7rrrfired rA_ps-laser_AlSirrrrrrfired 12.7-

70 60 50 40 30 20 10

10 900

950

1000

1050

1100

Wavelengthr(nm)

1150

0 1200

Reflectancer(-)

rA_norlaser_Al rA_norlaser_Alrfired rA_ns-laser_Alrfired rA_ps-laser_Alrfired

900

950

1000

1050

1100

1150

0 1200

Wavelengthr(nm)

Figure 5.4: Average reflectance measurements with 2 µm PVD pure Al (left) and with 2 µm PVD AlSi12.7% (right)

5.3.

Transmission electron microscopy (TEM)

Transmission electron microscopy (TEM) images were taken for both cases, with Al PVD and Al-Si12.7% . The results for the pure Al case with no laser holes are collected in figure 5.5. The whole structure is shown in a), depicting from top to bottom, Si/SiO2 /SiOx /SiNy /Al. A high angle annular dark field scanning TEM (HAADF-STEM)[124], taken from the marked area in a), is depicted on b). The interface between SiNy and Al is magnified in c). In HAADF, the contrast is proportional to the thickness of the TEM specimen and to ~2 where Z is the atomic number [125]. In this sample, the thickness is constant, and therefore differences in contrast are coming from the different elements. Due to the application of the Ga ion beam used for the thinning of the TEM lamellas, Ga is also present. Being this element the heaviest one among the present ones, it is the brightest in the image in figure 5.5b. It is known that Ga accumulates at the interfaces of Al with other materials [126, 127], and it has been confirmed by EDX analysis for this sample. It is still visible in the TEM close-up (right) image a limited reaction in

5.3 TRANSMISSION ELECTRON MICROSCOPY (TEM)

a)

TEM Al-N=SiO 2/SiOx/SiNy/Al fired

b) HAADF-STEM c) TEM

SiO 2/SiOx

Si SiO 2/SiOx SiN Al

SiN Interaction zone Ga (from FIB) Al

Figure 5.5: TEM image of the Si/SiO2 /SiOx /SiNy /Al stack fired without laser openings (a). The squared area has the HAADF displayed in b) where Ga is shown in brightest contrast. In c), a zoomed TEM image shows an interaction zone visible in dark contrast.

between SiNy and Al (dark zone at the interface, marked “interaction zone”). This configuration resulted in ~70 % reflectance at 1200 nm, as plotted in previous figure 5.3. If Al-Si12.7% is used instead of pure Al, with the same stack as before and with no laser openings, Si will be present in the Al side as deposited, as depicted in figure 5.6. Columnar Si crystals in the Al-Si12.7% layer with smaller crystals at the SiNy /Al-Si12.7% interface are observed (identified by EDX), causing a small reflectance loss (previous figure 5.4), between the Al and the Al-Si12.7% cases. In the absence of firing, no interaction zone is visible in c). Figure 5.7 shows the fired version of the previous sample. Si coming from the Al-Si12.7% is attached to the surface of the SiNy layer. Ga contrast indicates where the boundaries of the interface between these Si grains and Al are (or the SiNy /Al interface), which are also surrounding Si grains, instead of accumulating only over the interface between SiNy and Al, as previously observed in figure 5.6. The interaction zone visible in

95

96

DIELECTRIC DEGRADATION. IMPACT ON REFLECTANCE Si TEM AlSi-noIfiring=SiO2/SiOx/SiNy/AlSi12.7% a)

b)

HAADF-STEM

c)

SiO 2/SiOx

SiN

SiO 2/SiOx TEM SiN

Ga (fromIFIB) AlSi

AlSi SiIgrain

SiIgrains

Figure 5.6: TEM image of the Si/SiO2 /SiOx /SiNy /Al-Si12.7% stack as deposited (with no laser openings) and no firing (a). On the squared area, the HAADF is displayed in b) where Ga is found and shown in bright contrast, surrounding small crystals. The zoomed TEM image is depicted in c).

TEM is more or less similar in both fired cases, indicating a SiNy /Al reaction. The presence of a thin Si-rich layer at the Al/SiNy interface, causes the reflectance loss (~20 %) probably due to high parasitic absorption, which is not the case when no Si is present in Al as in figure 5.5. Laser/Pitch

Voc (mV)

Jsc (mA/cm2 )

FF (%)

h (%)

ps/600 µm

659.3 ± 0.8

38.3 ± 0.1

78.0 ± 1.1

19.7 ± 0.3

ns/600 µm

658.0 ± 1.9

38.3 ± 0.1

79.4 ± 0.5

20.0 ± 0.1

ps/400 µm

657.3 ± 1.0

38.3 ± 0.1

79.5 ± 0.7

20.0 ± 0.2

Table 5.1: Average and standard deviation I-V results over 6 i-PERC solar cells with PVD Al-Si12.7% and screen printed Ag front

5.4 CONCLUSIONS AlSi-N=SiO2/SiOx/SiNy/AlSi12.7) firedG Si TEM SiO 2/SiOx

a)

SiN AlSi voids

SiGgrains

HAADF-STEM

TEM SiO 2/SiOx

b)

c)

Ga (fromGFIB)

SiN

InteractionGzone AlSi

Figure 5.7: TEM image of the Si/SiO2 /SiOx /SiNy /Al-Si12.7% stack fired with no laser openings (a). On the squared area, the HAADF is displayed in b) where Ga is found around Si grains and SiNy /Al interface and shown in bright contrast. The zoomed TEM image is depicted in c).

In order to verify the reflectance results plotted in figure 5.3, i-PERC solar cells were processed on 125 mm Cz-Si p-type 1-3 W.cm wafers. The results are given in table 5.1. The cells feature screen printed Ag front contacts, PVD Al-Si12.7% metallization, and various rear opening schemes. The Jsc values correlate nicely with the rear reflectance results being identical for the various opening schemes used.

5.4.

Conclusions

In this chapter, the mechanism for the reflectance loss observed in iPERC cells when performing the last high temperature step (firing) is investigated. It has been demonstrated that the presence of Si (confirmed by TEM) at the SiNy /Al interface at the backside of the cells causes a loss in reflectance (up to 20 % absolute at 1200 nm). This is probably due to parasitic light absorption in that region, even when no pyramid are created for the contacts, performed by using an Al-Si12.7% PVD

97

98

DIELECTRIC DEGRADATION. IMPACT ON REFLECTANCE

metallization. If no Si is present in Al during firing in the absence of dielectric openings, the reflectance loss is almost negligible (~2 % absolute). SSRM and TEM images have resulted in helpful understanding the mechanism. SSRM showed the reaction between Al-Si and the dielectric layers. TEM images spot the presence of Si on the SiNy /Al interface if an Al-Si alloy is in contact with that interface, being a determining factor for the reflectance loss observed, which does not happen when no Si is found in that interface. Finally, using this Al-Si12.7% PVD metallization, solar cells have been fabricated, resulting in identical short circuit current values for different rear side dielectric openings, corroborating the reflectance results obtained in our measurements.

6

CONCLUSIONS AND OUTLOOK

6.1.

Conclusions

As a summary, in this chapter 6 the most important aspects of this thesis will be gathered. The thesis consists of six chapters. The work has been focused on the study of the characteristics of the back surface field that is used in the i-PERC type of silicon solar cells to efficiently repel the minority carriers from the back surface of the cells around the contacts. The parameters that influence the formation of the BSF region have been studied and different ways to improve it have been investigated and new methods and techniques have been employed. Hypotheses have been proposed to explain the local contact formation on i-PERC solar cells by analyzing the phase diagram between Al and Si, approximating the time limited case to this diagram. In particular, the use of the SSRM technique, along with the better known SEM, has been proven to be an extremely useful approach to characterize this BSF region. This analysis led to the finding of the degradation of the SiOx film, acting as the dielectric layer at the back side. And not only for the BSF, because thanks to this technique, the reaction that takes place when the Al-Si melt is in contact with the SiOx

99

100


dielectric layer used in the passivation layers, leads to possible shunt paths known in these type of cells has been identified. The introduction of Si in the Al deposition led to an improved performance at cell level, because of the enhancement of the open circuit voltage due to the smaller pyramid formation. The proximity of Si due to a smaller contact pyramid has been proven to benefit a thicker recrystallization of the BSF. Further modifications of this metallization have been investigated to improve the intrinsic lower conductivity of this type of alloy. For the first time, the dissolution of Si into the Al layer on the backside of the cells has been visualized in-situ, allowing for a better understanding of the contact formation. It has been seen that Al in contact with Si effectively reduces its melting point to the eutectic temperature, 577 ○ C. The impact of this alloying process on the reflectance at the back side of the cells has been also investigated, showing that the firing step degrades the back reflectance. Therefore, the use of a stack of dielectric layers seems to be required to avoid spiking of Al through the layer stack and to have the most advantageous optical and passivation properties of this dielectric stack. Both, in-situ microscopy analysis (as function of processing temperature) and cross sectional analysis of the contact formation process give hints that the interaction of molten Al-Si alloy with the underlying dielectric layers (SiNy :H and SiOx ) is likely to form reaction gases like for instance H2 and N2 that escape through the melt from the interface region. At the end, the interaction between the Al-Si alloy during firing and the dielectric layers with the presence of Si in the Al layer have been assessed in terms of rear reflectance losses, identifying the presence of Si as the main cause for this loss by means of transmission electron microscopy images combined with reflectance measurements.

6.2 OUTLOOK

After all the experiments done in this work, as a general guideline for an optimal efficiency output for i-PERC solar cells with the current technology, it is advised to use a distance pitch between openings of 500 µm, in combination with 1 µm Al-Si, optionally capping it with 1 µm Al. All these parameters are valid when using a laser opening around 50 µm in diameter and a resistivity material around 2-3 Ohm.cm. A peak firing temperature between 835 and 885○ C, depending on the front configuration. As a final word, all this work presented has been focused on the fundamental analysis of the interactions between the elements and materials involved during the back side processing of the cells within the i-PERC concept. The link between these interactions and the electrical performance at cell level has been given a mayor attention.

6.2.

Outlook

Beyond this work, there are other possible studies. Some suggestions and viable routes are given in this section as an outlook: • Analysis of the interconnection at module level of the alternative Al-Si PVD metallization. This might show the full potential of the utilization of the proposed solution within the i-PERC technology. In industry, Ag tabs are soldered on the back side of the full Al-BSF cells to interconnect the different cells in the module. The use of new products as the conductive adhesives might be combined with this technology to investigate its performance. • At cell level, the use of this metallization, with its application to thinner cell as the ITRPV roadmap predicts, can result in more

101

102


beneficial as the smaller penetration of the contacts obtained in this work, and it needs to be further investigated. • Other possible cell concepts as the PERL, depicted next to the PERC in chapter 1, in which the doping underneath the contacts is not made by a high temperature step as in PERC, but using diffused localized boron regions. The higher solubility of boron compared to aluminum [128, 129] allows a higher doping level providing a higher field effect of the BSF and therefore a reduced recombination of the minority carriers. Besides that, with this cell concept, the temperature profile used for the formation of the front and rear contact can be adapted separately, giving the possibility to apply different temperature profiles which can improve the contact quality.

APPENDIX A

Scanning Spreading Resistance Microscopy (SSRM) The technique was first reported by Vandervorst [130, 131]. It allows the measurement of the spreading resistance of a given material, being able to measure it independently from the contact resistance. The method explained in this section follows the explanations given by Eyben et al. in [104]. The resistance which is measured is as defined in equation A.1, in similarity with the SRP technique [132] R=

ρ , 4⋅a

(A.1)

where ρ is the resistivity of the material in contact with the probe and a is the electrical radius of the probe. The technique is based in a conductive AFM (Atomic Force Microscopy) [133]. Figure A.1 shows one schematic representation of the technique and the connection made to analyze the samples. The spreading resistance is measured applying a high force on the tip (in µN range) to discriminate it from the contact resistance (figure A.2) [134]. The spatial resolution is given by the radius of the tip of the probe. The hard conductive probe scans the surface under investigation, being the a DC bias applied between this probe and the back side of the sample. The resulting current is measured using a logarithmic current amplifier providing a typical range of 10 pA to 0.1 mA. The probes used are made of silicon coated with doped diamond. These probes, mounted on stiff cantilevers (10 to 100 Nm), are the only

103

APPENDIX A

Conductive)AFM)probe

Sample

lo log(I)

To)ADC

Bias)voltage 50)– 500)mV

Ag)back)contact

Figure A.1: Schematic representation of the SSRM tool

resistance (a.u.)

104

AFM/SCM 10-9

SSRM 10-6

10-3

force(N)

Figure A.2: Resistance measured as a function of the load. After [104]

APPENDIX A

ones that can survive the forces required for SSRM. The modeling using this equation A.1, is not fully precise as such, as there are inaccuracies due to the multiple parameter dependence of the electrical properties of the probe–semiconductor contact. As it is very difficult to express those mathematically, the solution in this case is to compare them with calibrated samples, expressing the difference by a barrier resistance as defined by De Wolf et al. [135]. R=

ρ + Rbarrier (ρ) 4⋅a

(A.2)

Moreover, for non-uniformly doped samples, the resistance value at every position is no longer exclusively determined by the carrier concentration at this same place, as the spreading of the current might be influenced by neighboring conducting or insulating regions (figure A.3). R = CF (a, ρ)

ρ + Rbarrier (ρ), 4⋅a

(A.3)

where CF is a correction factor as proposed by De Wolf et al. in [135], which depends on a and ρ. According to equation A.2, for a fixed probe, the resistance measured is directly proportional to the resistivity of the material in contact, which is, following equation A.4, ρ=

1 , q ⋅ µp ⋅ p

(A.4)

which relates the resistivity in a semiconductor with its doping level and the mobility of the considered carriers. Based on these two equations A.1

105

106

APPENDIX A

Figure A.3: Current distribution (solid lines) and potential contour lines (dashed lines) under a SSRM tip for three extreme situations: (a) nearby isolating boundary, (b) semi-infinite uniform layer, (c) nearby conducting boundary, taken from [135]

and A.4, the technique is able to quantify the doping concentration in a given sample. Due to the small radius of the tip (few nm), the resolution reached by this technique is in the order of 1 to 3 nm, with a dopant gradient resolution of 1-2 nm per decade, values, which are small enough for our application.

APPENDIX B

In-situ microscopic observation I have always imagined the possibility of introducing a camera while performing experiments into a tool to see in live what is happening at micrometer scale. For certain processes, such as the firing step explained in section 1.4, this might tell us a lot about on how the process is really happening. Unfortunately, it was difficult to find an appropriate camera which could withstand such extreme temperatures. Instead, the use of a heating stage that could be installed under a microscope was employed. It looks like the one pictured in figure B.4.

Experimental setup Having this tool, the following experimental setup was made. A cooling circuit was provided in order to maintain the outer case of the tool at room temperature. Like that, during the operation at high temperatures, water is circulating through the heating stage all the time. The sample cut is placed on the ceramic crucible in the center of the tool and covered by a lid with a central window for visualization. The heating stage is then positioned under a confocal microscope to record the images. The heating stage has a maximum heating rate of 200 ºC/min, which although being lower than the firing furnace, it is fast enough to realize the process within minutes.

107

108

APPENDIX B

Figure B.4: Heating stage detail

L I S T O F P U B L I C AT I O N S

Presentations • G. Beaucarne, P. Choulat, Y. Ma, F. Dross, A. Urueña, G. Agostinelli, J. Szlufcik, J. John. Local Al-alloyed contacts for next generation Si solar cells. 1st Metallization Workshop Utrecht, 2008. • P. Eyben, F. Seidel, T. Hantschel, A. Lorenz, A. Urueña, D. Van Gestel, J. John, J. Horzel, W. Vandervorst. “Development and optimization of scanning spreading resistance microscopy for measuring the two-dimensional carrier profile in solar cell structures”. EMRS Spring Meeting, 2010. • J. John, V. Prajapati, B. Vermang, A. Lorenz, C. Allebe, A. Rothschild, L. Tous, A. Urueña, K. Baert, J. Poortmans. “Evolutionary process development towards next generation crystalline silicon solar cells: A semiconductor process toolbox application”. Photovoltaic Technical Conference – PVTC, 2011. • P. Eyben, M.H.P. Pfeiffer, A. Urueña, K. Arstila, W. Vandervorst. “Electrical and morphological characterization at the nanometer scale of aluminum back surface field”. EMRS Spring Meeting, 2012. • J. John, L. Tous, N. Posthuma, F. Duerinckx, E. Cornagliotti, M. Haslinger, K. Wostyn, P. Choulat, M. Recamán, A. Urueña, E. Sleeckx, S. Singh, J. Poortmans. “20.5% conversion efficiency

109

110

LIST OF PUBLICATIONS

on large area n-type PERT cell with copper based front side metallization”. nPV Workshop, Chamberi, 2013.

International peer-reviewed journal contributions • P. Eyben, F. Seidel, T. Hantschel, A. Lorenz, A. Urueña, D. Van Gestel, J. John, J. Horzel, Wilfried Vandervorst. “Development and optimization of scanning spreading resistance microscopy for measuring the two-dimensional carrier profile in solar cell structures”. Phys. Status Solidi A. Volume 208, Issue 3, pages 596–599, 2011. http://dx.doi.org/10.1002/pssa.201000306. • B. Vermang, H. Goverde, A. Urueña, A. Lorenz, E. Cornagliotti, A. Rothschild, J. John, J. Poortmans, R. Mertens. “Blistering in ALD Al2 O3 passivation layers as rear contacting for local Al BSF Si solar cells”. Solar Energy Materials and Solar Cells, volume 101, pages 204–209, 2012. http://dx.doi.org/10.1016/j.solmat.2012.01.032. • A. Urueña, E. Cornagliotti, J. Horzel, I. Kuzma-Filipek, S. Singh, J. John, R. Mertens, J. Poortmans. “Rear contact and BSF formation for local Al-BSF solar cells”. Energy Procedia, volume 27, pages 561-566, 2012. http://dx.doi.org/10.1016/j.egypro.2012.07.110. • L. Tous, M. Alemán, H. Bender, J. Das, T. Emeraud, J. John, J. Lerat, J. Meersschaut, R. Mertens, J. Poortmans, R. Russell, A. Urueña. “Nickel silicide formation using excimer laser annealing”. Energy Procedia, volume 27, pages 503-509, 2012. http://dx.doi.org/10.1016/j.egypro.2012.07.101. • R. De Rose, K. Van Wichelen, L. Tous, J. Das, F. Dross, C. Fiegna, M. Lanuzza, E. Sangiorgi, A. Urueña, M. Zanuccoli.


“Optimization of rear point contact geometry by means of 3-D numerical simulation”. Energy Procedia, volume 27, pages 197-202, 2012. http://dx.doi.org/10.1016/j.egypro.2012.07.051 • J. John, V. Prajapati, B. Vermang, A. Lorenz, C. Allebe, A. Rothschild, L. Tous, A. Urueña, K. Baert, J. Poortmans. “Evolutionary process development towards next generation crystalline silicon solar cells: A semiconductor process toolbox application”. EPJ Photovoltaics, volume 3, pages 35005-35012, 2012. http://dx.doi.org/10.1051/epjpv/2012009. • A. Urueña, J. Horzel, P. Eyben, J. John, J. Poortmans, R. Mertens. “Interaction between Al-Si melt and dielectric layers during formation of local Al-alloyed contacts for rear passivated Si solar cells”. Phys. Status Solidi A 209, No. 12, pages 2615–2619 (2012). http://dx.doi.org/10.1002/pssa.201228353. • L. Tous, J.F. Lerat, T. Emeraud, R. Negru, K. Huet, A. Urueña, M. Alemán, J. Meersschaut, H. Bender, R. Russell, J. John, J. Poortmans, R. Mertens. “Nickel silicide contacts formed by excimer laser annealing for high efficiency solar cells”. Progress in Photovoltaics. 2013. http://dx.doi.org/10.1002/pip.2362. • A. Urueña, L. Tous, I. Kuzma-Filipek, E. Cornagliotti, F. Duerinckx, J. John, R. Mertens, and J. Poortmans. “Understanding the mechanisms of rear reflectance losses in PERC type solar cells”. Energy Procedia, 2013. http://dx.doi.org/10.1016/j.egypro.2013.07.349. • P. Eyben, M. P. Pfeiffer, A. Urueña, W. Vandervorst. “Electrical and morphological characterization at the nanometer scale of aluminum back surface field”. In preparation.

111

112


Conference proceedings • A. Urueña, J. John, G. Beaucarne, P. Choulat, P. Eyben, G. Agostinelli, E. van Kerschaver, J. Poortmans, R. Mertens. “Local Al-alloyed contacts for next generation Si solar cells”. Proceedings of the 24th European Photovoltaic Solar Energy Conference, pages 1483 - 1486, 2009. http://doi.org/10.4229/24thEUPVSEC20092CV.2.22. • D. Trusheim, M. Schulz-ruhtenberg, J.L. Hernández, A. Urueña, S. Krantz, C. Morilla, T.Sarnet, A. Olowinsky. “Damage-less laser ablation of thin films for silicon solar cells”. Proceedings of 5th Lasers in Manufacturing Conference 2009, pages 647-652. http://publica.fraunhofer.de/dokumente/N-113307.html. • J. John, V. Prajapati, C. Allebé, A. Urueña, J.L. Hernandez, B. Vermang, A. Rothschild, A. Lorenz, B.T. Chan, K. Baert, J. Poortmans. “A process technology toolbox for next generation large area crystalline silicon solar cells”. Proceedings of the 35th IEEE Photovoltaic Specialists Conference, 2010. http://dx.doi.org/10.1109/PVSC.2010.5616830. • A. Urueña, J.L. Hernández, J. John, J. Poortmans, R. Mertens. “Controlling the depth of the local Al BSF in PERC crystalline solar cells using alternative back side metallization”. Proceedings of the 25th European Photovoltaic Solar Energy Conference, pages 2562 - 2564, 2010. http://dx.doi.org/10.4229/25thEUPVSEC20102DV.1.40. • C. Allebé, E. Schlenker, J.L. Hernández, M. Alemán, L. Tous, V. Prajapati, K. Van Nieuwenhuysen, T. Janssens, A. Rothschild, B. Vermang, A. Lorenz, A. Urueña, E.


Rosseel, J. John, K. Baert, J. Poortmans. “Process integration towards PERL structure”. Proceedings of the 25th European Photovoltaic Solar Energy Conference, pages 1469 - 1474, 2010. http://doi.org/10.4229/25thEUPVSEC2010-2DO.3.1. • A. Urueña, J. John, P. Eyben, D. Vanhaeren, T. Werner, T. Hantschel, W. Vandervorst, J. Poortmans, R. Mertens. “Studying local aluminum back surface fields (Al-BSF) contacts through scanning spreading resistance microscopy (SSRM)”. Proceedings of the 26th European Photovoltaic Solar Energy Conference, pages 1530 - 1533, 2011. http://doi.org/10.4229/26thEUPVSEC20112BV.2.38. • B. Vermang, H. Goverde, A. Lorenz, A. Urueña, J. Das, P. Choulat, E. Cornagliotti, A. Rothschild, J. John, J. Poortmans, R. Mertens. “On the blistering of Al2 O3 passivation layers for Si solar cells”. Proceedings of the 26th European Photovoltaic Solar Energy Conference, pages 1129 - 1131, 2011. http://doi.org/10.4229/26thEUPVSEC2011-2DO.1.4. • P. Jaffrennou, A. Urueña, J. Das, J. Penaud, M. Moors, A. Rothschild, B. Lombardet, J. Szlufcik. “Laser ablation of SiO2 /SiNx and AlOx /SiNx back side passivation stacks for advanced cell architectures”. Proceedings of the 26th European Photovoltaic Solar Energy Conference, pages 2180 - 2183, 2011. http://doi.org/10.4229/26thEUPVSEC2011-2CV.4.15. • J. Horzel, A. Lorenz, E. Cornagliotti, A. Urueña, J. John, M. Izaaryene, D. Habermann, P. Jaffrennou, J. Penaud. “Development of Rear Side Polishing Adapted to Advanced Solar Cell Concepts”. Proceedings of the 26th European Photovoltaic Solar Energy Conference, pages 2210 - 2216, 2011. http://doi.org/10.4229/26thEUPVSEC2011-2CV.4.25.

113

114


• R. Labie, J.L. Hernández, J. Govaerts, C. Allebe, L. Tous, A. Urueña, R. Russell, I. Gordon, K. Baert. “Cu plated i-PERL cells: Ageing and humidity reliability tests”. Proceedings of the 26th European Photovoltaic Solar Energy Conference, pages 1195 - 1198, 2011. http://doi.org/10.4229/26thEUPVSEC2011-2EO.2.1. • B. Vermang, H. Goverde, A. Lorenz, A. Urueña, G. Vereecke, J. Meersschaut, E. Cornagliotti, A. Rothschild, J. John, J. Poortmans, R. Mertens. “On the blistering of atomic layer deposited Al2 O3 as Si surface passivation”. Proceedings of the 37th IEEE Photovoltaic Specialists Conference, 2011. http://dx.doi.org/10.1109/PVSC.2011.6185916. • P. Jaffrennou, M. Moors, A. Urueña, J. Das, F. Duerinckx, J. Penaud, A. Rothschild, B. Lombardet, J. Szlufcik. “Laser ablation of AlOx and AlOx /SiNx backside passivation layers for advanced cell architectures”. Proceedings of the 37th IEEE Photovoltaic Specialists Conference, 2011. http://dx.doi.org/10.1109/PVSC.2011.6186138. • A. Urueña, J.F. Lerat, T. Emeraud, J. Meersschaut, J. John, J. Poortmans, R. Mertens. “Studying the properties of laser annealed AlSi alloys”. Proceedings of the 27th European Photovoltaic Solar Energy Conference, pages 1725 - 1727, 2012. http://dx.doi.org/10.4229/27thEUPVSEC2012-2CV.5.30. • A. Rothschild, J. Penaud, J. Toman, P. Jaffrennou, P. Choulat, E. Cornagliotti, M. Recaman Payo, B. Pawlak, J. Das, A. Urueña, S. Singh, J. Horzel. “Impact of surface preparation prior to ALDAl2 O3 deposition for PERC type solar cell”. Proceedings of the 27th European Photovoltaic Solar Energy Conference, pages 1974 - 1977, 2012. http://dx.doi.org/10.4229/27thEUPVSEC2012-2CV.6.60.


• E. Cornagliotti, A. Urueña, J. Horzel, J. John, L. Tous, D. Hendrickx, V. Prajapati, S. Singh, R. Hoyer, F. Delahaye, K. Weise, D. Queisser, H. Nussbaumer, J. Poortmans. “How much rear side polishing is required? A study on the impact of rear side polishing in PERC solar cells”. Proceedings of the 27th European Photovoltaic Solar Energy Conference, pages 561 - 566, 2012. http://dx.doi.org/10.4229/27thEUPVSEC2012-2AO.1.6. • F. Duerinckx, M. Moors, T. Caremans, K. Baert, A. Cacciato, G. Leys, M. Mrcarica, E. Picard, J. Szlufcik, V. Prajapati, P. Choulat, A. Urueña, J. Horzel. “Accumulation versus inversion layer passivation on +19% screenprinted industrial crystalline silicon solar cells”. Proceedings of the 27th European Photovoltaic Solar Energy Conference, pages 1630 - 1633, 2012. http://dx.doi.org/10.4229/27thEUPVSEC2012-2BV.6.49. • J. Horzel, L. Tous, A. Urueña, A. Seidl, R. Russell, E. Cornagliotti, “Cz-Si material influence on PERL-Type Si solar cells”. Proceedings of the 27th European Photovoltaic Solar Energy Conference, pages 780 - 788, 2012. http://dx.doi.org/10.4229/27thEUPVSEC20122CO.15.6. • J. Horzel, P. Choulat, E. Cornagliotti, T. Janssens, J. John, I. Kuzma-Filipek, V. Prajapati, A. Rothschild, R. Russell, S. Singh, E. Sleeckx, L. Tous, A. Urueña, B. Vermang. “Overview on recent improvement for industrially applicable PERL-type Si solar cell processing”. Proceedings of the 27th European Photovoltaic Solar Energy Conference, pages 1602 - 1606, 2012. http://dx.doi.org/10.4229/27thEUPVSEC2012-2BV.6.40. • A. Urueña, J. F. Lerat, T. Emeraud, L. Tous, F. Duerinckx, J. John, J. Poortmans, R. Mertens. “Excimer laser annealing for boron implanted local BSF for PERC crystalline silicon solar

115

116


cells”. Proceedings of the 28th European Photovoltaic Solar Energy Conference, 2013. Accepted. • B.J. O’Sullivan, M. Debucquoy, S. Singh, A. Urueña, M. Recaman Payo, N. Posthuma, J. Poortmans. “Process simplification for high efficiency, small area interdigitated back contact silicon solar cells”. Proceedings of the 28th European Photovoltaic Solar Energy Conference, 2013. Accepted. • I. Kuzma-Filipek, S. Asad, G. Beaucarne, F. Campeol, P. Descamps, F. Duerinckx, J. Horzel, V. Kaiser, P. Leempoel, A. Urueña and J. Poortmans. “Alternative oxide technologies benchmarked with PECVD oxide as rear dielectric for i-PERC cell architecture”. Proceedings of the 28th European Photovoltaic Solar Energy Conference, 2013. Accepted. • M. Recaman, N. Posthuma, M. Debucquoy, A. Urueña, K. Warikoo and J. Poortmans “Boron doped selective silicon epitaxy: high efficiency and process simplification in IBC cells”. Proceedings of the 28th European Photovoltaic Solar Energy Conference, 2013. Accepted. • M. Aleman, L. Tous, S. Sukhvinder, E. Cornagliotti, J. John, R. Russell, E. Sleeckx, A. Urueña, F. Duerinckx, N.E. Posthuma, J. Szlufcik. “Large area high-efficiency n-type Si rear junction cells featuring laser ablation and Cu plated front contacts”. Proceedings of the 28th European Photovoltaic Solar Energy Conference, 2013. Accepted.


Patents • Method for the fabrication of back-contacted photovoltaic cells. Patent number WO 2013/020867. • Method for fabricating photovoltaic cells with plated contacts. Patent number EP 2013/055093. • Method of manufacturing a solar cell with local back contacts. Patent number WO 2013/050556 - EP 2579317 (A1).

Press journal articles • F. Duerinckx, E. Cornagliotti, V. Prajapati, A. Urueña, P. Choulat, P. Pieters and J. Poortmans. “i-PERC technology enables Si solar cell efficiencies beyond 20%”. Photovoltaics International 19th Edition, February, pages 44-50, 2013.

117

BIBLIOGRAPHY

[1] http://www.enerdata.net, accessed on May 2013. (Cited on page 1) [2] http://www.bp.com/liveassets, accessed on May 2013. (Cited on page 1) [3] http://www.nrel.gov/ncpv/images/efficiency_chart.jpg, accessed on August 2013. (Cited on page 2) [4] C. E. Fritts. On a new form of selenium photocell. In Proceedings of the American Association for the Advancement of Science, page 97, 1883. (Cited on page 2) [5] H. Hertz. Ueber einen Einfluss des ultravioletten Lichtes auf die electrische Entladung. Annalen der Physik, 267:983–1000, 1887. (Cited on page 3) [6] D.M. Chapin, C.S. Fuller, and G.L. Pearson. A new p-n junction photocell for converting solar radiation into electrical power. Journal of Applied Physics, 25:676–677, 1954. (Cited on page 3) [7] http://www.pvinsights.com, accessed on May 2013. page 3)

(Cited on

[8] A.W. Blakers, A. Wang, A.M. Milne, J. Zhao, and M.A. Green. 22.8 % efficient silicon solar cell. Applied Physics Letters, 55(13):1363, 1989. (Cited on pages 5, 6, and 7) [9] J. Zhao, A. Wang, P. Altermatt, and M.A. Green. Twenty-four percent efficient silicon solar cells with double layer antireflection

119

120

BIBLIOGRAPHY

coatings and reduced resistance loss. 66:3636, 1995. (Cited on page 6)

Applied Physics Letters,

[10] J. Zhao, A. Wang, and M.A. Green. 24 percent efficient PERL structure silicon solar cells. In Photovoltaic Specialists Conference, 1990, Conference Record of the Twenty First IEEE, volume 1, pages 333 –335, 1990. (Cited on pages 5 and 6) [11] G. Agostinelli, P. Choulat, H.F.W. Dekkers, E. Vermariën, and G. Beaucarne. Rear surface passivation for industrial solar cells on thin substrates. In 4th IEEE World Conference on Photovoltaic Energy Conversion, Hawaii, 2006. (Cited on page 6) [12] V. Meemongkolkiat, K. Nakayashiki, D.S. Kim, S. Kim, A. Shaikh, A. Kuebelbeck, W. Stockum, and A. Rohatgi. Investigation of modified screen-printing Al pastes for local back surface field formation. In 4th IEEE World Conference on Photovoltaic Energy Conversion; Hawaii, USA, 2006. (Cited on pages 6 and 73) [13] F. Huster. Aluminium - back surface field - bow investigation and elimination. In 20th European Photovoltaic Solar Energy Conference and Exhibition, 2005. (Cited on page 6) [14] Y.F. Wang, W. Wu, P.R. Li, L. Zhang, and Z.Q. Ma. Influence of cooling on the performance of silicon solar cells. In IEEE Proceedings of 16th IPFA, China, 2009. (Cited on page 6) [15] G. Agostinelli, P. Choulat, H.F.W. Dekkers, S. De Wolf, and G. Beaucarne. Screen printed large area crystalline silicon solar cells on thin substrates. In Proceedings of the 20th European Photovoltaic Solar Energy Conference, pages 647–650, 2005. (Cited on page 7) [16] G. Agostinelli, P. Choulat, H.F.W. Dekkers, Y. Ma, and G. Beaucarne. Silicon solar cells on ultra-thin substrates for large

BIBLIOGRAPHY

scale production. In Proceedings of the 21st European Photovoltaic Solar Energy Conference, volume 4, 2006. (Cited on page 7) [17] F. Heinemeyer, C. Mader, D. Münster, T. Dullweber, and R. Brendel. Inline high-rate thermal evaporation of aluminum for novel industrial solar cell metallization. In Proceedings of the 25th European Photovoltaic Solar Energy Conference, pages 2066 – 2068, 2010. (Cited on page 7) [18] M. J. Kerr. Surface, Emitter and Bulk Recombination in Silicon and Development of Silicon Nitride Passivated Solar Cells. PhD thesis, The Australian National University, 2002. (Cited on page 7) [19] T. Tiedje, E. Yablonovitch, G. D. Cody, and B. G. Brooks. Limiting efficiency of silicon solar cells. Electron Devices, IEEE Transactions on, 31(5):711–716, 1984. (Cited on page 8) [20] W. Michaelis and M. H. Pilkuhn. Radiative recombination in silicon p-n junctions. physica status solidi (b), 36(1):311–319, 1969. (Cited on page 8) [21] H. Schlangenotto, H. Maeder, and W. Gerlach. Temperature dependence of the radiative recombination coefficient in silicon. physica status solidi (a), 21(1):357–367, 1974. (Cited on page 8) [22] M. A. Green. Silicon solar cells: Advanced principles and practice. Bridge printery, 1995. (Cited on pages 8 and 11) [23] A. Hangleiter and R. Häcker. Enhancement of band-to-band auger recombination by electron-hole correlations. Phys. Rev. Lett., 65:215–218, Jul 1990. (Cited on page 9) [24] M.S. Tyagi and R. Van Overstraeten. Minority carrier recombination in heavily-doped silicon. Solid-State Electronics, 26(6):577 – 597, 1983. (Cited on page 9)

121

122

BIBLIOGRAPHY

[25] M. A. Green. Solar cells: Operating principles, technology, and system applications. 1982. (Cited on page 10) [26] W. Shockley and WT Read. Statistics of the recombinations of holes and electrons. Physical Review, 87(5):835–842, 1952. (Cited on page 10) [27] R. N. Hall. Electron-hole recombination in germanium. Phys. Rev., 87:387–387, Jul 1952. (Cited on page 10) [28] Y. Chen, H. Shen, and P. Altermatt. Analysis of recombination losses in screen-printed aluminum induced back surface fields of silicon solar cells by numerical device simulation. Solar Energy Materials and Solar Cells, In press, 2013. (Cited on pages 13 and 14) [29] S. M. Sze. Physics of Semiconductor Devices. Wiley, 1981. (Cited on page 17) [30] G. Aijuan, Y. Famin, G. Lihui, J. Dong, and F. Shimeng. Effect of the back surface topography on the efficiency in silicon solar cells. Journal of Semiconductors, 30:1–3, 2009. (Cited on pages 17 and 18) [31] H. W. Fritts and E. D. Verink. Aluminum alloys. Industrial & Engineering Chemistry, 43(10):2197–2199, 1951. (Cited on page 21) [32] S. Wang, C. Z. Wang, F. C. Chuang, J. R. Morris, and K. M. Ho. Ab initio molecular dynamics simulation of liquid Al88 Si12 alloys. The Journal of Chemical Physics, 122(3):034508, 2005. (Cited on pages 21 and 37) [33] W.D. Callister. Materials Science and Engineering. Wiley, 2007. (Cited on pages 21, 30, and 32)

BIBLIOGRAPHY

[34] T. Chung. Study of aluminum fusion into silicon. Journal of the Electrochemical Society, 109:229–234, 1962. (Cited on pages 22 and 49) [35] J. O. McCaldin and H. Sankur. Diffusivity and solubility of Si in the Al metallization of integrated circuits. Applied Physics Letters, 19(12):524 –527, 1971. (Cited on page 22) [36] M. Hansen, R. P. Elliott, F. A. Shunk, and Research Institute. Constitution of binary alloys. 1958. (Cited on page 22) [37] J. O. McCaldin and H. Sankur. Precipitation of Si from the Al metallization of integrated circuits. Applied Physics Letters, 20(4):171 –172, 1972. (Cited on page 22) [38] G.J. Van Gurp. Diffusion-limited Si precipitation in evaporated Al/Si films. Journal of Applied Physics, 44:2040, 1973. (Cited on page 22) [39] http://www.itrpv.net, accessed on January 2013. page 22)

(Cited on

[40] J. Mandelkorn and J.H. Lamneck. Simplified fabrication of back surface electric field silicon cells and novel characteristics of such cells. Solar Cells, 29(2-3):121–130, 1972, reprinted 1990. (Cited on pages 22 and 23) [41] H.F. Wolf. Silicon semiconductor data. International series of monographs on semiconductors. Pergamon Press, 1969. (Cited on page 23) [42] J.L. Murray and A.J. McAllister. The Al-Si (aluminium-silicon) system. Bulletin of alloy phase diagrams, 5, no.1:74–112, 1984. (Cited on pages 23, 24, 26, 27, 32, 35, 49, and 51)

123

124

BIBLIOGRAPHY

[43] A. Uruena, J. John, G. Beaucarne, P. Choulat, P. Eyben, G. Agostinelli, E. van Kerschaver, J. Poortmans, and R. Mertens. Local Al-alloyed contacts for next generation Si solar cells. In Proceedings of the 24th European Photovoltaic Solar Energy Conference, pages 1483–1486, 2009. (Cited on pages 23, 33, and 50) [44] A. Uruena, J.L. Hernández, J. John, J. Poortmans, and R. Mertens. Controlling the depth of the local Al-BSF in PERC crystalline solar cells using alternative back side metallization. In Proceedings of the 25th European Photovoltaic Solar Energy Conference, pages 2562 – 2564, 2010. [45] E. Urrejola, K. Peter, H. Plagwitz, and G. Schubert. Silicon diffusion in aluminum for rear passivated solar cells. Applied Physics Letters, 98(15):153508, 2011. (Cited on pages 43 and 48) [46] E. Urrejola, K. Peter, H. Plagwitz, and G. Schubert. Al-Si alloy formation in narrow p-type Si contact areas for rear passivated solar cells. Journal of Applied Physics, 107(12):124516, 2010. (Cited on pages 28, 48, and 71) [47] T. Lauermann, T. Luder, S. Scholz, B. Raabe, G. Hahn, and B. Terheiden. Enabling dielectric rear side passivation for industrial mass production by developing lean printing-based solar cell processes. In Proceedings of the 35th IEEE Photovoltaic Specialists Conference, 2010. (Cited on pages 28, 72, and 73) [48] M. Rauer, R. Woehl, K. Ruhle, C. Schmiga, M. Hermle, M. Horteis, and D. Biro. Aluminum alloying in local contact areas on dielectrically passivated rear surfaces of silicon solar cells. Electron Device Letters, IEEE, 32(7):916 –918, 2011. (Cited on pages 37 and 48) [49] M. Rauer, C. Schmiga, M. Hermle, and S. Glunz. Passivation of screen-printed aluminium-alloyed emitters for back junction n-type

BIBLIOGRAPHY

silicon solar cells. In Proceedings of the 24th European Photovoltaic Solar Energy Conference, 2009. [50] J. Krause, R. Woehl, M. Rauer, C. Schmiga, J. Wilde, and D. Biro. Microstructural and electrical properties of differentsized aluminum-alloyed contacts and their layer system on silicon surfaces. Solar Energy Materials & Solar Cells, 95:2151–2160, 2011. [51] J. Müller, K. Bothe, S. Gatz, and R. Brendel. Modeling the formation of local highly aluminum-doped silicon regions by rapid thermal annealing of screen-printed aluminum. physica status solidi (RRL) – Rapid Research Letters, 6(3):111–113, 2012. (Cited on pages 23, 34, and 48) [52] S. Baumann, D. Kray, K. Mayer, A. Eyer, and G. P. Willeke. Comparative study of laser induced damage in silicon wafers. In Conference Record of the 2006 IEEE 4th World Conference on Photovoltaic Energy Conversion, volume 1, pages 1142–1145. IEEE, May 2006. (Cited on page 27) [53] D. Kray, S. Baumann, K. Mayer, A. Eyer, and G.P. Willeke. Novel techniques for low-damage microstructuring of silicon. In Proceedings of the 20th European Photovoltaic Solar Energy Conference, pages 156–159, 2005. (Cited on page 27) [54] D. Kray, M. Aleman, A. Fell, S. Hopman, K. Mayer, M. Mesec, R. Muller, G.P. Willeke, S.W. Glunz, B. Bitnar, D.-H. Neuhaus, R. Ludemann, T. Schlenker, D. Manz, A. Bentzen, E. Sauar, A. Pauchard, and B. Richerzhagen. Laser-doped silicon solar cells by laser chemical processing (LCP) exceeding 20% efficiency. In Proceedings of the 33th IEEE Photovoltaic Specialists Conference, pages 1–3, 2008. (Cited on page 27)

125

126

BIBLIOGRAPHY

[55] B. Richerzhagen. Entwicklung und Konstruktion eines Systems zur Uebertragung von Laserenergie fuer die Laserzahnbehandlung. PhD thesis, Lausanne, 1994. (Cited on page 27) [56] K. H. Yang. An etch for delineation of defects in silicon. Journal of The Electrochemical Society, 131(5):1140–1145, 1984. (Cited on page 27) [57] A. Knorz, A. Grohe, C. Harmel, R. Preu, J. Luther, and F. Publica. Progress in selective laser ablation of dielectric layers. In Proceedings of the 22nd European Photovoltaic Solar Energy Conference, 2007. (Cited on page 27) [58] P. Engelhart, N. P Harder, T. Horstmann, R. Grischke, R. Meyer, and R. Brendel. Laser ablation of passivating SiNx layers for locally contacting emitters of high-efficiency solar cells. In Conference Record of the 2006 4th World Conference on Photovoltaic Energy Conversion, volume 1, pages 1024–1027. IEEE, May 2006. [59] A. Grohe, C. Harmel, A. Knorz, S.W. Glunz, R. Preu, and G.P. Willeke. Selective laser ablation of anti-reflection coatings for novel metallization techniques. In Conference Record of the 2006 IEEE 4th World Conference on Photovoltaic Energy Conversion, volume 2, pages 1399 –1402, may 2006. (Cited on page 27) [60] Z.R. Du, N. Palina, J. Chen, F. Lin, M.H. Hong, and B. Hoex. Impact of KOH etching on laser damage removal and contact formation for Al local back surface field silicon wafer solar cells. In Proceedings of the 27th European Photovoltaic Solar Energy Conference, 2012. (Cited on page 27) [61] F.S. Grasso, J. Rentsch L. Gautero and, R. Preu, and R. Lanzafame. Characterisation of local Al-BSF formation for PERC solar cell structures. In Proceedings of the 25th European Photovoltaic Solar

BIBLIOGRAPHY

Energy Conference, pages 371 – 374, 2010. (Cited on pages 28 and 73) [62] T. Lauermann, A. Dastgheib-Shirazi, F. Book, B. Raabe, G. Hahn, H. Haverkamp, D. Habermann, C. Demberger, and C. Schmid. InSECT: An inline selective emitter concept with high efficiencies at competitive process costs improved with inkjet masking technology. In Proceedings of the 24th European Photovoltaic Solar Energy Conference, page 1767, 2009. (Cited on page 28) [63] A.J. Lennon, R.Y. Utama, M.A.T.Anita Lenio, W.Y. Ho-Baillie, N.B. Kuepper, and S.R. Wenham. Forming openings to semiconductor layers of silicon solar cells by inkjet printing. Solar Energy Materials & Solar Cells, 92:1410 – 1415, 2008. (Cited on page 28) [64] T. C. Röder, E. Hoffmann, J. R. Kohler, and J. H. Werner. 30 µm wide contacts on silicon cells by laser transfer. In Proceedings of the 35th IEEE Photovoltaic Specialists Conference, 2010. (Cited on page 28) [65] E. Schneiderlöchner, R. Preu, R. Lüdemann, S. W. Glunz, and G. Willeke. Laser-fired contacts (LFC). In 17th European Photovoltaic Solar Energy Conference, 2001. (Cited on page 28) [66] E. Schneiderlöchner, R. Preu, R. Lüdemann, and SW Glunz. Laserfired rear contacts for crystalline silicon solar cells. Progress in Photovoltaics: Research and Applications, 10(1):29–34, 2002. (Cited on page 28) [67] A. Urueña, J.F. Lerat, T. Emeraud, J. Meersschaut, J. John, J. Poortmans, and R. Mertens. Studying the properties of laser annealed AlSi alloys. In Proceedings of the 27th European Photovoltaic Solar Energy Conference, pages 1725 – 1727, 2012. (Cited on page 28)

127

128

BIBLIOGRAPHY

[68] B. Thaidigsmann, C. Kick, A. Drews, F. Clement, A. Wolf, and D. Biro. Fire-through contacts: a new approach to contact the rear side of passivated silicon solar cells. Solar Energy Materials and Solar Cells, 108:164 – 169, 2013. (Cited on page 28) [69] B.-J. Oh, I.-H. Yeo, and D.-G. Lim. Co-firing optimization of crystalline silicon solar cell using rapid thermal process. Journal of the Korean Institute of Electrical and Electronic Material Engineers, 25:236 – 240, 2012. (Cited on page 31) [70] D.L. Bätzner, K. Hanton, A. Li, and A. Cuevas. Investigation of the co-firing process for contact formation in screen printed silicon solar cells. In 21st European Photovoltaic Solar Energy Conference, 2006. (Cited on page 31) [71] V. Meemongkolkiat, K. Nakayashiki, D.S. Kim, R. Kopecek, and A. Rohatgi. Factors Limiting the Formation of Uniform and Thick Aluminum–Back-Surface Field and Its Potential. Journal of the Electrochemical Society, 153:G53–G58, 2006. (Cited on page 32) [72] A. Kaminski, B. Vandelle, A. Fave, JP Boyeaux, L.Q. Nam, R. Monna, D. Sarti, and A. Laugier. Aluminium BSF in silicon solar cells. Solar Energy Materials and Solar Cells, 72(1-4):373– 379, 2002. (Cited on pages 32 and 38) [73] S. Narasimha, A. Rohatgi, and A. W Weeber. An optimized rapid aluminum back surface field technique for silicon solar cells. Transactions on Electron Devices, 46(7):1363–1370, July 1999. (Cited on page 32) [74] P. Loelgen. Surface and volume recombination in silicon solar cells. PhD thesis, University of Utrecht, 1995. (Cited on page 32) [75] F. Huster. Investigation of the alloying process of screen printed aluminium pastes for the BSF formation on silicon solar cells.

BIBLIOGRAPHY

In 20th European Photovoltaic Solar Energy Conference and Exhibition, 2005. (Cited on page 33) [76] G. Beaucarne. Local Al-alloyed contacts for next generation Si solar cells. 1st Metallization Workshop Utrecht, 2008. (Cited on page 33) [77] J. Del Alamo, J. Eguren, and A. Luque. Operating limits of Al-alloyed high-low junctions for BSF solar cells. Solid State Electronics, 24:415–420, 1980. (Cited on page 33) [78] C.P. Sealy, M.R. Castell, and P.R. Wilshaw. Mechanism for secondary electron dopant contrast in the SE. Journal of Electron Microscopy, 49 (2):311–321, 2000. (Cited on pages 35 and 41) [79] S.W. Jones. Diffusion in silicon. IC Knowledge LCC, 2008. (Cited on page 35) [80] F. Gao, H. Zhao, D.P. Sekulic, Y. Qian, and L. Walker. Solid state Si diffusion and joint formation involving aluminum brazing sheet. Materials Science and Engineering: A, 337(1-2):228 – 235, 2002. (Cited on page 37) [81] Y. Mishin, C. Herzig, J. Bernardini, and W. Gust. Grain boundary diffusion: fundamentals to recent developments. International Materials Reviews, 42(4):155–178, 1997-01-01T00:00:00. [82] J.R. Terrill. Diffusion of silicon in aluminum brazing sheet. Welding Journal, 45(5):202S – 209S, 1966. (Cited on page 37) [83] A. Uruena, J. John, P. Eyben, D. Vanhaeren, T. Werner, T. Hantschel, W. Vandervorst, J. Poortmans, and R. Mertens. Studying local aluminum back surface fields (Al-BSF) contacts through scanning spreading resistance microscopy (SSRM). In Proceedings of the 26th European Photovoltaic Solar Energy Conference, pages 1530 – 1533, 2011. (Cited on page 38)

129

130

BIBLIOGRAPHY

[84] G. Hahn, P. Geiger, and G. Schubert. Influence of BSF thickness and Al-gettering steps on IQE and cell parameters in mc Si solar cells. In Proceedings of the 19th European Photovoltaic Solar Energy Conference, 2004. (Cited on page 38) [85] M. Rauer, C. Schmiga, M. Glatthaar, and S.W. Glunz. Alloying from screen-printed aluminum pastes containing boron additives. IEEE Journal of Photovoltaics, 3(1):206 –211, 2013. [86] A. Kränzl, A. Schneider, I. Melnyk, A. Hauser, E. Rüland, and P. Fath. Different aspects of back-surface field (BSF) formation for thin multi-crystalline silicon wafers. Asian J. Energy Environ, 5(4):275 – 283, 2004. (Cited on page 38) [87] J.A. Amick, F.J. Bottari, and J.I. Hanoka. The effect of aluminum thickness on solar cell performance. J. Electrochem. Soc., Vol. 141, No. 6, June 1994, 141:1577–1585, 1994. (Cited on page 38) [88] G. Agostinelli, J. Szlufcik, P. Choulat, and G. Beaucarne. Local contact structures for industrial perc-type solar cells. In Proceedings of the 20th European Photovoltaic Solar Energy Conference, pages 942–945, 2005. (Cited on page 41) [89] G. Agostinelli, G. Beaucarne, and P. Choulat. Photovoltaic cell with thick silicon oxide and silicon nitride passivation fabrication, 2006. (Cited on page 41) [90] Y. Pauleau. Microelectronic Materials and Processes. Springer, 1989. (Cited on pages 41 and 71) [91] A.D. Smigelskas and E.O. Kirkendall. Zinc diffusion in alpha brass. Trans. Aime, 171:130 – 142, 1947. (Cited on page 41) [92] H. Nakajima. The discovery and acceptance of the Kirkendall effect - The result of a short research career. JOM, 49(6):15–19, June 1997. (Cited on pages 41 and 42)

BIBLIOGRAPHY

[93] A. Paul, M.J.H. Van Dal, A.A. Kodentsov, and F.J.J. Van Loo. The Kirkendall effect in multiphase diffusion. Acta materialia, 52(3):623 – 630, 2004. (Cited on page 42) [94] A. Paul. The Kirkendall effect in solid state diffusion. PhD thesis, 2004. (Cited on page 42) [95] E. Urrejola, K. Peter, H. Plagwitz, and G. Schubert. Effect of gravity on the microstructure of Al-Si alloy for rear-passivated solar cells. Journal of Applied Physics, 110:056104, 2011. (Cited on page 42) [96] E. Urrejola. Aluminum-silicon contact formation through narrow dielectric openings: Application to industrial rear passivated high efficiency solar cells. PhD thesis, Konstanz University, 2012. (Cited on pages 42 and 85) [97] J. Goldstein. Scanning Electron Microscopy and X-Ray Microanalysis. Springer London, Limited, 2003. (Cited on page 43) [98] C. E. Roberts. CXXIX - The alloys of aluminium and silicon. J. Chem. Soc., Trans., 105:1383–1386, 1914. (Cited on page 48) [99] W. Fraenkel. Metallographische Mitteilungen aus dem Institut fuer physikalische Chemie der Universitaet Goettingen. LXIII. Ueber Silicium-Aluminiumlegierungen. Zeitschrift fuer Anorganische Chemie, 58(1):154 – 158, 1908. (Cited on page 48) [100] J.G. Fossum. Physical operation of back-surface-field silicon solar cells. Electron Devices, IEEE Transactions on, 24(4):322 – 325, apr 1977. (Cited on page 48) [101] L. Sardi, S. Bargioni, C. Canali, P. Davoli, M. Prudenziati, and V. Valbusa. Some features of thick film technology for the back metallization of solar cells. Solar Cells, 11(1):51 – 67, 1984. (Cited on page 48)

131

132

BIBLIOGRAPHY

[102] C. Mader, J. Muller, S. Gatz, S. Dullweber, and R. Brendel. Rearside point-contacts by inline thermal evaporation of aluminum. In Proceedings of the 35th IEEE Photovoltaic Specialists Conference, 2010. (Cited on page 49) [103] J. A. Aboaf, D. R. Kerr, and E. Bassous. Charge in SiO2 - Al2 O3 double layers on silicon. Journal of The Electrochemical Society, 120(8):1103–1106, 1973. (Cited on pages 54 and 56) [104] P. Eyben, W. Vandervorst, D. Alvarez, M. Xu, and M. Fouchier. Scanning Probe Microscopy. Springer, 2006. (Cited on pages 54, 103, and 104) [105] P. De Wolf, T. Clarysse, W. Vandervorst, L. Hellemans, P. Niedermann, and W. Haenni. Cross-sectional nano-spreading resistance profiling. Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures, 16:355, 1998. (Cited on page 54) [106] V. Prajapati, E. Cornagliotti, A. Lorenz, B. Vermang, J. John, J. Poortmans, and R. Mertens. Illumination level independence in relation to emitter profiles on industrial high efficiency local AlBSF cells. In Proceedings of the 26th European Photovoltaic Solar Energy Conference, pages 1052 – 1057, 2011. (Cited on page 56) [107] B. Vermang, E. Cornagliotti, V. Prajapati, J. John, J. Poortmans, and R. Mertens. Assessment of the illumination dependency of Al2 O3 and SiO2 rear-passivated p-type silicon solar cells. physica status solidi (RRL) – Rapid Research Letters, 6(6):259–261, 2012. (Cited on page 56) [108] B. Fischer. Loss analysis of crystalline silicon solar cells using photoconductance and quantum efficiency measurements. PhD thesis, Cuvillier, 2003. (Cited on page 59)

BIBLIOGRAPHY

[109] H. Plagwitz and R. Brendel. Analytical model for the diode saturation current of point-contacted solar cells. Progress in Photovoltaics: Research and Applications, 14(1):1–12, 2006. (Cited on page 59) [110] P. Saint-Cast, M. Rudiger, A. Wolf, M. Hofmann, J. Rentsch, and R. Preu. Advanced analytical model for the effective recombination velocity of locally contacted surfaces. Journal of Applied Physics, 108(1):013705, 2010. (Cited on page 60) [111] A. Wolf, D. Biro, J. Nekarda, S. Stumpp, A. Kimmerle, S. Mack, and R. Preu. Comprehensive analytical model for locally contacted rear surface passivated solar cells. Journal of Applied Physics, 108(12):124510, 2010. (Cited on page 60) [112] A. Cacciato, F. Duerinckx, K. Baert, M. Moors, T. Caremans, G. Leys, K. De Keersmaecker, and J. Szlufcik. Investigating manufacturing options for industrial PERL-type si solar cells. Solar Energy Materials and Solar Cells, 113(0):153 – 159, 2013. (Cited on page 62) [113] J. Bass. Landolt-Bornstein - Numerical data and functional relationships in science and technology. Springer-Verlag Berlin, Heidelberg, New York, 1982. (Cited on page 66) [114] A. Uruena, J. Horzel, S. Singh, I. Kuzma-Filipek, E. Cornagliotti, J. John, R. Mertens, and J. Poortmans. Rear contact and BSF formation for local Al-BSF solar cells. Energy Procedia, 27:561 – 566, 2012. Proceedings of the 2nd International Conference on Crystalline Silicon Photovoltaics SiliconPV 2012. (Cited on page 68) [115] T. Lauermann, A. Zuschlag, S. Scholz, G. Hahn, and B. Terheiden. The influence of contact geometry and sub-contact passivation

133

134

BIBLIOGRAPHY

on the performance of screen-printed Al2 O3 passivated solar cells. In Proceedings of the 26th European Photovoltaic Solar Energy Conference, pages 1137 – 1143, 2011. (Cited on page 72) [116] D. Chen, Z. Liang, Y. Yang, H. Shen, and Y. Liu. Structure simulation of screen printed local back surface field for rear passivated silicon solar cells. In Proceedings of the 38th IEEE Photovoltaic Specialists Conference, pages 001018–001022, 2012. (Cited on pages 72 and 73) [117] J. Müller, K. Bothe, S. Gatz, H. Plagwitz, G. Schubert, and R. Brendel. Contact formation and recombination at screen-printed local aluminum-alloyed silicon solar cell base contacts. Electron Devices, IEEE Transactions on, 58(10):3239–3245, 2011. (Cited on page 73) [118] T. Trupke and R. Bardos. Photoluminescence: A versatile characterization technique for crystalline silicon. In NREL, 2005. (Cited on page 75) [119] T. Trupke and R.A. Bardos. Photoluminescence: a surprisingly sensitive lifetime technique. In Conference Record of the 31st IEEE Photovoltaic Specialists Conference, pages 903 – 906, 3-7 2005. (Cited on page 75) [120] F. J. Bottari, W. Montanew-Ortiz, D. C. Wong, P. J. Richter, F. C. Dimock, M. Bowers, and T. Bao. High-ramp industrial firing processes for the metallization of crystalline silicon solar cells. In Proceedings of the 35th IEEE Photovoltaic Specialists Conference, 2010. (Cited on page 79) [121] M. Hermle, E. Schneiderlöchner, G. Grupp, and S. W. Glunz. Comprehensive comparison of different rear side contacting methods for high-efficiency solar cells. In 20th European

BIBLIOGRAPHY

Photovoltaic Solar Energy conference, 10 june, 2005. (Cited on page 89) [122] J.A. Rand and P.A. Basore. Light-trapping silicon solar cellsexperimental results and analysis. In Photovoltaic Specialists Conference. Conference Record of the Twenty Second IEEE, volume 1, pages 192 –197, 1991. (Cited on page 90) [123] D.K. Schroder, R.N. Thomas, and J.C. Swartz. Free carrier absorption in silicon. Solid-State Circuits, IEEE Journal of, 13(1):180 – 187, 1978. (Cited on page 90) [124] S.J. Pennycook. Z contrast STEM for materials science. Ultramicroscopy, 30(1):58–69, 1989. (Cited on page 94) [125] K. Ishizuka. A practical approach for STEM image simulation based on the FFT multislice method. Ultramicroscopy, 90(2):71– 83, 2002. (Cited on page 94) [126] P. Favia and H. Bender. On the gallium accumulation at the boundaries of Al layers in FIB prepared TEM specimens. In EMC 2008 14th European Microscopy Congress, Germany, pages 393–394. Springer Berlin Heidelberg, 2008. (Cited on page 94) [127] R.C. Hugo and R.G. Hoagland. Gallium penetration of aluminum: in-situ TEM observations at the penetration front. Scripta Materialia, 41(12):1341 – 1346, 1999. (Cited on page 94) [128] D. Nobili. Properties of Silicon, Solubility of B in Si. EMIS, 1987. (Cited on page 102) [129] T. Yoshikawa and K. Morita. Solid solubilities and thermodynamic properties of aluminum in solid silicon. Journal of The Electrochemical Society, 150(8):G465–G468, 2003. (Cited on page 102)

135

136

BIBLIOGRAPHY

[130] W. Vandervorst and M. Meuris. Method for resistance measurements on a semiconductor element with controlled probe pressure, January 1992. (Cited on page 103) [131] W. Vandervorst and M. Meuris. Method for resistance measurements on a semiconductor element with controlled probe pressure, November 1994. (Cited on page 103) [132] R. G. Mazur and D. H. Dickey. A spreading resistance technique for resistivity measurements on silicon. Journal of The Electrochemical Society, 113(3):255–259, 1966. (Cited on page 103) [133] A. Olbrich, B. Ebersberger, and C. Boit. Conducting atomic force microscopy for nanoscale electrical characterization of thin SiO2 . Applied Physics Letters, 73(21):3114–3116, 1998. (Cited on page 103) [134] P. Eyben, M. Xu, N. Duhayon, T. Clarysse, S. Callewaert, and W. Vandervorst. Scanning spreading resistance microscopy and spectroscopy for routine and quantitative two-dimensional carrier profiling. Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures, 20:471, 2002. (Cited on page 103) [135] P. De Wolf, T. Clarysse, and W. Vandervorst. Quantification of nanospreading resistance profiling data. Journal of Vacuum Science Technology B: Microelectronics and Nanometer Structures, 16(1):320 –326, 1998. (Cited on pages 105 and 106)

C U R R I C U L U M V I TA E

29 Sep. 1980

Born in Valladolid (Spain)

1998-2004

Telecommunication engineering degree by Universidad de Valladolid.

2004

Internship for final project done as a Bachelor of Engineering in Electronic Systems in Dublin City University (Ireland), with the maximum grade.

2004-2006

Electronic engineering degree by Universidad de Valladolid.

2006-2007

Internship for master thesis at imec (interuniversity micro-electronics center, Leuven, Belgium), obtaining the maximum grade.

2008-2013

Ph.D. researcher at the KU Leuven - Department of Electrical Engineering (ESAT) and performed at imec.

137