POLY-SI FILMS ON ZnO:Al COATED GLASS PREPARED
BY THE ALIMINIUM-INDUCED LAYER EXCHANGE
PROCESS
von
Master of Engineering
KYU YOUL LEE
geboren in Seoul/Südkorea
der Fakultät IV - Elektrotechnik und Informatik
der Technischen Universität Berlin
zur Erlangung des akademischen Grades
Doktor der Ingenieurwissenschaften (Dr. -Ing.)
genehmigte Dissertation
Promotionsausschuss:
Vorsitzender: Herr Professor Dr. Karlheinz Bock
Gutachter: Herr Professor Dr. Bernd Rech
Herr Professor Dr. Jürgen Müller (TU Hamburg-Harburg)
Tag der wissenschaftlichen Aussprache: 17.5.2010
Berlin 2010
D83
i
C
ONTENT
2.1 Aluminium-induced layer exchange (ALILE) process.....................................7
2.1.1 The growth modeling of the ALILE process................................................9
2.1.2 The influence of the thickness ratio of a-Si/Al...........................................10
2.1.3 The influence of the interlayer between Al and a-Si layer .........................11
2.1.4 The influence of the process temperature...................................................12
2.1.5 The influence of hydrogen in a-Si layer .....................................................12
2.1.6 The removal of Si islands ...........................................................................13
2.1.7 The effect of hydrogen passivation.............................................................13
2.1.8 The formation of n-type poly-Si layer........................................................14
2.1.9 The use of a substrate coated with a conducting layer ...............................14
2.1.10 Further applications of the ALILE process.............................................15
2.2 The absorber layer growth..............................................................................15
2.3 Aluminium doped zinc oxides........................................................................17
2.3.1 Basic properties of ZnO.................................................................................17
2.3.2 Deposition of zinc oxide thin films................................................................19
3.1 Preparation......................................................................................................23
3.1.1 Substrates....................................................................................................24
3.1.2 Layer deposition .........................................................................................25
3.1.3 Oxidation ....................................................................................................26
3.1.4 Annealing....................................................................................................26
3.1.5 Chemical mechanical polishing (CMP)......................................................26
3.2 Characterization..............................................................................................28
3.2.1 In-situ optical microscopy ..........................................................................28
3.2.2 Raman spectroscopy...................................................................................29
ABSTRACT……...............................................................................................................I
CHAPTER 1 INTRODUCTION................................................................................1
CHAPTER 2 STATE OF THE ART..........................................................................7
CHAPTER 3 EXPERIMENTAL..............................................................................23
ii
3.2.3 Electron backscatter diffraction (EBSD).....................................................31
3.2.4 X-ray diffraction spectroscopy....................................................................32
3.2.5 4-point probe measurement.........................................................................34
3.2.6 Hall measurement........................................................................................34
3.2.7 UV-VIS spectroscopy .................................................................................36
4.1 Motivation.......................................................................................................37
4.2 Properties of ZnO:Al layers ............................................................................38
4.3 Properties of ZnO:Al /poly-Si stacks..............................................................40
4.4 Electrical properties of ZnO:Al /poly-Si stacks..............................................44
4.5 Comparison .....................................................................................................46
4.6 Influence of SiN
x
as a barrier layer.................................................................48
4.7 Influence of post treatments............................................................................49
4.8 Application of SnO
2
:f films.............................................................................51
4.9 Conclusion.......................................................................................................53
5.1 Kinetics of crystallization................................................................................55
5.1.1 Nucleation ...................................................................................................60
5.1.2 Grain growth ...............................................................................................61
5.2 Structural properties........................................................................................63
5.2.1 Crystalline quality.......................................................................................63
5.2.2 Preferential orientation................................................................................71
5.2.3 Grain size.....................................................................................................79
5.2.4 Defect analysis ............................................................................................84
5.2.5 Concentration of impurities.........................................................................89
5.3 Conclusion.......................................................................................................92
6.1 Preparation and structure.................................................................................95
6.2 Solar cell results..............................................................................................97
6.2.1 Solar cell on ZnO:Al coated glass...............................................................97
6.2.2 Solar cell on glass......................................................................................100
6.3 Conclusion and outlook.................................................................................105
CHAPTER 4 TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS......37
CHAPTER 5 POLY-Si FILMS ON ZnO:Al COATED GLASS..............................55
CHAPTER 6 POLY-Si THIN-FILM SOLAR CELLS.............................................95
iii
[Chapter 1]................................................................................................................111
[Chapter 2]................................................................................................................113
[Chapter 3]................................................................................................................119
[Chapter 4]................................................................................................................121
[Chapter 5]................................................................................................................124
[Chapter 6]................................................................................................................127
CHAPTER 7 CONCLUSIONS ..............................................................................107
ABBREVIATIONS, SYMBOLS AND UNITS...........................................................109
REFERENCES… .........................................................................................................111
LIST OF PUBLICATIONS..........................................................................................129
CONFERENCES..........................................................................................................130
ACKNOWLEDGEMENT............................................................................................131
iv
I
A
BSTRACT
The formation of large-grained polycrystalline silicon (poly-Si) films on transparent
conductive oxide (TCO) coated glass opens up new possibilities for the fabrication of
photovoltaic devices. This work addresses the growth of large-grained poly-Si films on
Al doped ZnO (ZnO:Al). A fundamental prerequisite was the realization of temperature
stable Al doped ZnO and the successful formation of poly-Si layers on the ZnO:Al layer.
The investigated key aspects have been (i) the temperature stability of the ZnO:Al films
which are capped with a poly-Si layer, (ii) the study of poly-Si thin films formed on
ZnO:Al coated glass and on bare glass (for comparison) by the aluminium-induced
layer exchange (ALILE) process, and (iii) the fabrication of poly-Si thin film solar cells
on ZnO:Al coated glass.
While uncoated ZnO:Al films show a strong increase of resistivity upon heat treatment,
Si coating of the ZnO:Al layers used in this study resulted in electrical properties that
were not only stable but considerably improved in their electrical properties. The
kinetics of Si crystallization on ZnO:Al coated glass and on bare glass by the ALILE
process was studied. Although the activation energy for the nucleation and the grain
growth on ZnO:Al coated glass showed no significant difference as compared to the
activation energy for the nucleation and the grain growth on bare glass, it was found
that the ALILE process time on ZnO:Al coated glass is shorter than the ALILE process
time on bare glass. Structural properties of poly-Si thin films on ZnO:Al coated glass
and on bare glass were studied by Raman spectra, EBSD, and XRD measurements. The
preferential (100) orientation of poly-Si films on ZnO:Al coated glass and on bare glass
had a similar value of 60% and did not depend on annealing temperatures. The grain
size of poly-Si films on ZnO:Al coated glass was slightly smaller than the grain size of
poly-Si films on bare glass. The underlying ZnO:Al film did not influence the defect
density of thickened poly-Si films as determined by Secco etching. Finally, poly-Si
films formed on ZnO:Al coated glass using the ALILE process have been successfully
introduced as seed layers in poly-Si thin film solar cells, showing an efficiency and
open-circuit voltage of 2% and 389 mV, respectively.
II
INTRODUCTION 1
CHAPTER 1
I
NTRODUCTION
The European photovoltaic industry association (EPIA) announced that world’s future
is bright with solar electricity [EPIA]. As EPIA said, the photovoltaic (PV) market is
booming and dramatically growing. The cumulative installed capacity of PV systems
around the world had reached more than 9,200 megawatts (MW) by the end of 2007.
The yearly installed capacity of PV around the world in 2007 reached a record of 2,826
MW, representing growth of 62% compared to the previous year [Sol08]. Installations
of PV cells and modules around the world have been growing at an average annual rate
of more than 35% since 1998 [EPI08]. The lowest price for a PV module, excluding
installation and other system costs, has dropped from almost $100 per watt in 1975 to
less than $4 per watt at the end of 2008. Average PV prices are projected to drop to $2
per watt in 2010 with expanding polysilicon supplies and production costs of thin film
PV are expected to reach $1 per watt in 2010. According to Earth Policy Institute solar
electricity is poised to take a prominent position in the global energy economy with
concerns about rising oil prices and climate change spawning political momentum for
renewable energy [Ear07].
PV cells are generally fabricated either from crystalline silicon, sliced from ingots or
from grown ribbons, or from thin films on a low-cost substrate. The crystalline silicon
PV cell (mono and multi silicon wafer) has monopolized 87% of PV cell production in
2007. Meanwhile, thin film production more than doubled from 181 MW in 2006 to 400
MW in 2007, accounting for 12% of total PV production [Sol08].
Crystalline silicon wafer is still the mainstay of most PV modules in despite of the
highest price. Efficiencies of more than 20% have been obtained with silicon cells
already in mass production [Tan03]. The thickness of wafers is also an important factor
as well as the efficiency of the solar cells. Thinner wafers mean less silicon needed per
solar cell and therefore lower cost. The average thickness of wafers has been reduced
INTRODUCTION 2
from 0.32 mm in 2003 to 0.17 mm in 2008 [EPI08]. Over the same period, the average
efficiency has increased from 14% to 16%. Nevertheless, the thickness of wafers is
limited by the mechanical stability during wafer handling. The problem has to be solved
to reduce the thickness of wafers for reducing PV prices.
Thin film PV modules are produced by depositing extremely thin films by either silicon
or other materials onto a low-cost substrate such as glass, stainless steel or plastic.
These have a benefit compared to the crystalline silicon PV technology in lower
production costs. Several new companies are working on the development of thin film
production. The successful implementation of such thin film technologies will offer
opportunities for significantly higher throughput in the factory and lower costs. EPIA
expects a growth in the thin film market share to reach about 20% of the total
production of PV modules by 2010 [EPI08]. Mainly, three types of thin film modules
are commercially available at the moment. These are manufactured from copper indium
diselenide (CIS) or copper indium gallium diselenide (CIGS), hydrogenated amorphous
silicon (a-Si:H) and/or microcrystalline silicon (c-Si:H), and cadmium telluride (CdTe).
All of these have active layers in the thickness range of less than a few micrometers.
This allows higher automation once a certain production volume is reached, while a
more integrated approach is possible in module construction. The process is less labor-
intensive compared to the assembly of crystalline modules, where individual cells have
to be interconnected. In order to collect all the current, the thin film device is
sandwiched between two contact layers. The front contact ought to be conductive and
transparent. Therefore transparent conducting oxides (TCOs) like ZnO:Al, SnO
2
:F or
In
2
O
3
:SnO
2
(ITO) are used.
On a-Si:H/µc-Si:H tandem solar modules of sizes up to 5.7 m
2
remarkable initial
efficiencies of 11.6% were obtained by Applied Materials in 2008 [She08]. Even
though the result is quite impressive, it is believed that a-Si:H/µc-Si:H tandem
structures are limited compared to multi-crystalline silicon wafer due to the
recombination at grain boundaries and the light-induced degradation of a-Si:H caused
by the Staebler-Wronski effect [Sta77]. A large grain size is desirable in order to
minimize recombination at grain boundaries.
INTRODUCTION 3
Large-grained polycrystalline silicon (poly-Si) thin-film solar cells feature the potential
to combine the advantages of both crystalline silicon wafers (high material quality) and
thin-film technologies (low costs). Recent theoretical calculations show that it is
possible to achieve 17% efficiency in 2 µm thick silicon films if the grain size is larger
than 10 µm and the dislocation density is less than 10
6
cm
-2
[Deb00]. Ideally a large
grained silicon thin film should be formed on a foreign inexpensive substrate, like glass.
Large grained can be defined as laterally larger than the layer thickness. In this case the
efficiency is limited by the recombination at the front and back contacts. Thin poly-Si
films on glass for solar cells can be prepared for example by (i) solid phase
crystallization (SPC) [Mat90, Gre04], where amorphous Si (a-Si) is transformed into
poly-Si by annealing at about 600 °C, (ii) laser crystallization (LC), where amorphous
Si is melted by laser pulses [Im93] and subsequently into solidified poly-Si, or (iii) the
‘seed layer approach’ [Fuh04], where in a first step a very thin large-grained poly-Si
film (seed layer) is formed and in a second step this seed layer is thickened homo-
epitaxially [Rau04, Ges06].
So far the most successful way to prepare poly-Si thin-film solar cells is based on SPC
of amorphous silicon. Poly-Si thin film solar cells based on the SPC process are
fabricated by CSG Solar. The Mini-module efficiencies of up to 10.4% have already
been obtained [Kee07]. However, the SPC process requires long crystallization times.
In case of LC technique, the achievable Si grain sizes of the crystallized material are in
the range of hundred micrometers [Vou03] and of good crystallographic quality
[And06]. The major disadvantage of LC, based on excimer laser technology, is its low
throughput due to small laser spot size, which is not suitable for large area devices such
as solar cells.
Another attractive method to prepare such large-grained seed layers is the aluminum-
induced layer exchange (ALILE) [Nas98, Mat01, Sch05] which is used in this work.
The ALILE process has a shorter process time compared to SPC. Commercial
deposition techniques can be used such as sputtering or e-beam evaporation which is not
expensive. The ALILE process is a special form of aluminum-induced crystallization
(AIC). During the ALILE process the initial glass/Al/oxide/a-Si layer stack is
INTRODUCTION 4
transformed into a glass/poly-Si/oxide/Al(+Si) layer stack by a simple annealing step
below the eutectic temperature of the Al/Si system (577 °C).
The resulting poly-Si layer is continuous and features large grains (over 10 µm) with a
preferential (100) orientation [Kim02]. The process temperature and time of the ALILE
process are more cost effective compared to a process temperature and time of SPC.
Even though the films feature a large grain size, the poly-Si films are not suitable for
absorber layers due to the very low thickness and the very high doping level of Al. But
due to the large grain sizes such layers are used as templates for subsequent epitaxial
thickening. Although the poly-Si seed layer is highly doped with Al (doping
concentration of 1×10
19
cm
2
[Nas00]) the sheet resistance is too high for high-efficiency
solar cells. An additional contacting scheme is requested to improve a lateral
conductance.
An appealing option for an additional contacting scheme is the seed layer formation on
a transparent conductive oxide (TCO) coated substrate because TCOs show a high
transparency in the spectral region where the solar cell is operating and they provide a
high electrical conductivity. TCOs are well established for thin-film solar cells based on
amorphous Si (a-Si:H) or microcrystalline Si (µc-Si:H) [Mul04, Klu03, Rec06] but are
not used so far for poly-Si thin-film solar cells. The formation of poly-Si seed layers on
Al doped ZnO (ZnO:Al) coated glass using the ALILE process has been introduced
recently [Dim07].
The formation of poly-Si films on ZnO:Al coated glass substrates using the ALILE
process is the topic of this work. The thesis is divided into following chapters:
Chapter 2: ‘State of the art’ provides the background and the current status of the
aluminum-induced layer exchange (ALILE) process and the Aluminium-doped ZnO
(ZnO:Al).
Chapter 3: ‘Experimental’ lists the sample preparation steps from the preparation of
substrates to chemical mechanical polishing to produce a flat surface of the resulting
thin films and to remove the Al layer on top. The characterization methods are
INTRODUCTION 5
explained.
Chapter 4: ‘Temperature stability of ZnO:Al’ shows the electrical and optical properties
of ZnO:Al/poly-Si stacks. In particular, the properties of ZnO:Al layer covered with
poly-Si film fabricated using the ALILE process are studied.
Chapter 5: ‘Poly-Si films on ZnO:Al coated glass’ addresses the influence of the two
main parameters on the poly-Si properties. First, the influence of ALILE process
temperature is investigated as it is crucial for the layer exchange process. Second, the
properties of poly-Si films on ZnO:Al coated glass are compared with the properties of
poly-Si films on glass.
Chapter 6: ‘Solar cells’ shows poly-Si thin film solar cells fabricated on ZnO:Al coated
glass. Sample structure and solar cell performance are characterized in detail and the
suitability of the applied ZnO:Al coated glass substrates is evaluated.
Chapter 7: ‘Conclusion’ summarizes the most important results and closes with an
outlook on future perspectives of the ALILE process.
INTRODUCTION 6
STATE OF THE ART 7
CHAPTER 2
S
TATE OF THE ART
2.1 ALUMINIUM-INDUCED LAYER EXCHANGE
(ALILE) PROCESS
In this chapter the history of the aluminum-induced layer exchange (ALILE) process is
summarized. The ALILE process is a special form of aluminium-induced crystallization
(AIC) where a substrate/Al/a-Si stack is transformed into a substrate/poly-Si/oxide/Al(+
Si islands) stack by a simple annealing step below the eutectic temperature of the
Al/oxide/Si system (577 °C). The ALILE process is well explained by O. Nast [Nas00]
and illustrated schematically in Fig 2.1.
The crystallization of a-Si with Al was used in a type of layer exchange process already
by Boatright et al. and Greene et al. [Boa76, Gre76]. The experiment investigated by
Boatright et al. is shown schematically in Fig. 2.2. A hole was made by
photolithographic means in a Si oxide layer covering a silicon wafer. After deposition
of Al on top of the oxide covered wafer and annealing at 550 °C for 5...25 min, the Al
dissolved the Si beneath the hole. Amorphous silicon was evaporated onto the Al layer
(Fig. 2.2(a)) and this stack was annealed again at 475...525 °C for another 10...20 min.
It was found that the hole is refilled with crystalline silicon (Fig. 2.2(b)). At that time
many researcher investigated the phenomena for understanding the crystallization of a-
Si in contact with Al. Majni et al. [Maj77, Maj78, Maj79] investigated the growth
kinetics of the crystallization of a-Si by deposition of Al (700 nm)/a-Si (650 nm) double
layers on differently orientated Si wafers and actually mentioned for the first time that
Si and Al exchange their position during heat treatment at lower temperatures. Tsaur et
al. [Tsa81] used the crystallization of a-Si through an Al layer to form highly p-doped
emitters on n-doped c-Si wafers. The separated Al layer was removed and the resulting
solar cells reached an efficiency of 10.4 %. The Al inclusions in crystalline silicon films
were mentioned by Scranton et al. [Scr78]. They concluded that fast growth rates can
STATE OF THE ART 8
cover defects, such as Al inclusions, in the epitaxial film but the epitaxial films prepared
with slower growth rates by Majni et al. [Maj77] also contained Al inclusions.
QingHeng et al. [Qin82] investigated the diffusivity and growth rate of silicon in Al and
revealed that the activation energy is almost identical to that of the Al-Si system but the
change of diffusivity is caused by different experimental conditions. They also
mentioned that the existence of a thin oxide layer at the Al/a-Si interface reduces the
diffusion process significantly and the growth time usually should be longer to produce
higher quality growth layers.
Glasssubstrate
Al
a-Si
Glasssubstrate
Poly-Si
Al
Oxide Si
Figure 2.1: Schematic of the ALILE process. The layer stack before (left) and after
(right) the layer exchange are shown [Nas00]. The oxide interface layer remains in
position during the process.
Poly-Si
Al
OXIDE
(a) (b)
Figure 2.2: Schematic illustration of a layer exchange phenomenon by Boatright et al.
[Boa76]: (a) Evaporating amorphous Si (~200 nm) on thick Al layer and (b) after heat
treatment at 475-525 °C for 10 or 20 min Si fill the dissolution pit.
STATE OF THE ART 9
Researchers investigated Al-induced crystallization (AIC) process on oxidized silicon
wafer or on silicon wafer at that time. Nast et al. [Nas98] started to investigate the AIC
process on glass substrate at the University of New South Wales (UNSW) in Sydney,
Australia. And then many other institutions started research in this field.
2.1.1 THE GROWTH MODELING OF THE ALILE PROCESS
Konno et al. [Kon92, Kon93, Kon94] studied the crystallization mechanism of a-Si in
an a-Si/Al multilayer system using transmission electron microscopy (TEM). They
found that the Si atoms diffuse through the Al grain from amorphous to crystalline
phases and suggested that the fastest reaction path occur to reduce the excess free
energy for crystallization of a-Si. Nast et al. [Nas00S] developed a model for the ALILE
process dividing the overall process into four steps. This model was extended by
Widenborg et al. [Wid02] as shown in Fig 2.3. They found that the formation of Si
islands in the top layer is completed after the crystallization process in the bottom layer.
Step 1 involves the interaction of amorphous silicon with the Al and the dissociation of
the amorphous phase across the oxidized Al/Si interface. Once Si is dissolved in the
metal, the Si atoms diffuse within the Al layer (step 2). The Si nucleates within the Al
layer (step 3). While process steps 1, 2 and 3 correspond to the steps suggested by Nast,
here step 4 describes the diffusion of Al into the top layer. Al is displaced by Si grains
and forced to diffuse into the a-Si top layer (step 4a). The a-Si crystallizes in presence
of the Al and small Si crystallites form in the Al top layer. The Al in this mixed phase
dissolves further a-Si and grows (step 4b) competing with the Si growth in the bottom
layer in this model. While the poly-Si in the bottom layer grows supplied by step 5,
silicon is also crystallized in the top layer. Small nanocrystalline Si (nc-Si) crystallites
will agglomerate at the more compact central Si islands, even when the actual layer
exchange is completed (step 6).
STATE OF THE ART 10
Figure 2.3: ALILE growth model by Widenborg et al. [Wid02]. Details are explained
in the text.
Various deposition methods and processing parameters are known to influence the layer
exchange process. In the following these factors are highlighted for the ALILE process.
2.1.2
THE INFLUENCE OF THE THICKNESS RATIO OF a-Si/Al
The relevance of the thickness ratio of Al and a-Si layer was recognized by Harris et al.
[Har77]. However, they did not succeed in producing continuous poly-Si films. Nast et
al. [Nas00S] investigated the influence of the thickness ratio of Al and a-Si layer and
found that the a-Si layer must be at least as thick as the Al layer for obtaining a
STATE OF THE ART 11
continuous layer of poly-Si. Sugimoto et al. [Sug05] found that the crystal direction was
strongly influenced by the thicknesses of Al and Si and showed preferred orientation
with Si (111) with decreasing thicknesses. Hsu et al. [Hsu03] found that the Al has
compressive stress and the a-Si layer has tensile stress and there is a compressive/tensile
stress turning point at a thickness ratio of 4/1 for the a-Si/Al stacked structure.
2.1.3 THE INFLUENCE OF THE INTERLAYER BETWEEN Al AND a-Si
LAYER
The ALILE process requires a thin permeable membrane between the Al and the a-Si
layer which controls the diffusion of Al and Si [Gal02a]. This membrane does not
participate in the layer exchange process and therefore remains in position throughout
the annealing process. The interlayer between the initial Al and a-Si layer plays a
crucial role because it influences grain size, preferential orientation and crystallization
time.
The influence of an Al oxide layer formed at an Al/a-Si interface was studied by
Ottaviani et al. and Cai et al. [Ott79, Cai92]. Ottaviani et al. suggested that the oxide
layer act as a diffusion barrier for the Al atoms, but allows Si atoms to diffuse into the
metal layer with subsequent crystallization. However, Cai et al. concluded that a gradual
Al/Si interface reduces the crystallization temperature even further. However, the
presence of oxygen at the Al/a-Si interfaces inhibits the Al-induced crystallization.
The role of the Al/a-Si interface on the overall crystallization process was also
investigated by Nast et al. [Nas00] and by Sieber et al. [Sie03]. According to the report
by Sieber et al., an oxide layer introduced at the a-Si/Al interface is very important from
the viewpoint of a diffusion of Si into Al film and the local nucleation of Si in the Al
film, which lead to ALILE of the a-Si/Al bilayer and poly-Si formation of large grain
size. The influence of the aluminium oxide interlayer produced by different oxidation
processes was investigated by Schneider et al. [Sch05] and Pihan et al. [Pih04]. Usually
a native Al oxide layer is formed on top of the Al prior to a-Si deposition. This oxide
STATE OF THE ART 12
layer remains in position and separates top and bottom layer throughout the layer
exchange process. Schneider et al. [Sch05] investigated the effect of aluminium oxide
layers formed by air and by thermal oxidation. Pihan et al. [Pih04] fabricated the
aluminium oxide using HNO
3
solution. Both aluminium oxides produced by thermal
oxidation and using HNO
3
solution can increase the grain size of poly-Si films
compared to the aluminium oxides formed by air. The influence of oxygen present
within Al layers was investigated by Klein et al. [Kle04]. They found that the increasing
oxygen content within Al layers leads to faster crystallization.
Kim et al. [Kim96] studied glass/a-Si/SiO
2
/Al structures to analyze the process
mechanism at an oxidized interface between Al and a-Si layer during aluminium-
induced crystallization. They presented that the silicon oxide acts as a membrane that
allows the local interaction of the Al and the amorphous silicon due to the reduction of
the oxide and the formation of Al spikes.
Schneider investigated the influence of molybdenum (Mo) as an interlayer and it was
found that the grain size of poly-Si film is much bigger (up to 40 µm), but Mo reacts
with Al and Si during the ALILE process [Sch05].
2.1.4 THE INFLUENCE OF THE PROCESS TEMPERATURE
The process temperature is one of the most important parameters which influence the
grain size and the preferential orientation. It was shown that the grain size decreases and
the preferential (100) orientation increases with decreasing annealing temperature
[Kim02, Gal06].
2.1.5 THE INFLUENCE OF HYDROGEN IN a-Si LAYER
Amorphous silicon (a-Si) layers can be deposited by either physical vapor deposition
(PVD) or plasma enhanced chemical vapor deposition (PECVD) methods. Amorphous
STATE OF THE ART 13
silicon layers deposited by PECVD method using Silane (SiH
4
) as source gas contain
hydrogen. The influence of hydrogenated a-Si (a-Si:H) was investigated by Kishore et
al. [Kis03]. It was found by in-situ XRD measurements that the crystallization of a-Si:H
occurs faster than crystallization of a-Si without hydrogen at a process temperature of
150 °C. The influence of different annealing atmospheres and hydrogen concentration
in a-Si on the layer exchange process was investigated by Dimova-Malinovska and
Grigorov [Dim06, Gri06]. They showed that the use of hydrogen in the annealing
atmosphere leads to improved structural properties of the resulting poly-Si film when
the a-Si:H precursor layers contain 9 at.% hydrogen. Zou et al. [Zou05] concluded that
tensile stress in a-Si:H can have a significant effect on AIC of a-Si:H. The
crystallization of intrinsic and boron-doped a-Si:H was studied by Al-Dhafiri et al.
[Al05]. They noticed that crystallization was further enhanced in the p-type a-Si:H.
2.1.6 THE REMOVAL OF Si ISLANDS
Si islands present on the surface of ALILE poly-Si seed layers have a negative effect on
the epitaxy [Dog08]. The removal of these islands could therefore lead to an increased
absorber layer quality and solar cell performance. To remove the residual Si islands on
top of poly-Si films various treatments are used. A wet chemical lift-off process for the
Si islands within the Al layer has been developed in order to produce a smooth and flat
surface on the ALILE layer [Wid03]. A chemical mechanical polishing (CMP) method
was used to remove Si islands [Sch05]. Plasma selective etching process with SF
6
gas
was used by Van Gestel et al. [Ges08a].
2.1.7
THE EFFECT OF HYDROGEN PASSIVATION
Walter Schottky Institute in Germany investigated the influence of hydrogen
passivation on the electronic properties of ultra-thin poly-Si films prepared by ALILE
[Jae08]. They found a drastic increase of the resistivity for very thin hydrogenated
STATE OF THE ART 14
samples, which is attributed to a combination of two effects: (1) the reduction of free
holes due to acceptor passivation and (2) compensation of free holes remaining after H-
passivation by interface states.
2.1.8
THE FORMATION OF N-TYPE POLY-Si LAYER
Recently, the formation of n-type ALILE seed layers was investigated by Tüzün et al.
[Tuz08]. It was found that p-type poly-Si layers obtained by ALILE of amorphous
silicon can be successfully converted into n
+
-type by phosphorus diffusion from a
highly doped solid source in a furnace at 950 °C for 1hour.
2.1.9 THE USE OF A SUBSTRATE COATED WITH A CONDUCTING
LAYER
Although the poly-Si films formed using ALILE process feature a high Al concentration
of 10
19
cm
-3
, the lateral conductivity of poly-Si film itself is not sufficient for solar cell
applications. Glass substrates coated with a conductive material coated were
investigated in order to increase the lateral conductivity. Widenborg et al. [Wid02]
investigated the formation of poly-Si films on molybdenum (Mo) and on titanium
nitride (TiN) coated substrate. It was found that although Mo is not compatible due to
the reaction with Si during the AIC process TiN has a medium to high resistivity (120–
500 µΩ cm) and an estimated low back reflectance at the near-infrared wavelengths
critical for light trapping. But TiN is the only possible choice for process temperatures
above 540 °C.
Dimova-Malinovska et al. [Dim07] produced poly-Si films on ZnO:Al coated glass by
Aluminium-Induced Crystallisation (AIC) in different annealing atmospheres (air, N
2
,
and N
2
+H
2
) and analyzed only the structural properties of poly-Si thin films on ZnO:Al
coated glass substrates using Raman spectra and XRD method. It was found that an
STATE OF THE ART 15
annealing in forming gas (N
2
+H
2
) led to a better structural quality of the poly-Si films,
compared to annealing in air or nitrogen. The (111) preferential orientation was
observed by XRD.
2.1.10 FURTHER APPLICATIONS OF THE ALILE PROCESS
Zou et al. [Zou07] investigated the formation of silicon nanowires (SiNWs) using
ALILE process. They asserted that both aluminum thickness and native silicon oxide
between Al and a-Si play critical roles in the formation of the nanostructures. The
results suggest that the rapid in-plane growth of these SiNWs can be attributed to self-
assembly of nanosilicon islands, which is different from traditional SiNWs growth
mechanisms such as vapor-liquid-solid.
Some research groups investigate the formation of poly-SiGe layers using aluminium-
induced layer exchange process. The use of the ALILE process for crystallization of
amorphous SiGe is investigated by Gjukic et al. [Gju04, Gju05]. Here, it was found that
ALILE is suitable for crystallizing binary amorphous material without segregating
different phases [Gju04]. The optical and electrical properties of the resulting
polycrystalline films were found to be comparable with high quality crystalline SiGe
films [Gju05]. Low-temperature Al-induced crystallization of hydrogenated amorphous
silicon–germanium thin films has been investigated by Peng et al. [Pen08]. They found
that the increase of the Ge fraction and annealing temperature yields an enhancement in
film crystallinity and grain size.
2.2 THE ABSORBER LAYER GROWTH
ALILE poly-Si layers are not suitable as absorber layers due to the high hole density of
2×10
18
cm
−3
[Nas01] which is inherent due to the remaining Al dopant content in the
line. Several institutes have investigated the epitaxial thickning of ALILE poly-Si films
STATE OF THE ART 16
by using various silicon deposition methods. There are two types of approaches: One is
the low temperature approach where poly-Si films are thickened at temperatures upto
700 °C, and the other is the high temperature approach where poly-Si films are
thickened at a temperature above 1000 °C. The UNSW and HZB focus on the low
temperature approach which promises a higher potential for cost reduction than the high
temperature approach. The low temperature approach allows for the use of cheap
substrates like a glass. The ALILE seed layers prepared on glass substrates were
successfully thickened epitaxially for the first time by ion assisted deposition (IAD)
[Har00]. Different cell concepts have been developed involving IAD and solid phase
epitaxy (SPE) from a-Si layers deposited on the seed layer [Abe05]. Widenborg et al.
[Wid07] announced an open-circuit voltage (Voc) of 480 mV for poly-Si solar cells.
ALILE poly-Si seed layers have been also thickened epitaxially using pulsed sputter
deposition system (SPUTTER) [Rei02], electron cyclotron resonance chemical vapor
deposition (ECRCVD) [Rau04, Ree06] or electron beam (e-beam) evaporation [Dog08].
So far, the best crystal qualities have been achieved by the later method. A Voc of 346
mV was measured for a solar cell fabricated by epitaxially thickening of poly-Si film by
e-beam evaporation. Ito et al. [Ito03] explored the use of the crystallization of a-Si
deposited on ALILE seed layers for solar cell applications. Some simulations were done
to extrapolate the future goals for solar cells involving ALILE seed layers which
claimed that efficiencies of about 13% could be achieved when various solar cell
parameters are improved [Ito04]. A low temperature epitaxy using ECRCVD on an
inverse ALILE seed layer structures (glass/a-Si/Al) has been studied by Ekanayake et al.
[Eka03, Eka04]. The structural quality of the films was analyzed using electron back
scatter diffraction (EBSD) and X-ray diffraction (XRD) analysis. Some researchers
[Sla06, Gor07] focus on the high temperature approach to thicken ALILE poly-Si films.
ALILE seed layers on ceramics substrates are studied using atmospheric pressure
chemical vapor deposition (APCVD), Gordon et al. have recently achieved solar cell
efficiencies of 8% [Gor07].
STATE OF THE ART 17
2.3 ALUMINIUM DOPED ZINC OXIDES
In this section the basic properties and the current status of ZnO are reviewed. The
deposition methods for ZnO are summarized. Nowadays ZnO:Al films are widely
applied as a front contact layer in order to fabricate thin film solar cells, because
ZnO:Al films are highly transparent in the active range of the absorber layer (400 –
1100 nm in case of µc-Si:H) and also highly conductive to avoid ohmic losses. In the p-
i-n structure, the front contact additionally has to possess a rough surface to provide
efficient light scattering. Furthermore, the ZnO:Al film features favourable physico-
chemical properties for the growth of silicon. In general, the TCO/p-contact plays an
important role for cell performance [Kub96, Boh96, Mul01].
Obvious parameters of the front contact for thin film solar cells are electrical
conductivity and transparency. An improved transparency of the TCO results in a higher
short circuit current density, while an improved conductivity allows a higher fill factor
of an otherwise optimized solar cells. A low sheet resistance is most important for thin
film solar cells due to the high short circuit currents. ZnO:Al is a suitable material.
2.3.1 BASIC PROPERTIES OF ZnO
ZnO is a II-VI compound semiconductor whose ionicity resides at the borderline
between the covalent and ionic semiconductors. It crystallizes in the hexagonal Wurtzite
structure. The wurtzite structure has a hexagonal unit cell with two lattice parameters a
and c in the ratio of c/a =
3/8
= 1.633 (in an ideal wurtzite structure) and belongs to
the space group
4
6
C
ν
in the Schoenflies notation and P6
3
mc in the Hermann–Mauguin
notation [Mor09]. A schematic representation of the wurtzitic ZnO structure is shown in
Fig 2.4. The structure consists of two pervasive grids together. One of the grids is filled
with Zinc atoms, the other with oxygen atoms. The two hexagonal densest lattices
packing lie along the anisotropy c-axis a fraction of the lattice parameter offset. Each
atom is fourfold coordinated, with the nearest neighbors of each atom by the other
STATE OF THE ART 18
atomic sites originate. The lattice parameters are a
0
= 0.325 nm and c
0
= 0.52066 nm at
room temperature [Ell01]. The melting temperature is 1975 °C and the density is 5.67
g⋅cm
-3
at room temperature [Ell01]. The direct band gap of Zinc oxide is 3.2 eV to 3.4
eV [Ell01, Iba03]. The thermal expansion coefficient of ZnO is given as α
a
= 4.31×10
−6
K
−1
and α
c
= 2.49×10
−6
K
−1
at 300K [Hel82, Ada05].
C0
a0
O
Zn
Figure 2. 4: Schematic representation of a wurtzitic ZnO structure. The c-axis is the
anisotropy axis.
STATE OF THE ART 19
2.3.2 DEPOSITION OF ZINC OXIDE THIN FILMS
Different deposition methods are introduced for the production of ZnO (for review see
[Cho83, Hel82, Man81, Vos76, Cha80, Gro01, Pul84,]). These are, for example, the
sputtering [Szy99], chemical vapor deposition (CVD) [Fay00], the pulsed laser
deposition (PLD) [Lor03], electrochemical deposition from aqueous solution [Iza96],
the spray pyrolysis [Par99] and evaporation method (thermally or through electron)
[Wu00]. The sputtering method allows a large area deposition making ZnO films the
industrially most advanced deposition technique for ZnO [Gla00].
First sputtering processes for ZnO deposition were developed in the late 1960’s for
manufacturing surface acoustic wave devices [Hic73]. Large-area sputtering of ZnO
was established in the field of energy-efficient glazing in the early 1980’s. ZnO was
chosen as dielectric material because of its high sputtering rate and its suitability for
reactive DC sputtering. Transparent and conductive, sputtered ZnO-based films with
resistivity below 1,000 µΩ were first reported by Brett et al. [Bre83]. Breakthroughs
such as high-rate magnetron sputtering of ZnO:Al [Ell94] guided the developments
towards large-area manufacturing technology, with main emphasize on thin-film
photovoltaics [Men98, Mul03].
Many attempts have been made to deposit ZnO:Al-based films by magnetron sputtering.
Table 2.1 summerizes a literature survey on sputtered ZnO:Al film. The classification
criteria are: plasma excitation used (DC/MF/RF), reactive or ceramic deposition;
material deposited and film properties relevant for TCO applications.
Some important deposition parameters to control ZnO:Al film properties are deposition
pressure, substrate temperature and amount of oxygen in the sputter gas mixture [Hup06,
Klu03, Min84, Szy99]. Doping concentration in ZnO films plays another important role
for opto-electronic properties [Aga04, Min84, Ell01].
STATE OF THE ART 20
TABLE 2.1. Literatures survey on magnetron sputtering of ZnO:Al-based TCO films
Ref. Scope Growth conditions Layer properties
Process
Material
Year
Coater and magnetron
Process mode and control
Target
Substrate, T
s
(°C)
P/A (W cm
-2
)
a
s
(nm s
-1
)/a
d
(nm s
-1
)
d (nm), c
d
(at.%), ρ (Ωcm),
n
e
(cm
-3
), µ (cm
2
V
-1
s
-1
)
[Bre99] RDCMS
ZnO:Al
1999
Batch, 90 mm planar
Low power
Zn:2 wt% Al
Glass, RT
1.6 W cm
-2
,0.8-2 nm s
-1
170-400 nm, 1.4 – 1.8 at.%
6×10
-4
Ωcm
1.0×10
21
cm
-3
10 cm
2
V
-1
s
-1
[Har91]
RDCMS
ZnO:Al
ZnO:In
1991
Batch
Low power
Zn target, Al or In pieces
Glass, 180 – 220 °C
< 2 nm s
-1
700 nm
2 at.% (Al), ~11 at.% (In)
5×10
-4
Ωcm (Al)
1.4×10
-3
Ωcm (In)
[Ell94]
RDCMS
ZnO:Al
1994
Batch, 3inch planar
Low power
Zn:2 wt% Al
Glass, < 150 °C
2.2 W cm
-2
< 2 nm s
-1
~ 500 nm
1.8 – 2.2 at.%
4.5×10
-4
Ωcm
8×10
20
cm
-3
17 cm
2
V
-1
s
-1
[Wal03]
RDCMS
RMFMS
ZnO:Al
2003
In-line, dual magnetron
tm, q(O
2
) = f(PEM)
1400 ×100 mm
2
each
Zn:2 wt% Al
Glass, RT
< 5 W cm
-2
125 nm m min
-1
1040 nm
9.1×10
-4
Ωcm
[Min90]
DCMS
ZnO:Al
1990
Batch, 120 mm planar
Zn:2-3 wt% Al
2
O
3
Glass, 250 °C
300-600 nm
2.7×10
-4
Ωcm
[Mal00]
RMFMS
ZnO:Al
2001
Batch, dual magnetron
tm, q(O
2
) = f(U)
500 ×88 mm
2
Zn:1.5 wt% Al
Glass, RT, 250 °C
~ 6 W cm
-2
7 nm s
-1
500 nm
7.5×10
-4
Ωcm
2.5×10
-4
Ωcm
[Szy99]
RMFMS
ZnO:Al
1999
Batch, 500 ×88 mm
2
DMS
Met. mode, Zn deposition
Zn:0.9-1.5 wt% Al
(segmented)
100 °C, 250 °C
~ 10 W cm
-2
8.2, 8.8 nm s
-1
~ 500 nm
3.6 at.%, 2.2 at.%
4.8, 3.0×10
-4
Ωcm
8.6, 8.5×10
20
cm
-3
15, 25 cm
2
V
-1
s
-1
STATE OF THE ART 21
Ref. Scope Growth conditions Layer properties
[Kon03]
RMFMS
ZnO:Al
2003
In-line, 400 ×120 mm
2
DMS
tm, q(O
2
) = f(PEM)
Zn:1.5 wt% Al
RT, 300 °C
2.9 W cm
-2
~ 10, 4.4×10
-4
Ωcm
5.8×10
20
cm
-3
24 cm
2
V
-1
s
-1
[Iga88]
RMFMS
ZnO:Al
1988
Batch, 125 mm planar
Zn, Al wires
Glass, 100 °C
0.01 nm s
-1
540 nm
4.9×10
-4
Ωcm
4.3×10
20
cm
-3
24 cm
2
V
-1
s
-1
[Iga91]
RFMS
ZnO:Al
1991
Batch, 125 mm magnetron
Zn:2.0 wt% Al
2
O
3
Sapphire, 230 °C
< 1.4 W cm
-2
0.5 nm s
-1
Hetero epitaxial growth
380-670 nm
2 at.%
1.2 nm min-1: 140 µΩcm
1.3×10
21
cm
-3
34 cm
2
V
-1
s
-1
22.3 nm min-1: 300 µΩcm
[Bos96]
RFMS
ZnO:Al
1996
Batch, 4 inch planar
Zn:2.0 wt% Al
2
O
3
, sintered
Glass, 150 °C
~ 700 nm
6.9×10
-4
Ωcm
[Par97]
RFMS
ZnO:Al
1997
Batch, 4 inch planar
Zn:0.5 wt% Al
2
O
3
, sintered
Glass, 150 °C
140 nm
4.7×10
-4
Ωcm
7.5×10
20
cm
-3
15 cm
2
V
-1
s
-1
[Add99]
RFMS
ZnO:(Al, H)
1999
Batch
Zn:2.0 wt% Al
2
O
3
, sintered
RT
2 W cm
-2
500 nm
4 ×10
-4
Ωcm
6×10
20
cm
-3
27 cm
2
V
-1
s
-1
[Hau01]
RFMS
ZnO:Al
CIS
2001
Batch, 100 mm magnetron
ZnO:2.0 wt% Al
2
O
3
200-250 °C
3.2 W cm
-2
0.5 nm s
-1
~ 10
-3
Ωcm
[Ceb98]
RFDCMS
M: ZnO:Al
1998
Batch, 3 inch magnetron
ZnO:2.0 wt% Al
2
O
3
RT
0.5 W cm
-2
< 0.5 nm s
-1
dev CIS
10-370 nm
3.3-4.0 at.%
~ 7 ×10
-4
Ωcm
5-7×10
20
cm
-3
~ 15 cm
2
V
-1
s
-1
[Jin88]
Co-deposition
ZnO:Al
1988
Batch
Zn:2 wt% Al
Glass, < 100 °C
0.2 nm s
-1
~ 300 nm
2.14 at.%
5.4 ×10
-4
Ωcm
4.5×10
20
cm
-3
26 cm
2
V
-1
s
-1
(DCMS: DC magnetron sputtering; RDCMS: reactive DC magnetron sputtering; RMFMS:
reactive MF magnetron sputtering; RFMS: RF magnetron sputtering; tm: Transition mode; Met.
mode: Metallic mode)
STATE OF THE ART 22
EXPERIMENTAL 23
CHAPTER 3
E
XPERIMENTAL
In the first part of this chapter the sample preparation is described in detail. The second
part of this chapter describes the measurement methods for characterizing structural,
electrical, and optical properties of polycrystalline silicon (poly-Si) films and
ZnO:Al/poly-Si stacks.
3.1 PREPARATION
The ALILE process on ZnO:Al coated glass is shown schematically in Fig. 3.1. The
layer stacks before (left) and after (center) the layer exchange and after chemical
mechanical polishing (right) are shown.
The substrates and their cleaning procedure are described in section 3.1.1. The
conditions for the deposition of ZnO:Al, Al, and a-Si layers are described in section
3.1.2. The oxide interface layer was formed by simple oxidation method as described in
section 3.1.3. The layer stacks have been annealed either within a furnace or a heating
stage of an optical microscope (section 3.1.4). After the layer exchange the Al (+Si) top
layer has to be removed in order to uncover the poly-Si layer. Chemical mechanical
polishing was used to produce a smooth surface as described in section 3.1.5. The last
step is of importance because the poly-Si layer is used as template for epitaxial
thickening in order to form the absorber layer of the thin film solar cell.
EXPERIMENTAL 24
TACMP
a-Si
Al
ZnO:Al
Al(+Si)
Poly-Si
ZnO:Al
Poly-Si
ZnO:Al
Glass
Glass
Glass
Figure 3. 1: Schematic of ALILE process: The layer stacks before (left) and after
(center) the layer exchange and after CMP (right) are shown. The oxide interface layer
(red layer) remains in position during the ALILE process and is removed by chemical-
mechanical polishing (CMP) process.
3.1.1 SUBSTRATES
Borofloat33 glasses from Schott were used as substrates in this study. The chemical
composition of this glass is shown in table 3.1. The strain point lies at 518 °C, which is
lower than the strain point (666 °C) of Corning1737 glass. The coefficient of linear
thermal expansion (C.T.E) of Borofloat33 glass is 3.25×10
-6
K
-1
in the range of 20 to
300 °C. This value is lower than the value (3.76×10
-6
K
-1
) of Corning1737. The C.T.E
of silicon is 2.6×10
-6
K
-1
at 25 °C and 3.25 - 3.6×10
-6
K
-1
in the range of 400 - 500 °C
[Hul99, Wat04]. This means that the C.T.E of Borofloat33 glass has a better match to
the C.T.E of silicon than that of Corning1737 glass. The better match should lead to less
stress and cracking.
The glass substrates were cleaned with a normal glass cleaning solution (Mucasol). The
temperature of the cleaning solution was 80 - 90 °C and the cleaning time was 10 min.
After that the samples were rinsed in de-ionized (DI) water and dried by N
2
. Square 1×1
inch samples with a thickness of 0.7 mm were used. For some experiments silicon
EXPERIMENTAL 25
nitride (SiN
X
) coated Borofloat33 glasses supplied by CSG Solar were used. The
thickness of the SiN
X
layer and the glass is about 80 nm and 3 mm, respectively.
TABLE 3.1: Chemical composition of Borofloat33 glass [Schott].
SiO
2
Al
2
O
3
B
2
O
3
Na
2
O/K
2
O
Chemical
composition 81% 2% 13% 4%
3.1.2 LAYER DEPOSITION
ZnO:Al films used in this work were obtained by Forschung Zentrum Jülich (FZJ).
They were deposited in an in-line sputtering system in dynamic mode using non-
reactive RF-sputtering from ceramic targets containing 1 wt.% Al
2
O
3
. A constant
deposition pressure of 0.1 Pa and a substrate temperature of 300 °C were used. Details
of the sputtering procedure can be found in [Aga04]. The thickness of the ZnO:Al films
used in this thesis, determined by fitting optical transmission and reflection spectra, was
between 690 and 770 nm. The ZnO:Al coated glasses were cleaned by a 10 minute
immersion in an ultrasonic bath filled with DI water heated at 80 ~ 90 °C and dried with
nitrogen gas.
The initial layer stacks for the ALILE process (Al and a-Si) were deposited onto the
ZnO:Al coated glass and on Borofloat33 glass (for comparison) by DC magnetron
sputtering. The base pressure in the sputter chamber was 7×10
-7
mbar. The initial layer
stacks were deposited at room temperature with an Ar pressure of 6.5×10
-3
mbar. For Al
deposition the purity of the Al target was 4N. A sputtering power of 500 W, resulting in
a deposition rate of 160 nm/min, was used to deposit a 300 nm thin film. For a-Si
deposition a Si target with a boron concentration of 1×10
16
atoms/cm
3
was used. Here a
sputtering power of 300 W was used and a film of 375 nm was grown using a
deposition rate of 66 nm/min.
EXPERIMENTAL 26
3.1.3 OXIDATION
An Al oxide layer was formed by exposure to air of the Al-coated substrate prior to the
a-Si deposition in order to obtain continuous poly-Si films. Al oxide layers are formed
rapidly upon exposure to an oxygen containing atmosphere such as air. The oxidation of
aluminium is self-limited. This means that upon reaching a certain oxide thickness, no
further oxidation takes place. It has been shown that the thickness of the native Al oxide
formed on aluminium in dry air at room temperature is in the range 1-2 nm [Mar93,
Nyl94]. In this work the samples were oxidized by exposure to air for 2 hours.
3.1.4 ANNEALING
In order to fabricate poly-Si films on ZnO:Al coated glass and on bare glass the stacked
Al and a-Si layers were annealed in a tube furnace at an annealing temperature between
425 °C and 525 °C in nitrogen ambient. The samples annealed at 425 °C and 450 °C
were annealed for 16 hours and the samples annealed between 475 °C and 525 °C were
annealed for 4 hours. In addition to the annealing experiments in the tube furnace,
annealing experiments were carried out in a heating stage of an optical microscope (see
section 3.2.1.1). These experiments enable the in-situ observation of crystallization
(nucleation and subsequent growth) during the ALILE process.
3.1.5 CHEMICAL MECHANICAL POLISHING (CMP)
Chemical Mechanical Polishing (CMP) is a process that uses abrasive, corrosive slurry
to physically grind flat and chemically remove the microscopic topographic features on
a substrate so that subsequent processes can begin from a flat surface. The oxide
interface layer is also removed by CMP process as shown in Fig 3.1.
EXPERIMENTAL 27
In the CMP process the sample is put into a carrier head which is pressed onto a pad
covering the main plate. Both carrier and main plate rotate while a defined amount of
slurry is poured onto the pad. The slurry is a liquid including colloidal particles.
Between sample and pad a small film of slurry is formed. While the chemical
component modifies the surface at a homogeneous depth the combined mechanical
polishing effect of colloids and pad take off the most protruding parts. This combined
effect results in a very smooth surface.
A Logitech CDP machine and silicic acid based slurry (Syton
®
HT-50 colloidal silica
slurry from DuPont Air Product NanoMaterials L.L.C) were used to polish the samples.
After the CMP process the samples were cleaned using hot DI water (~ 80 °C) in an
ultrasonic bath. The standard parameters for the main plate and carrier rotation
velocities, the download pressure on the sample and the flow of slurry are shown in
table 3.2. The polishing process takes about 2 min and four samples in size of 1×1 inch
can be polished at once.
TABLE 3.2: Main CMP parameters.
Main plate Carrier Download pressure Slurry
60 RPM 40 RPM 0.5 ... 0.7 mbar 50 ml/min
EXPERIMENTAL 28
3.2 CHARACTERIZATION
A number of characterization techniques are used to assess the quality of the poly-Si
films and solar cells fabricated in this thesis. The characterization methods used in this
study can be divided roughly into four categories: structural, electrical, and optical
characterization and solar cell performance analysis (see chapter 6). In this section, the
measurement methods and analytical procedures are described.
In-situ optical microscopy, Raman spectroscopy, electron backscatter diffraction
(EBSD), and X-ray diffraction (XRD) were used to characterize structural properties.
In-situ optical microscopy (section 3.2.1) was used to analyze the kinetics of the
crystallization and to calculate an average estimated grain size of poly-Si films. Raman
spectra (section 3.2.2) were performed to characterize the crystalline quality of the
resulting poly-Si films. Electron backscatter diffraction (EBSD) was used to analyze the
crystallographic surface orientation and the grain size (section 3.2.3). X-ray diffraction
(XRD) was measured to investigate the crystallographic orientation (section 3.2.4). In
order to characterize the electrical properties of the ZnO:Al/poly-Si stacks formed using
the ALILE process 4-point probe and Hall measurements were used (section 3.2.5 and
3.2.6). UV-VIS spectroscopy (section 3.2.7) measurements were used to characterize
the optical and structural properties of poly-Si films on ZnO:Al films coated glass. For
the extended defect analysis, poly-Si films have been Secco-etched and then
investigated by Scanning Electron Microscopy (SEM) using a Hitachi S4100 (25 keV)
SEM equipped with a cold field emission cathode. The concentration of impurities
within the poly-Si layers was measured by secondary ion mass spectrometry (SIMS).
3.2.1 IN-SITU OPTICAL MICROSCOPY
Samples were annealed in the optical microscope to determine the process time, an
average estimated grain size and the kinetics of the crystallization of the ALILE process.
In-situ images were recorded within time intervals of 1 sec to 10 min, depending on the
EXPERIMENTAL 29
process temperature, through the glass substrates. The interface of the ZnO:Al layer and
the Al layer was investigated. The spatial resolution limit of the in-situ optical
microscopy is about 0.5 µm. Due to the high reflectivity of the Al the micrographs
appear bright at the beginning of the process. As soon as the size of the Si grains
exceeds the resolution limit of the microscope, the crystallized Si becomes visible as
small dark spots at the ZnO:Al/Al interface due to the lower reflectivity of Si compared
to Al. The black fraction of the images recorded is used as crystallized fraction. The
crystallized fraction curves are evaluated as function of the annealing time. The
crystallized fraction is 0% at the beginning of the process and corresponds to all Al at
the interface. The time when first dark areas are detected is defined as the nucleation
time. The process time is defined as the time required reaching a crystallized fraction of
99%. Counting the number n of black objects in the black and white image of area A
0
determines the nucleation density N
G
in each image (N
G
= n/A
0
). The maximum of the
nucleation density was used to evaluate the final average estimated grain size d
G
of
grains which are assumed to be square shaped:
GG
N/1d =
(3.1)
The setup of the in-situ optical microscopy and the evaluation methods of the obtained
images are described in detail in the doctoral thesis of J. Schneider [Sch05]. Five
assumptions are established to estimate the average grain size: (i) the maximum number
of grains is detected when no new grains are formed, (ii) at this point no grains have
coalesced, i.e. all grains are separated and counted, (iii) all grains are averaged in size,
(iv) a square shape of the grains is assumed, (v) the areas grown from one nucleus
correspond to one grain.
3.2.2 RAMAN SPECTROSCOPY
Raman is a powerful light scattering technique for characterizing the structural
properties such as the crystal quality of thin films [Sch90] or any stress present in thin
films [Nic00]. Light of a known frequency and polarization is scattered from a sample.
EXPERIMENTAL 30
The scattered light is then analyzed for frequency and polarization. Raman scattered
light is frequency-shifted with respect to the excitation frequency, but the magnitude of
the shift is independent of the excitation frequency. This "Raman shift" is therefore an
intrinsic property of the sample. Because Raman scattered light changes in frequency,
the rule of conservation of energy dictates that some energy is deposited in the sample.
A definite Raman shift corresponds to excitation energy of the sample (such as the
energy of a free vibration of a molecule). In general, only some excitations of a given
sample are "Raman active," that means, only some may take part in the Raman
scattering process. Hence the frequency spectrum of the Raman scattered light maps out
part of the excitation spectrum. For solids phonon excitation and crystal structure can be
identified. As can be seen in Fig 3.2, the normalized Raman peak of crystalline silicon
(c-Si) and amorphous silicon (a-Si) is 520 and 480 cm
-1
, respectively. Details about
Raman spectrum for thin Si films are descriebed in [Wu96, Bo02, Owe03].
450 500 550
0
1
Intensity [a.u]
Raman shift [cm-1]
a-Si
FZ c-Si
Figure 3. 2: Normalized Raman spectra of a-Si layer (2 µm) and FZ c-Si (320 µm).
EXPERIMENTAL 31
In this thesis Raman Spectroscopy (set-up: LABRAM from Jobin Yvon Horiba) with a
HeNe-Laser at wavelength of 632.8 nm was used as a standard measurement for the
crystallographic quality of poly-Si layer formed by the ALILE process. The resolution
of Raman spectroscopy is 1 cm
-1
. Lorentzian line shape is fitted to the Raman spectra of
poly-Si films formed using the ALILE process and FZ c-Si as a reference to get the
peak position and full width half maximum (FWHM). It was separately tested that the
laser beam did not induce any crystallization in a-Si films on glass even after extended
periods of exposure (over 1 hour).
3.2.3 ELECTRON BACKSCATTER DIFFRACTION (EBSD)
Electron Back Scatter Diffraction (EBSD) method was used to investigate the
crystallographic structure of the samples. The EBSD measurements are performed with
an SEM tool equipped with a phosphor screen detector. EBSD provides quantitative
microstructural information about the crystallographic nature of materials. It reveals
grain size, grain boundary character, and grain orientation of the sample under the beam.
EBSD operates by arranging a flat sample at a shallow angle, usually 20°, to the
incident electron beam (since the SEM stage is often used to tilt the plane of the sample
to this shallow angle, the value of stage tilt is often referred to and is typically 70°). Fig
3.3 shows the electron interaction with crystalline silicon. Electron diffraction occurs
from the incident beam point on the sample surface. With the beam stationary, an EBSD
pattern emanates spherically from this point. This EBSD pattern is the Kikuchi pattern.
The Kikuchi pattern contains the angular relationship between the planes, the symmetry
of the crystal and orientation information. Bright Kikuchi bands correspond to planes in
the crystal lattice. The width of bands depends upon the electron wavelength and the
lattice plane spacing. Relationship is given by the Bragg’s equation (
θ
λ
sin2dn
=
)
[Tro93]. The variable
λ
is the wavelength of the incident beam,
θ
is a certain angles of
the incidence beam, the variable
d
is the distance between atomic layers in a crystal, and
n is an integer determined by the order given.
EXPERIMENTAL 32
70o
Camera
Highly
polished
specimen
surface
tiltied70o
Electronbeam
Figure 3. 3: Electron interaction with crystalline silicon [Day01].
The EBSD conditions used in this thesis are an acceleration voltage of 20 kV, a spot
size of 5.4 µm and a step size of 0.2 µm. A matrix of 400×400 points was measured for
all samples corresponding to an area of 80×80 µm
2
. For determining the grain sizes of
samples, only grains with a size of at least 0.6 µm (3 times the step size) were taken into
account. EBSD measurements were in cooperation with IMEC in the framework of the
European project ATHLET.
3.2.4 X-RAY DIFFRACTION SPECTROSCOPY
X-ray diffraction measurements were carried out to investigate the orientation of the Si
grains of the polycrystalline material. X-ray diffraction (XRD) is a powerful non-
destructive diagnostic technique analyzing the crystalline phases of thin-films and
determining the structural properties of these phases such as the preferred orientation
and average grain size [Klu74, Sch90]. X-ray diffractograms of the poly-Si films were
measured in symmetric θ-2θ geometry. These X-rays are generated by a cathode ray
tube, filtered to produce monochromatic radiation, collimated to concentrate, and
EXPERIMENTAL 33
directed toward the sample. The interaction of the incident rays with the sample
produces constructive interference (and a diffracted ray) when conditions satisfy
Bragg's Law (nλ=2d sin θ). Here d is the spacing between diffracting planes in a
crystalline sample, θ is the incident angle, n is any integer, and λ is the wavelength of
the beam. Fig 3.4 illustrates schematically the interference between waves scattered
from two adjacent rows of atoms in a crystal. This law relates the wavelength (λ) of
electromagnetic radiation to the diffraction angle (θ) and the lattice spacing (d) in a
crystalline sample. h is a half of the path difference between the two reflected rays.
These diffracted X-rays are then detected and counted.
For the thin film silicon samples the angle 2θ is scanned through a range of 20 - 100°.
The lattice spacing in the films can be determined from the position of the detected
peaks. By comparison of the relative peak intensities with the spectrum of a powder
sample in which a random orientation should be present, preferred orientation of crystal
grains can be determined.
q
d
h
Figure 3. 4: Schematic diagram for determining Bragg's law.
EXPERIMENTAL 34
3.2.5 4-POINT PROBE MEASUREMENT
The 4-point probe is a simple technique used to measure the resistivity of a thin film on
an insulating material. By forcing a current through two outer probes and measuring the
voltage between the inner probes the substrate resistivity can be measured. While
simple in principle, there are experimental issues to take into account when using a 4-
point probe. Very high or very low resistivity samples require adjustment of the current
to obtain a reliable measurement.
3.2.6 HALL MEASUREMENT
The conductivity, the free carrier concentration and the mobility of samples were
determined with Hall measurements. The measurements (magnetic flux of 0.64 T) were
performed on 5×5 mm
2
samples and in the Van der Pauw geometry. In order to form
ohmic contacts gold contacts with a thickness of 20 nm were deposited at room-
temperature by thermal evaporation. An overview of the measuring principle and
experimental details are given in [Blo92]. The Van der Pauw technique can be used to
measure resistivity of a thin, arbitrary-shaped sample with 4 ohmic contacts placed on
the periphery. Van der Pauw demonstrated that there are two characteristic resistances
associated with the four terminals.
The measurement is performed as follows:
First the sheet resistance R
S
is calculated. This is related to the resistivity ρ by the film
thickness t,
]sq/[
t
R
S
Ω
ρ
=
(3.2)
Next, a current I is applied between two contacts and a perpendicular magnetic field B
is applied. The Hall voltage V
H
is measured across the two remaining contacts. The
schematic of the Hall measurement is illustrated in Fig 3.5.
EXPERIMENTAL 35
From this the sheet carrier density N
S
can be calculated:
]cm[
Vq
BI10
N
2
H
4
S
−
−
⋅
=
(3.3)
where the sheet carrier density N
S
is given by N
e
×t, N
e
is the carrier density [cm
-3
], and
q is the charge. Once R
S
and N
S
are known the Hall mobility can be calculated,
]sVcm[
RqN
1
112
SS
H
−−
=µ
(3.4)
Further details on this Hall measurement can be taken from [Pau58].
There are practical aspects which must be considered when carrying out Hall and
resistivity measurements. Primary concerns are (1) ohmic contact quality and size, (2)
sample uniformity and accurate thickness determination, (3) thermo-magnetic effects
due to non-uniform temperature, and (4) photoconductance and photovoltaic effects
which can be excluded by measuring in a dark environment. Also, the sample lateral
dimensions must be large compared to the size of the contacts and the sample thickness.
VH
VH
I
B
d
Figure 3. 5: The Hall effect measurement.
EXPERIMENTAL 36
3.2.7 UV-VIS SPECTROSCOPY
UV-VIS-IR spectroscopy (Lamba19 from Perkin-Elmer, spectral range: 250 ~ 2000 nm,
double-beam spectrophotometer with an integrating sphere) was used in order to
investigate the following: (i) the quality of poly-Si film formed using the ALILE
process and (ii) the properties of the ZnO:Al layer.
The characteristic peaks in the ultraviolet (UV) reflectance at ~ 360 and ~ 275 nm (e1
and e2, respectively) are related to direct optical transitions at the critical points in c-Si
and hence are a measure of the crystalline quality of the silicon material investigated
(note that defects in the material lead to a decrease and broadening of the peaks
[Kam88]).
The absorption in near infrared (NIR) is related to the free carriers of ZnO:Al. Near-
infrared (NIR) spectroscopy offers the characterization of the free carrier properties
since this spectral domain is well described by the Drude theory of free electrons,
allowing the determination of the free carrier density and local conductivity [Cou00].
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 37
CHAPTER 4
T
EMPERATURE STABILITY OF
ZnO:Al/POLY-Si STACKS
This chapter investigates the sputtered ZnO:Al films on glass which are capped with a
poly-Si layer formed during a thermal treatment. Two types of glass/ZnO:Al/poly-Si
samples were investigated: The poly-Si thin film was formed by either (i) the ALILE
process or (ii) the SPC process. The sample preparation for the ALILE process is
explained in chapter 3.1. For SPC process the sample preparation will be described later.
The most important finding of this chapter is that the capped ZnO:Al film can withstand
the high temperature treatments while its electrical properties even improve.
4.1 MOTIVATION
In the case of the aluminium-induced layer exchange (ALILE) process, an annealing
step is necessary so that only temperature-stable substrates can be used. The formation
of poly-Si films on transparent conductive oxides (TCOs), however, would be an
appealing option, especially for the preparation of thin-film solar cells because they
allow for a simple contacting scheme and light trapping [Mue04, Rec06]. Among the
available TCO materials, SnO
2
:F and ZnO:Al films are commonly used as front contact
material in a-Si:H solar cells and are considered here. The ALILE process is processed
in nitrogen ambient and at higher temperature compared to amorphous silicon (a-Si)
and/or microcrystalline silicon (µc-Si) solar cells. Studies on the stability of ZnO:Al
upon treatment at temperatures above the deposition temperature, as used for the
crystallization of Si, have so far concentrated on annealing of ZnO:Al films on glass
under various ambient conditions. Usually a strong decrease of electrical conductivity is
observed during annealing in air at temperatures between 300 and 400 °C [Min90,
Cha01], while much higher temperatures can be applied in vacuum [Iga88] or, in the
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 38
case of rapid thermal annealing, nitrogen [Kim05]. As the films do not show structural
degradation but rather an improved crystallinity, Minami et al. [Min99] suggested
oxygen to be responsible for both the decreased carrier density and mobility upon
annealing in air.
Recently the formation of poly-Si layers on ZnO:Al coated glass using the ALILE
process has been introduced [Dim07], but no comments on the evolution of ZnO:Al
properties were given. The formation of poly-SiGe layers on ZnO:Al coated glass using
the ALILE process has been introduced by Lechner [Lec03]. He concluded that Zinc
oxide proved to be thermally stable in the pertaining temperature region (550 °C), but
the removal of aluminium residuals after the ALILE process turned out to be
problematic.
4.2 PROPERTIES OF ZnO:Al LAYERS
First of all, in order to study the influence of the annealing temperature (425 ~ 525 °C)
on the properties of the ZnO:Al films used for this study, the electrical properties of
ZnO:Al films deposited on glass without Si on top before and after annealing were
measured. The sheet resistance of ZnO:Al films on bare glass as a function of annealing
time and annealing temperature is shown in Fig 4.1. The sheet resistances were
measured by 4-point probe. The sheet resistance of as-deposited ZnO:Al films was 6.3
/sq. The ZnO:Al films on glass were annealed in a tube furnace with nitrogen ambient
for 0 to 16 hours. After one hour annealing the sheet resistance increased by about one
order of magnitude. The influence of the annealing temperature is relatively small. For
long annealing times the sheet resistance depends strongly on the annealing temperature.
The sheet resistances of the samples annealed at 450 °C have similar values as the
samples annealed at 425 °C. But if the annealing temperature is higher than 450 °C, the
sheet resistances of the annealed ZnO:Al films are significantly increased with
increasing annealing temperature. The sheet resistance of the ZnO:Al film annealed at
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 39
525 °C for 16 hours increased to ~ 10
4
/sq (degradation of more than 3 orders of
magnitude).
Compared to results obtained for annealing in vacuum [Yu05, Ber06] the annealing in
nitrogen atmosphere used in this work leads to a faster degradation. Hamad et al.
[Ham05] concluded that the changes of the electrical properties upon annealing in
nitrogen ambient are due to both, the reduction of the free carrier density and the
reduction of carrier mobility. One possible reason is a chemical reaction with residual
oxygen within the nitrogen gas which is used in order to prevent a contamination of
oxygen from outside [Ber08].
0 4 8 12 16
10
0
10
1
10
2
10
3
10
4
10
5
Sheet resistance [
Ω
/sq]
Annealing time [hours]
Annealing temperature
525
o
C
500
o
C
475
o
C
450
o
C
425
o
C
N
2
ambient
Figure 4. 1: Sheet resistances of ZnO:Al films on borofloat33 glass as a function of
annealing time and annealing temperature.
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 40
4.3 PROPERTIES OF ZnO:Al/POLY-Si STACKS
In this section, the properties of ZnO:Al films capped with Si are investigated. For the
ALILE-based experiments the sample is annealed in a tube furnace with nitrogen
ambient at 425 and 450 °C for 16 hours and at 475…525 °C for 16 hours. For the SPC-
based experiments intrinsic a-Si layers were deposited on ZnO:Al coated glass by
electron beam (e-beam) evaporation at room temperature. The thickness of the a-Si
layer was about 290 nm. The initial glass/ZnO:Al/a-Si stacks were annealed in a tube
furnace at 600 °C for 24 hours in nitrogen ambient. During annealing the initially
amorphous silicon was crystallized and therefore finally glass/ZnO:Al/poly-Si stacks
were formed. For comparison, the different annealing steps were also applied to (i)
glass/ZnO:Al samples, (ii) glass/Al/a-Si samples (ALILE process without ZnO:Al), and
(iii) glass/a-Si samples (SPC process without ZnO:Al). The sheet resistances were
measured by a 4-point probe.
Fig 4.2 shows the sheet resistance of the different samples as a function of the annealing
temperature: glass/ZnO:Al (open circles), glass/poly-Si (open triangles), and
glass/ZnO:Al/poly-Si (solid squares). For comparison the sheet resistance of an as-
deposited ZnO:Al layer (6.3 /sq) is indicated with a dashed line. The sheet resistance
of the annealed glass/ZnO:Al samples (open circles) is significantly increased compared
to the as deposited value despite the nitrogen ambient as shown in Fig 4.1, in which
ZnO:Al is known to more stable as compared to air or oxygen ambient [Ham05]. In
contrast, the sheet resistances of the glass/ZnO:Al/poly-Si samples (solid squares) are
even lower than the sheet resistance of as deposited ZnO:Al layer. This is due to an
increased conductivity of the ZnO:Al as the poly-Si on top does not contribute to the
total resistance of the stack considerably. The resulting sheet resistance of the
glass/ZnO:Al/poly-Si samples formed at 425 °C by the ALILE process and at 600 °C by
the SPC process is around 4 /sq. Hence, the Si layer on top of the ZnO:Al effectively
prevents the degradation of the ZnO:Al films. For the SPC sample annealed at 600 °C
for 24 hours the difference between the sheet resistance of the glass/ZnO:Al sample
(open circle) and the glass/ZnO:Al/poly-Si sample (solid square) is more than three
orders of magnitude as shown in Fig 4.2. This very promising result strongly motivated
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 41
further effects to study the application of sputtered ZnO:Al layers in poly-Si thin-film
solar cells because the capped ZnO:Al layers do not degrade and provide a high quality
transparent front contact.
In addition to the electrical properties the influence of the annealing on the optical
properties was studied. Fig 4.3 shows the absorption spectra of three different samples
(SPC experiments): (i) glass/ZnO:Al before annealing (dashed line), (ii) glass/ZnO:Al
after annealing (solid line) at 600 °C for 24 hours, (iii) glass/ZnO:Al/poly-Si after
annealing (solid circle) at 600 °C for 24 hours. It can be clearly seen that the heat
treatment of the glass/ZnO:Al samples leads to a disappearance of free carrier
absorption in the near infra-red (NIR) region indicated by an arrow. The reduction of
free carriers can explain the strong increase of resistivity as can be seen in Fig 4.2. For
the glass/ZnO:Al/poly-Si stack, however, the free carrier absorption in the NIR is
preserved although the stack has been annealed for the same time. Bare poly-Si films
have no absorption in this spectral region. In case of the glass/ZnO:Al/poly-Si sample
the presence of the absorption in the ultra-violet (UV) region is caused by the
absorption of the poly-Si film.
For a better understanding of the change of the electrical properties during
crystallization Hall measurements were carried out on three samples, namely the as-
deposited ZnO:Al film, the glass/ZnO:Al/poly-Si stack formed by ALILE at 425 °C for
16 hours (sample was treated by CMP after the ALILE process) and the
glass/ZnO:Al/poly-Si stack formed by SPC at 600 °C for 24 hours. The resistivity of the
glass/ZnO:Al/poly-Si (ALILE) and the glass/ZnO:Al/poly-Si (SPC) sample was
(2.2±0.1)×10
-4
cm and (3.4±0.1)×10
-4
cm, respectively. This is lower than the
resistivity of the as-grown glass/ZnO:Al sample ((4.3±0.1)×10
-4
cm). It should be
noted that in all cases the Si layer does not contribute to the electrical transport
noticeably, so the measured values could be solely attributed to the ZnO:Al layer. Table
4.1 shows resistivity ρ, carrier concentration N
e
and hall mobility µ
H
. As can be seen
from Table 4.1, while the decreased resistivity of the ALILE sample is due to higher
carrier concentration N
e
in the ZnO:Al film, no change in carrier concentration can be
observed for the SPC sample. Instead the mobility µ increases strongly. The measured
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 42
value of 52.6 cm
2
/Vs is close to the theoretical limit proposed by Ellmer [Ell01], and
thus higher than the experimental limit investigated by C. Agashe [Aga04]. This
indicates that two different processes are responsible for the decrease of the resistivity.
The higher mobility for the SPC sample is most likely caused by an improved
crystallinity of the ZnO:Al film after the annealing. The increased electron
concentration of the ALILE sample probably results from a diffusion of Al into the
ZnO:Al film during the annealing and thus is a peculiarity of the ALILE process. The
effect of the improved crystallinity of the ZnO:Al film expected after the annealing on
the mobility is probably compensated by increased ionized impurity scattering.
425 450 475 500 525 600
10
0
10
1
10
2
10
3
10
4
10
5
10
6
10
7
4 h
Sheet resistance [
Ω
/sq]
Annealing temperature [
o
C]
as-deposited ZnO:Al layer
16 h
ALILE SPC
(24h)
Figure 4. 2: The sheet resistance of glass/ZnO:Al/poly-Si (squares), glass/ZnO:Al
(circles) and glass/poly-Si (triangles) stacks as a function of the annealing temperature.
For reference, the sheet resistance of the glass/ZnO:Al layers before annealing is
indicated with a dashed line. The poly-Si films were formed by either ALILE (450 -
525 °C) or SPC (600 °C).
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 43
250 500 750 1000 1250 1500
0
20
40
60
80
100 T
A
= 600
o
C for 24 hours
Glass/ZnO:Al (as-deposited)
Glass/ZnO:Al (annealed)
Glass/ZnO:Al/poly-Si (annealed)
Absorption [%]
Wavelength [nm]
Figure 4. 3: Absorption spectra of ZnO:Al films on glass before (dashed line) and after
(solid line) annealing, as well as of a glass/ZnO:Al/poly-Si stack after annealing (solid
circles). Annealing was performed at 600 °C for 24 hours in nitrogen ambient (SPC
experiments). The arrow highlights that the remaining free carrier absorption of the
glass/ZnO:Al/poly-Si stack is maintained after annealing. In case of the
glass/ZnO:Al/poly-Si sample the presence of the absorption in the ultra-violet (UV)
region is caused by the absorption of the poly-Si film.
TABLE 4.1: Resistivity ρ, carrier concentration N
e
and hall mobility µ
H
of the as
deposited ZnO:Al film and ZnO:Al films coated with poly-Si produced by the ALILE
and the SPC process.
Sample Crystallization
conditions
ρ
[×10
-4
cm]
N
e
[×10
20
cm
-3
] µ
H
[cm
2
/Vs]
ZnO:Al as dep.
- 4.3 ± 0.1 3.5 ± 0.1 42.0 ± 0.1
SPC 24 h @ 600 °C 3.4 ± 0.1 3.5 ± 0.1 52.6 ± 0.1
ALILE 16 h @ 425 °C 2.2 ± 0.1 6.8 ± 0.2 41.8 ± 0.2
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 44
4.4 ELECTRICAL PROPERTIES OF ZnO:Al/POLY-Si
STACKS FORMED BY THE ALILE PROCESS
Fig 4.4 shows the resistivity, the hall mobility and the carrier concentration of
ZnO:Al/poly-Si stacks which were formed by the ALILE process as a function of the
annealing temperature. The black straight line is a linear fit to the open triangles data
points to emphasize the increasing carrier concentration with increasing annealing
temperature. Thus, it does not mean that the increasing carrier concentration is only due
to the increasing Al concentration in the ZnO:Al films. In the case of ZnO:Al, the
carrier concentration is not the same as the Al concentration. Usually, the Al
concentration is a factor of 2-10 higher than the carrier concentration. This increasing
carrier concentration is accompanied by a decreasing Hall mobility. The highest
mobility value of 42 cm
2
/Vs was obtained for a ZnO:Al/poly-Si stack formed at 425 °C
and equals the as-deposited value. However, annealing at 525 °C leads to a mobility of
33 cm
2
/Vs. Again, the red dashed line is a linear fit to the open squares to show the
general trend of decreasing Hall mobility. Nevertheless, measured Hall mobilities were
typically between 39 and 42 cm
2
/Vs for all the samples up to 500 °C. These values are
in the range of the mobility of single ZnO crystals which have similar doping levels
[Ell08]. It is unclear why the sample annealed at 525 °C shows a different behavior.
In order to investigate whether Al atoms diffuse into the ZnO film during the ALILE
process the sheet resistance of poly-Si on intrinsic ZnO (i-ZnO) coated glass is
compared with that of poly-Si on ZnO:Al coated glass. The sheet resistance of (i)
glass/i-ZnO, (ii) glass/i-ZnO/poly-Si, (iii) glass/ZnO:Al, (iv) glass/ZnO:Al/poly-Si
stacks and (v) glass/poly-Si were measured by 4-point probe. The results are
summarized in Table 4.2. The sheet resistance of the i-ZnO film could not be measured
due to a high sheet resistance (> 1 M). After ALILE process the sheet resistance of the
ZnO/poly-Si stack is more than one order lower than the sheet resistance of the poly-Si
film on glass. The sheet resistance of i-ZnO/poly-Si stack and poly-Si film on glass is
around 400 /sq and around 10 k/sq, respectively. The change of the sheet resistance
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 45
of glass/i-ZnO/poly-Si could be explained by Al diffusion into the i-ZnO film during the
ALILE process.
298 300 400 425 450 475 500 525 550
1
2
3
4
5
30
35
40
45
50
2
4
6
8
10
Resistivity
ρ
[
x
10-4
Ω
cm]
Annealing temperature [oC]
Glass/ZnO:Al/poly-Si
as-deposited ZnO:Al
16 hrs 4 hrs
Annealing time
Hall mobility
µ
H [cm2/ Vs]
µ
Η
ρ
Carrier concentration N
e
[
x
1020 cm-3]
N
e
Figure 4. 4: Resistivity ρ (open circle), Hall mobility µ
H
(open square), and carrier
concentration N
e
(open triangle) of ZnO:Al/poly-Si stacks formed by the ALILE
process as a function of annealing temperature. For comparison the resistivity (solid
circle), Hall mobility (solid square), and carrier concentration (solid triangle) of the
ZnO:Al film deposited on glass at 300 °C are shown.
TABLE 4.2: Sheet resistance of (i) glass/i-ZnO (as-deposited), (ii) glass/i-ZnO/poly-Si,
(iii) glass/ZnO:Al (as-deposited), (iv) glass/ZnO:Al/poly-Si, and (v) glass/poly-Si stacks.
The poly-Si films were formed at 425
o
C for 16 hours in nitrogen ambient.
Structure Sheet resistance
[/sq]
Glass/i-ZnO (as-deposited) > 1M
Glass/i-ZnO/poly-Si 413.4
Glass/ZnO:Al (as-deposited) 3.8
Glass/ZnO:Al/poly-Si 3.5
Glass/poly-Si ~ 10k
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 46
4.5 COMPARISON
The hall mobility of ZnO:Al/poly-Si stacks formed at 425 °C for 16 hours by the
ALILE (star) and at 600 °C for 24 hours by SPC (solid circle) and the hall mobility of
the as-deposited ZnO:Al on glass (open circle) are shown in Fig 4.5. The mobility is
plotted as a function of carrier concentration. Indicated are mobility data from various
publications [Ell06, Bre03, Kon03, Min92]. The data for single crystals were fitted by
Ellmer using Masetti’s formular [Ell05], resulting in the green line. Additionally the
influence of grain boundary scattering for polycrystalline layers was calculated for
different trap densities (blue and red straight lines). The trap density is 1.3×10
13
cm
-2
for
the blue line and 3×10
13
cm
-2
for the red line.
The Masetti mobility model µ
Hma
[Mas83] was used to simulate the doping dependent
mobility in Si and takes into account the scattering of the carriers by charged impurity
ions which lead to a degradation of the carrier mobility (ionized impurity scattering). It
is a model that combines lattice and impurity scattering.
2
2ref
1
1
1ref
minmax
min
Ma
H)n/n(1)n/n(1 αα +
µ
−
+
µ−µ
+µ=µ
(4.1)
where
max
is the lattice mobility at low carrier concentrations,
min
is the ionized
impurity mobility at higher carrier concentrations, and
min
-
1
is the clustering mobility
at very higher carrier concentrations. n is the carrier concentration.
The dotted line in Fig 4.5 by M. Kon et al. was fitted by Brooks–Herring–Dingle (BHD)
theory [Kon03]. The BHD theory is based on charged impurity scattering.
I
2
)x(
2*3
32
r0
3
1nZgme
n)(24
ε
ε
π
=µ
(4.2)
where ε
0
and ε
r
are the dielectric constant of free space and the relative permittivity, and
Z and n
I
are the charge and the density of the ionized scattering centers, respectively.
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 47
m* and g
(x)
is the effective mass and the screening function, respectively. n is the
integer and h is Planck’s constant.
60
40
20
0
1018 24 6 8
1019 24 6 8
1020 24 6 8
1021
Hallmobility[cm/Vs]
2
Chargecarrierdensity[cm ]
-3
Figure 4. 5: Hall mobilities achieved in this thesis are shown for ZnO:Al/poly-Si stack
formed at 425 °C for 16 hours by the ALILE () and at 600 °C for 24 hours by the SPC
() process and for the as-deposited ZnO:Al film on glass ().The straight green line is
the hall mobility of single crystal ZnO:Al which is fitted by Masetti’s formula [Mas83].
The dotted line was fitted by M. Kon [Kon03]. For comparison mobility data of other
groups were added, which have been measured for ZnO films deposited by magnetron
sputtering: Ellmer (□) [Ell06], Minami (, #) [Min92], Brehme et al. (, ∇) [Bre03]
and Kon et al. () [Kon03]. The mobility values of ZnO single crystals are indicated by
green triangles. Additionally the influence of grain boundary scattering for
polycrystalline layers was calculated for different trap densities (blue and red straight
lines). The trap density is 1.3×10
13
cm
-2
for the blue line and 3×10
13
cm
-2
for the red line.
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 48
The hall mobility of ZnO:Al/poly-Si stacks formed at 425 °C for 16 hours by the
ALILE (star) and at 600 °C for 24 hours by SPC (solid circle) process are comparable to
the hall mobility of single ZnO crystals showing the high quality of the films. Ellmer et
al. showed the improvement of the hall mobility of annealed ZnO:Al samples at 500 °C
in vacuum and concluded that no recrystallization occurred and instead, it is plausible
that point defects or dislocations have been annealed, reducing the scattering at such
centers [Ell05]. Thus, the improvement of the hall mobility of ZnO:Al/poly-Si stack
annealed at 600 °C in a furnace with nitrogen ambient seems to be a reduction of points
defect or dislocations within ZnO:Al layer during an annealing process.
4.6 INFLUENCE OF SiN
X
AS A BARRIER LAYER
SiN
X
layers were applied to prevent a contamination with impurities from the glass
substrate during the ALILE process [Abe06]. Resistivity ρ, carrier concentration N
e
and
hall mobility µ
H
of ZnO:Al/poly-Si stacks with and without SiN
X
as a barrier layer are
summarized in Table 4.3. For comparison, resistivity
ρ
, Hall mobility µ
H
, and carrier
concentration N
e
of as-deposited ZnO films on glass are shown. The resistivity, hall
mobility, and carrier concentration did not change significantly in both cases (with and
without SiN
X
layer). SiN
X
does not influence the properties of the ZnO:Al.
TABLE 4.3: Resistivity ρ, Hall mobility
H
, and carrier concentration N
e
of
ZnO:Al/poly-Si stack formed on glass and on SiN
X
coated glass at 425 °C. For
comparison resistivity, Hall mobility, and carrier concentration of as-deposited ZnO
film on glass is shown.
Sample ρ
[×10
-4
cm]
N
e
[×10
20
cm
-3
]
µ
H
[cm
2
/Vs]
ZnO:Al as dep. 4.3 ± 0.1 3.5 ± 0.1 42.0 ± 0.1
Glass/ZnO:Al/poly-Si 2.4 ± 0.1 6.8 ± 0.2 41.8 ± 0.2
Glass/SiN
X
/ZnO:Al/poly-Si
2.3 ± 0.1 7.3 ± 0.2 38.8 ± 0.2
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 49
4.7 INFLUENCE OF POST TREATMENTS
The investigations shown so far demonstrated that the capped ZnO:Al films can
withstand process temperatures up to 650 °C. However, high quality poly-Si further
process steps - i.e. Hydrogen (H)-passivation and rapid thermal annealing (RTA). The
influence of these process steps on the electrical properties of the ZnO:Al layer are
described in the following. For this investigation a 1"×1" sample after the CMP process
was divided into 4 pieces (0.5"×0.5"). Each piece was treated with different post
treatments. The electrical resistivity (triangles), Hall mobility (circles), and carrier
concentration (squares) of the ZnO:Al/poly-Si stack formed on SiN
X
coated glass at
425 °C for 16 hours with different kinds of post treatments (H-passivation, RTA, and
RTA & H-passivation) are shown in Fig 4.6. For the H-passivation the hydrogen plasma
was applied for 10 minutes at a substrate temperature of 600 °C. The RTA process was
applied at 850 °C for 200 seconds. Even though the electrical properties of
ZnO:Al/poly-Si stack did not change significantly after H-passivation, the electrical
properties of ZnO:Al/poly-Si stack were improved dramatically after RTA treatment.
The H-passivation is practically interesting, as hydrogen has been identified as a donor
in ZnO [Wal03]. The theoretical findings are backed by many experimental results.
Thomas et al. [Tho56] investigated the influence of hydrogen on the conductivity of
ZnO single crystal. They annealed the ZnO single crystal at 650 °C for a short time in
hydrogen ambient. They found that the conductivity in ZnO is increased by the
hydrogen. Also Oh et al. [Oh05] investigated the influence of H-passivation on the
electrical properties of ZnO:Al layer. They applied an H-passivation process at 300 °C
for 10 ~ 120 min. They found that the resistivity of the ZnO:Al layer is reduced after the
H-passivation. The influence of hydrogen gas during the deposition of Al doped and
undoped ZnO:Al layers on the hall mobility and carrier concentration was investigated
by many researchers [Due08, Myo03]. They found a decrease of the Hall mobility of
the Al doped and undoped ZnO layer with increasing hydrogen content during the
deposition. The decrease of the Hall mobility could be explained by excessive hydrogen
atoms acting as centers of interstitial sites in the lattice or let to an increase ionized or
neutral impurity scattering. Both Al and H could constitute scattering centers or
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 50
combine to form the neutral defect complexes [Li05]. It has been stated by several
groups that the carrier scattering at ionized impurities limits the mobility in the ZnO:Al
films for carrier concentrations above 10
19
cm
-3
[Bel92, Min00, Ell01, Ell08]. From the
literature the increase of carrier concentration and the decrease of the Hall mobility
could be expected after H-passivation treatment. The results obtained here, show that
the resistivity of ZnO:Al/poly-Si stack was slightly improved after H-passivation. But
the improvement of the resistivity is within the error range. The Hall mobility and the
carrier concentration of the as-grown ZnO:Al/poly-Si stacks and the RTA treated
ZnO:Al/poly-Si did not exhibit significant changes after H-passivation. The reason
could be that the highly Al doped poly-Si thin film acts as a barrier layer and prevents a
diffusion of H atoms into the ZnO:Al layer.
1
2
3
4
5
30
40
50
60
2
4
6
8
10
RTA &
H-passivation
RTA
H-passivation
µ
H
N
e
Resistivity
ρ
ρ
ρ
ρ
[
[
[
[
x
10-4
Ω
Ω
Ω
Ω
cm
]
]
]
]
ρ
as-grown
Hall mobility
µ
µ
µ
µ
H
[
[
[
[
cm2/ Vs
]
]
]
]
Carrier concentration Ne
[
[
[
[
x
1020 cm-3
]
]
]
]
Glass/ZnO:Al/poly-Si
TA = 425 oC
Figure 4. 6: Electrical resistivity ρ (squares), Hall mobility µ
H
(circles), and carrier
concentration N
e
(triangles) of ZnO:Al/poly-Si stacks formed on SiN
X
coated glass at
425 °C for 16 hours with different kinds of post treatment (H-passivation and RTA).
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 51
The hall mobility of ZnO:Al/poly-Si stack after RTA treatment was increased compared
to the hall mobility of the as-grown ZnO:Al/poly-Si stack. The hall mobility of
ZnO:Al/poly-Si stack before and after RTA treatment is 36 and 53 cm
2
/Vs, respectively.
This result is well matched with the results by Kim et al. [Kim05b]. They investigated
the influence of RTA at 900 °C for 3 min in nitrogen ambient on the electrical
properties of Al doped and undoped ZnO layer. They concluded that the enhancement
of the mobility could be attributed to the combined effects of the improved crystallinity
and the activation of Al dopants through de-oxidation of Al-oxide. As can be seen in
Fig 4.7 the Hall mobility of ZnO:Al/poly-Si stack after H-passivation and after RTA
treatment is 34 and 56.4 cm
2
/Vs, respectively. It is concluded that in the ZnO:Al/poly-Si
stack the RTA is more effective than the H-passivation in order to improve the Hall
mobility and the resistivity of the ZnO:Al layer. From this results it is extracted that
ZnO:Al/poly-Si stack withstand post-treatments (RTA and H-passivation), ZnO:Al can
be improved by RTA and can be applied for poly-Si thin film solar cells.
4.8 APPLICATION OF SnO
2
:F FILMS
SnO
2
:F thin films could be an alternative to ZnO:Al layers as mentioned in the
beginning of this chapter. Kawashima et al. [Kaw03] investigated the dependence of the
resistivity of the SnO
2
:F layer grown on Corning7059 glass substrate on annealing
temperature (up to 600 °C) for 1 hour in the air. The resistivity of SnO
2
:F layer was
maintained after annealing. Bae et al. [Bae07] investigated the sheet resistance of the
SnO
2
:F layer in different ambient at 450 °C for 30 min. The sheet resistance of the
SnO
2
:F layer does not depend on the ambient. Thus SnO
2
:F thin films should be suitable
for the high temperature process in poly-Si formation.
The sheet resistance of SnO
2
:F/poly-Si stacks formed using the ALILE process and the
SPC process were investigated and compared. Fig 4.7 shows the sheet resistance of
glass/SnO
2
:F/poly-Si stacks (solid squares), glass/SnO
2
:F (open circles) as a function of
annealing temperature. The poly-Si films were formed by either ALILE at 425 and
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 52
450 °C for 8 hours or SPC at 600 °C for 24 hours. The SnO
2
:F layers were deposited on
borofloat33 glass (from Schott) using atmosphere pressure chemical vapor deposition
(APCVD) method by Saint-Gobain-Recherche (SGR) in France. The thickness of
SnO
2
:F films used in this experiment is about 450 nm. For SPC experiments the
thickness and the B doping concentration of amorphous silicon (a-Si) films deposited
using electron beam (e-beam) evaporation is 290 nm and 4×10
16
cm
-3
, respectively. The
sheet resistance of the SnO
2
:F film in the as-deposited state is 27.3 /sq. The sheet
resistance of the SnO
2
:F/poly-Si stack formed at 425 °C and 450 °C for 8 hours by the
ALILE process is about 33 and 50 /sq, respectively. The sheet resistance of the
SnO
2
:F/poly-Si stack formed by the SPC process is around 70 /sq. The sheet
resistance of the glass/SnO
2
:F/poly-Si stacks treated in parallel increases with
increasing temperature. However, the sheet resistance of the glass/SnO
2
:F/poly-Si
stacks is still suitable for the application in solar cells. The sheet resistance of the poly-
Si films formed on bare glass using ALILE process is about 10 k/sq. The reason for
the degradation of the SnO
2
:F layer capped with poly-Si formed by either the ALILE
process or the SPC process is unclear.
For the properties of SnO
2
:F/poly-Si stack the following study will be investigated in a
future.
1. The interface of SnO
2
:F layer and poly-Si layer is investigated in order to
understand the degradation of SnO
2
:F layers capped with Si.
2. Electrical properties (i.e. Hall mobility, carrier concentration and resistivity) of
SnO
2
:F/poly-Si stack are investigated.
3. Optical properties (i.e. absorption and HAZE) of SnO
2
:F/poly-Si stack are
investigated.
4. Characterization of solar cells on SnO
2
:F coated substrate
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 53
425 450 600
20
40
60
80
Sheet resistance
[
[
[
[
Ω
Ω
Ω
Ω
/sq
]
]
]
]
Annealing temperature
[
[[
[
o
C
]
]]
]
Glass/SnO
2
:F/poly-Si
Glass/SnO
2
:F
SPC
ALILE
Figure 4. 7: The sheet resistance of glass/SnO
2
:F/poly-Si stacks (solid squares), glass/
SnO
2
:F (open circles) as a function of annealing temperature. The poly-Si films were
formed by either ALILE (at 425 and 450 °C for 8 hours) or SPC (at 600 °C for 24
hours).
4.9 CONCLUSION
It was shown that crystallization of a-Si layers can also be carried out on ZnO:Al coated
glass substrates using both ALILE and SPC. While uncoated ZnO:Al films show a
strong increase of the sheet resistance upon heat treatment, ZnO:Al layers with Si on top
used in this study resulted in electrical properties that were not only stable but
considerably improved. While for SPC this is a consequence of a higher mobility, a
strong increase of carrier density was observed during the ALILE process. The
temperature-stable conductivity of ZnO:Al/Si stacks opens up appealing options for
poly-Si thin-film solar cells including transparent conductive oxides (TCOs).
TEMPERATURE STABILITY OF ZnO:Al/POLY-Si STACKS 54
In the investigation of the influence of post-treatments (H-passivation and/or RTA) the
electrical properties of ZnO:Al/poly-Si stack were not changed by H-passivation while
the electrical properties (mobility and resistivity) of ZnO:Al/poly-Si stacks were
improved by RTA treatment. The highst mobility of ZnO:Al/poly-Si stack was achieved
56.4 cm
2
/Vs after the RTA treatment.
Although the sheet resistance of the glass/SnO
2
:F/poly-Si stacks treated in parallel
increases with increasing temperature, the sheet resistance of the glass/SnO
2
:F/poly-Si
stacks is still suitable for the application in solar cells.
POLY-Si FILMS ON ZnO:Al COATED GLASS 55
CHAPTER 5
P
OLY-Si FILMS ON ZnO:Al COATED GLASS
In this chapter the properties of poly-Si films formed on ZnO:Al coated glass and on
glass are characterized and compared in order to understand and optimize the
aluminium induced layer exchange (ALILE) process for poly-Si thin film solar cells.
5.1 KINETICS OF CRYSTALLIZATION
In this section the kinetics of the crystallization on glass and on ZnO:Al coated glass in
the aluminium-induced layer exchange (ALILE) process is studied in order to
understand the influence of the ZnO:Al layer. Studies of the nucleation and grain
growth using appropriate tools can elucidate the mechanisms of the crystallization of a-
Si and will allow the control of the quality of the resulting polysilicon silicon (poly-Si)
film formed by ALILE process.
The kinetics of crystallization by aluminium-induced crystallization (AIC) process was
already addressed by several researchers [Manj78, Nas00, Gal02, Wid02, Kis03, Hsu03,
Sch05, Pih07]. The driving force of the AIC process is described in literatures by
[Nas00, Wid02, Rob04, Yu97, Sch06]. From some literature [Nas00b, Kon95, Sch06]
the fact is agreed that the concentration of Si dissolved in the Al in the Al/a-Si system is
different from that in the Al/c-Si system. Nast et al. [Nas00] and Konno et al. [Kon95]
suggested that this concentration gradient is generated by the difference in Gibbs free
energy between the initial and the final state. Pihan et al. [Pih07] argued that this
difference in Gibbs free energy of the initial and the final state in the case of a
crystalline Si layer as the precursor is associated to the grain boundary density and
energy changes during crystallization.
POLY-Si FILMS ON ZnO:Al COATED GLASS 56
A wide range of activation energies is found for the layer exchange process varying
from 1.2 to 2 eV in the literature [Man78, Kim02, Pih07, Gal02]. The difference in
activation energies from one author to another may originate from the complexity in the
kinetics of the process since dissolution of Si, diffusion through the oxide at the
interface Al/a-Si, and competition between grains influence the Si oversaturation in the
Al layer and therefore modify the activation energy. The activation energy deduced
from the diffusion coefficient of Si in Al layers and for Si diffusion in a thin Al layer is
in the range of 0.8~1.36 eV [Hul98, McC71]. The activation energy of self-diffusion of
Si and of the epitaxial regrowth of implanted Si is 4.1~4.6 eV) [Hir79, Kal79] and 2.4
eV [Cse75, Cse78], respectively. The activation energy of grain-boundary diffusion of
Al in polycrystalline Si is 2.64 eV [Hwa80]. The activation energy for a-Si is 3~4 eV in
the solid phase crystallization (SPC) process [Spi98, Lee97].
In the presented study, the crystallization (nucleation and grain growth) was observed in
a heating stage of an optical microscope described in section 3.2.1.1. Therefore the
samples were placed upside down on the heating stage such that the initial ZnO:Al/Al
interface could be observed through the glass/ZnO:Al during annealing. The formation
of poly-Si grains at the initial ZnO:Al/Al interface leads to a local change of the
reflectivity as Si reflects less light than Al. The corresponding optical micrographs were
analyzed to determine the crystallized fraction. The crystallized fraction as a function of
the annealing time for different annealing temperatures (T
A
) is shown in Fig 5.1. The
annealing temperature was varied from 375 to 550 °C with an increment of 25 °C. At
higher temperatures, the crystallized fraction increases very rapidly and then saturates
while at low temperature a long incubation time is necessary before the crystallization
starts and then the crystallized fraction increases moderately and saturates after a long
time. The time necessary to get complete poly-Si films is found to depend strongly on
the annealing temperature. The crystallized fraction reached nearly 100% at all
annealing temperatures, which means that at all annealing temperatures continuous
poly-Si film were formed on top of the ZnO:Al coated glass. The crystallization process
on ZnO:Al coated glass at 425 °C takes around 4 hours. The crystallization process at
annealing temperatures above 500 °C is very fast (below 20 min). Pihan et al. [Pih07]
mentioned that the dentritic growth behaviour of Si grains is typical for metal-induced
POLY-Si FILMS ON ZnO:Al COATED GLASS 57
crystallization of a-Si below the eutectic point, especially during the Al-induced layer
exchange process and the shape of the Si grains is strongly depending on the
crystallization temperature. In spite of the dentritic growth behaviour of Si grains
formed on thermal oxidized Si wafer at 450 °C in his studies, the growth behaviour of
Si grains in our experiment have circular shape at higher temperatures (> 425 °C).
Dentritic growth behaviour was obtained on some Si grains formed at 400 °C and
obviously on most of Si grains formed at 375 °C.
10
0
10
1
10
2
10
3
0
20
40
60
80
100
Crystallized fraction [%]
Annealing time [min.]
T
A
375
o
C
550
o
C
Figure 5. 1: Crystallized fraction versus annealing time for poly-Si films formed on
ZnO:Al coated glass at eight different annealing temperature T
A
(375…550 °C). The
arrow indicates increasing annealing temperature.
POLY-Si FILMS ON ZnO:Al COATED GLASS 58
0 50 100 150 200 250 300
0
20
40
60
80
100
T
A
= 425
o
C
on bare glass
on ZnO:Al coated glass
Crystallized fraction [%]
Annealing time [min.]
Figure 5. 2: Comparison of the crystallized fraction of poly-Si films on bare glass and
on ZnO:Al coated glass at 425 °C.
The crystallized fraction of poly-Si films on bare glass and on ZnO:Al coated glass is
compared in Fig 5.2. It was observed that the process time for the ALILE process on
ZnO:Al coated glass is slightly shorter compared to the process time of the ALILE
process on bare glass by the investigation of in-situ annealing. The reason could be that
the surface of the ZnO:Al coated glass is rougher than the surface of the bare glass itself.
The root-mean-squared (RMS) roughness of the ZnO:Al film and the Borofloat33 glass
measured by atomic force microscopy (AFM) were 1.37 and 0.75 nm, respectively.
Fig 5.3 shows the increase of the crystallized fraction and the variation of the grain
density on ZnO:Al coated glass as a function of an annealing time at the annealing
temperature of 425 °C. Nucleation and grain growth does not occur in region (I). Both
nucleation and grain growth occur in region (II). In region (III) only grain growth
occurs (no nucleation). Here the grains grow until the film is fully crystallized. During
the isothermal transformation, the cystallized fraction (α) of Si is represented by
Avrami's equation [Joh39, Avr39, Avr40].
POLY-Si FILMS ON ZnO:Al COATED GLASS 59
)Ktexp(1)t(
n
−−=α
(5.1)
Here t is the process time, K is rate constant and n is the order parameter which depends
upon the mechanism of crystal growth. n is held to have an value between 1-4 which
reflected the nature of the transformation in question [Jen92]. The lines in Fig 5.2 and
Fig 5.3 are based on Avrami's equation shown in Eq (5.1). For the fitting of the
crystallized fration on ZnO:Al coated glass in Fig 5.2 the fitting parameter K and n was
5.21×10
-14
and 3.76, respectively. For the fitting of the crystallized fration on bare glass
in Fig 5.2 the fitting parameter K and n was 8.45×10
-11
and 2.68, respectively.
0
20
40
60
80
100
0 20 40 60 80 100 120 140 160 180
0.0
0.5
1.0
1.5
2.0
2.5
Crystallized fraction [%]
T
A
= 425
o
C
on ZnO:Al coated glass
(III)(II)
Annealing time [min.]
Grain density [/cm
2
]
(I)
Figure 5. 3: Crystallized fraction and the grain density as a function of the annealing
time at 425 °C as the annealing temperature; region (I): no nucleation and no grain
growth, region (II): nucleation and grain growth, region (III): no further nucleation and
grain growth but no nucleation until full crystallization.
POLY-Si FILMS ON ZnO:Al COATED GLASS 60
5.1.1 NUCLEATION
Fig 5.4 shows the nucleation time of silicon on glass (solid squares) and on ZnO:Al
coated glass (open circles) as a function of the annealing temperature. The nucleation
time corresponds to the time when the first grain exceeding the optical microscope
resolution (~ 0.5 µm) appears in the optical microscope. Open circles and open squares
are the experimental data and straight lines are the Arrhenius fitting curves. The
equation used for calculation of the activation energy E
A
is shown below.
−=
A
kT
A
E
expAt
(5.2)
where A is the pre-exponential factor or simply the prefactor, t is the nucleation time,
E
A
is the activation energy for the nucleation, k is the Boltzmann constant and T
A
is the
annealing temperature (K). In Eq. (5.2), E
c
and A are calculated to be practically
independent of the temperature (at least in the temperature interval accessible in the
calorimetric measurements).
The slope of the curve is hence related to the magnitude of the activation energy E
A
that
reflects the rate-limiting mechanism during the entire nucleation process. The activation
energy for nucleation on glass and on ZnO:Al coated glass below 500 °C is 1.4 and 1.2
eV, respectively. The activation energy for the nucleation on ZnO:Al coated glass is
lower than that for the nucleation on glass. This means that ZnO:Al films have an
influence on the Si nucleation. Generally, the surface roughness of a substrate
influences the nucleation time. The surface roughness of ZnO:Al films is higher than
that of glass. It might be the reason why the crystallization time on ZnO:Al coated glass
is shorter than that on glass substrates. The activation energy on ZnO:Al coated glass
was calculated to be 1.2 eV for T < 500 °C and 0.2 eV for T > 500 °C. And also the
activation energy of poly-Si films on bare glass shows the same behaviour. This
behaviour is different to the behaviour reported in the literature [Sch04]. According to
Schneider et al., the activation energy is 1.9 eV at the annealing temperature range of
400 ~ 540 °C. The activation energy for the nucleation on bare glass is lower than 1.9
POLY-Si FILMS ON ZnO:Al COATED GLASS 61
eV reported by Gall et al. [Gal02b] and Schneider et al. [Sch04]. It can be explained
with the different glass substrate used. Schneider et al. used Corning1737 glasses as a
substrate. The reason could be the change of glass substrate from Corning 1737 glass
(from Corning) to Borofloat33 glass (from Schott). Corning1737 glass and Borofloate33
glass have different thermal expansion coefficients and different strain points as seen in
section 3.1.1. Hsu et al. [Hsu03] argued that the thermal energy at higher temperatures
can release the film stress in a short time.
1.2 1.3 1.4 1.5
10
0
10
1
10
2
E
A, on glass, T
A
> 500
o
C
= 0.3
±
0.1 eV
E
A, on ZnO:Al, T
A
> 500
o
C
= 0.2
±
0.1 eV
E
A, on ZnO:Al, T
A
< 500
o
C
= 1.2
±
0.2 eV
Glass/poly-Si
Glass/ZnO:Al/poly-Si
Nucleation time [min.]
1/T [
x
10
-3
K
-1
]
E
A, on glass, T
A
< 500
o
C
= 1.4
±
0.1 eV
Figure 5. 4: Arrhenius-plot of the temperature dependence of the nucleation time of
poly-Si on glass (solid squares) and on ZnO:Al coated glass (open circles). The
nucleation time corresponds to the time when the first grain appears in the optical
microscope (~ 0.5 µm).
5.1.2 GRAIN GROWTH
As can be seen in Fig 5.5, the activation energy for the grain growth determined in
region III (see Fig 5.3) for the crystallization of 99 % on ZnO:Al coated glass and on
POLY-Si FILMS ON ZnO:Al COATED GLASS 62
bare glass is 1.9 and 1.8 eV, respectively. The activation energy for the grain growth on
ZnO:Al coated glass and on bare glass is quite similar. The value of 1.8 ~ 1.9 eV
compares well with the activation energy of 1.8 eV calculated from the grain growth
velocity reported by Schneider [Sch04]. This means that the ZnO:Al film and the kind
of glass substrate should not influence the activation energy for the grain growth. But
the ZnO:Al film can reduce the total crystallization time because the pre-factor A is
lower. The pre-factor for the grain growth on ZnO:Al coated glass and on bare glass is
4×10
-12
and 2×10
-11
min, respectively. The pre-factor for the grain growth on ZnO:Al
coated glass is one order magnitute lower than that on bare glass. In the perspective of
an industrial application in the future the enhanced crystallization speed on ZnO:Al is
desired to reduce a process time.
1.2 1.3 1.4 1.5 1.6
10
-1
10
0
10
1
10
2
10
3
10
4
E
A, on glass
= 1.8
±
0.2 eV
E
A, on ZnO:Al
= 1.9
±
0.2 eV
Glass/poly-Si
Glass/ZnO:Al/poly-Si
Grain growth time [min.]
1/T [
x
10
-3
K
-1
]
Prefactor
A
on glass
= 1E-11
A
on ZnO:Al
=2E-12
Figure 5. 5: Arrhenius-plot of the temperature dependence of the grain growth time on
ZnO:Al coated glass (open circles) and on bare glass (solid squares). The error range is
smaller than indicators.
POLY-Si FILMS ON ZnO:Al COATED GLASS 63
5.2 STRUCTURAL PROPERTIES
The structural properties of poly-Si films formed on ZnO:Al coated glass and on glass
were intensively studied and are compared in this section. In the ALILE process the
process temperature is a crucial parameter which can change grain size and preferential
(100) orientation. O. Nast [Nas00b] concluded that the process temperature is one of the
most important parameters which influence the grain size.
5.2.1 CRYSTALLINE QUALITY
SEM tilted image and surface image of a glass/ZnO:Al/poly-Si stack after chemical
mechanical polishing (CMP) are shown in Fig 5.6(a) and (b), respectively. The
annealing is applied at 450 °C for 16 hours. As shown in Fig 5.6 a continuous poly-Si
film is formed on ZnO:Al coated glass but few holes are visible at the grain boundaries.
These holes are a typical feature of poly-Si films formed by ALILE process. O. Nast et
al. [Nas00S] concluded that these holes in the poly-Si film are caused by intergrain Al
clusters. The straight lines from top left the bottom right shown in Fig 5.6(a) were
caused by the CMP process.
The crystalline nature of poly-Si thin films on ZnO:Al coated glass and on glass after
ALILE process was verified using Raman spectroscopy. Raman measurements can also
give a first impression of the crystallographic quality of the poly-Si material. Fig 5.7
shows the normalized Raman spectra of poly-Si thin films (a) on glass and (b) on
ZnO:Al coated glass for different annealing temperatures. The Raman peaks
corresponding to the transverse optical (TO) phonon of poly-Si thin films on glass and
on ZnO:Al coated glass annealed within the whole process temperature range from
425 °C to 525 °C are at 520 cm
-1
and exhibit a similar line shape for all the samples. In
all cases, there is no Raman signal at 480 cm
-1
that would indicate an amorphous phase
in the poly-Si thin films [Sch98].
POLY-Si FILMS ON ZnO:Al COATED GLASS 64
ZnO:Al
Glass
Poly-Si
(a)
holes
(b)
Figure 5. 6: (a) SEM tilted image and (b) surface image of glass/ZnO:Al/poly-Si stack
after CMP.
POLY-Si FILMS ON ZnO:Al COATED GLASS 65
500 510 520 530
0
1
Intensity [a.u]
Raman shift
[
[[
[
cm
-1
]
]]
]
Glass/poly-Si
Annealing temperature
425
o
C
450
o
C
475
o
C
500
o
C
525
o
C
(a)
500 510 520 530
0
1
Intensity [a.u]
Raman shift
[
[[
[
cm
-1
]
]]
]
Glass/ZnO:Al/poly-Si
Annealing temperature
425
o
C
450
o
C
475
o
C
500
o
C
525
o
C
(b)
Figure 5. 7: Normalized Raman spectra of poly-Si thin films (a) on glass and (b) on
ZnO:Al coated glass for different annealing temperatures.
POLY-Si FILMS ON ZnO:Al COATED GLASS 66
To determine the Full-Width at Half Maximum (FWHM) of the TO phonon line a
Lorentzian peak has been used as fit function [Kam98, Vou95]. The spectral width is
interpreted in terms of phonon confinement, i.e. it is a measure for the density of defects
in the films. The FWHM of poly-Si thin films on ZnO:Al coated glass and on glass as a
function of the annealing temperature are shown in Fig 5.8. For comparison the FWHM
of FZ-Si wafer is indicated with a dashed line. The FWHM value of both poly-Si films
on glass and on ZnO:Al coated glass shows no significant change with increasing
annealing temperature. The FWHM values of 4.1 cm
-1
and 3.8 cm
-1
have been
calculated for the poly-Si films formed on ZnO:Al coated glass and on glass at 425 °C
as a ALILE process temperature, respectively. The value of the FWHM of poly-Si thin
films formed on glass and on ZnO:Al coated glass at 525
°C is 5 and 4.8 cm
-1
,
respectively. The FWHM of FZ Si wafer is 3 cm
-1
, which is still much lower than the
obtained values for poly-Si films. Hence, no significant difference has been observed
for both types of poly-Si films. From the Raman results it could be concluded that poly-
Si films formed on glass and on ZnO:Al coated glass have the same crystalline quality.
425 450 475 500 525
3
4
5
6 Glass/poly-Si
Glass/ZnO:Al/poly-Si
FWHM
[
[
[
[
cm
-1
]
]
]
]
Annealing Temperature
[
[[
[
o
C
]
]]
]
FZ-Si wafer
Figure 5. 8: FWHM of poly-Si films on glass and on ZnO:Al coated glass as a function
of the annealing temperature. For comparison the FWHM of FZ-Si wafer is indicated
with a dashed line.
POLY-Si FILMS ON ZnO:Al COATED GLASS 67
500 510 520 530
0
1
Intensity [a.u]
Raman shift
[
[[
[
cm
-1
]
]]
]
Glass/ZnO:Al/poly-Si
Glass/poly-Si
FZ-Si wafer
Figure 5. 9: Normalized Raman spectra of poly-Si films grown on ZnO:Al coated glass
(red solid line) and on glass (blue dashed line). The samples were annealed at 425 °C
for 16 hours. The normalized Raman spectrum of a FZ silicon wafer (black dotted line)
is shown as a reference. Each spectrum is normalized to its maximum.
The normalized Raman spectra of poly-Si thin films grown on glass (blue dashed line)
and on ZnO:Al coated glass (red solid line) annealed at 425 °C for 16 hours are
compared in Fig 5.9. The normalized Raman spectrum of a FZ silicon wafer is shown as
a reference. Each spectrum is normalized to its maximum. The Raman spectra of poly-
Si films on glass and on ZnO:Al coated glass have no significant difference.
UV reflectance is an excellent characterization tool for the evaluation of the near-
surface crystalline quality of poly-Si thin-film materials [Wid05, Str04, Wid07]. Fig
5.10 shows the measured UV reflectance of poly-Si films on glass and ZnO:Al coated
glass, together with that of a polished c-Si wafer for comparison. The UV reflectance of
poly-Si films on glass and on ZnO:Al coated glass are quite similar to the UV
reflectance of the Si wafer. Using the ‘‘crystalline quality figure of merit’’ (Q) proposed
in Ref. [Str04, Wid05], the poly-Si films formed on glass and on ZnO:Al coated glass at
425 °C have a crystalline quality in the surface region of about 99 %, compared to a
POLY-Si FILMS ON ZnO:Al COATED GLASS 68
figure of merit of 100% of a polished high-quality Si wafer. The crystalline quality
factor Q is obtained by dividing the measured reflectances of the sample at two specific
wavelengths, ~ 367 nm (e1 peak) and ~ 275 nm (e2 peak), by the reflectances of a
polished high-quality Si wafer at these wavelengths and then calculating the mean of the
two values. The equation is the following:
)
R
R
R
R
(
2
1
100Q
Sic,2e
2e
Sic,1e
1e
−−
+×=
[%] (5.3)
where R
e1
and R
e2
is the measured reflectance of the sample at 367 nm and 275 nm,
respectively. R
e1, c-Si
and R
e2, c-Si
is the measured reflectance of the polished high-quality
Si wafer as a reference at 367 nm and 275 nm.
260 280 300 320 340 360 380
50
55
60
65
70
75
80
e1
Reflectance [%]
Wavelength [nm]
Si wafer
Glass/poly-Si
Glass/ZnO:Al/poly-Si
e2
Figure 5. 10: Measured UV reflectance of a polished Si wafer (black line) and poly-Si
films formed on glass (red circles) and on ZnO:Al coated glass (blue circles) at 425 °C
for 16 hours.
POLY-Si FILMS ON ZnO:Al COATED GLASS 69
425 450 475 500 525
95
96
97
98
99
100
Quality factor [%]
Annealing temperature
[
[[
[
o
C
]
]]
]
Glass/poly-Si
Glass/ZnO:Al/poly-Si
Figure 5. 11: The quality factor Q of poly-Si films on ZnO:Al coated glass (red
squares) and on glass (blue circles) as a function of annealing temperature. The quality
factors were calculated by the equation (5.2). The marked area indicates data from
Widenborg et al. [Wid05].
The crystalline quality of poly-Si films on glass and on ZnO:Al coated glass calculated
as a function of annealing temperature by Eq (5.3) is shown in Fig 5.11. All poly-Si
films formed on glass and on ZnO:Al coated glass have a similar crystalline quality
(98.5 ~ 99.4%). Hence Raman results and UV reflectance are in agreement with respect
to the poly-Si film quality. The quality factor Q of poly-Si films formed using the
ALILE process from Widenborg et al. [Wid05] is 97.4 ~ 98% as shown in Fig 5.11. The
crystalline quality of poly-Si films on ZnO:Al coated glass is slightly better than the
data of Widenborg et al. for both on glass and on ZnO:Al coated glass.
Some researcher investigated the influence of post treatments (H-passivation, rapid
thermal annealing (RTA)) in terms of open-circuit voltage and short-current density of
solar cells [Ter07, Gor07, Rau06]. But the characterization of the ALILE poly-Si films
itself after post treatments were not studied so far. Raman was used to investigate the
influence of post-treatments on the crystalline quality of ZnO:Al/poly-Si stacks.
POLY-Si FILMS ON ZnO:Al COATED GLASS 70
Fig 5.12 shows the Raman spectra of glass/ZnO:Al/poly-Si stacks after H-passivation
(red), RTA (green), and RTA+H-passivation (blue). The black line is the Raman
spectrum of glass/ZnO:Al/poly-Si stack after the ALILE process and before the
application of post-treatments. H-passivation treatment does not have any influence on
the position of the Si-Si TO (transverse optical)-LO (longitudinal optical) phonon band.
But the Si-Si TO-LO phonon band position was shifted from 520.3 to 521.3 cm
-1
by the
RTA treatment. It is known that stress causes the shift of the optical-phonon line at 520
cm
-1
[Cer72]. The stress can be estimated from the frequency shift in the case of
FWHM < 8 cm
-1
[Kit02]. The dominant origin of stress in poly-Si thin films was
attributed to thermal stress. The frequency is also modified by the presence of micro-
crystals. It can be concluded that the RTA treatment introduces compressive stress in
ZnO:Al/poly-Si stacks. The FWHM of poly-Si films treated by H-passivation does not
change.
500 510 520 530 540
0
1
518 520 522 524
Glass/ZnO:Al/poly-Si
as-grown
H-passivation
RTA
RTA+H-passivation
Intensity [a.u]
Raman shift
[
[[
[
cm
-1
]
]]
]
T
ALILE
= 425
o
C
t
ALILE
= 16 hours
T
RTA
= 850
o
C
t
RTA
= 200 seconds
Figure 5. 12: Raman spectra of glass/ZnO:Al/poly-Si stacks treated by H-passivation
(red), RTA (green), and RTA+H-passivation (blue). The black line is the Raman
spectrum of glass/ ZnO:Al/poly-Si stack after ALILE process and before post-
treatments.
POLY-Si FILMS ON ZnO:Al COATED GLASS 71
5.2.2 PREFERENTIAL ORIENTATION
To investigate the crystallographic orientation of poly-Si thin films formed on glass and
on ZnO:Al coated glass, electron backscatter diffraction (EBSD) and X-ray diffraction
measurements were performed.
Fig 5.13 shows the EBSD orientation maps of poly-Si films formed on glass and on
ZnO:Al coated glass by the ALILE process at different annealing temperatures. The
color scale for the orientation maps is: Red color is (100) orientation, blue color is (111)
orientation, and green color is (101) orientation. From the orientation maps shown in
Fig 5.13, two important parameters concerning the poly-Si thin films can be extracted:
The grain size and the orientation of the grains. The orientation maps are mainly red
what is assigned to (100) orientation. Fig 5.14 shows the preferential (100) orientation
R
(100)
of poly-Si films on ZnO:Al coated glass (solid squares) and on glass (open
circles). The fraction of the poly-Si surface showing an orientation within 20° of the
perfect (100) direction is defined as the preferential (100) orientation R
(100)
. The
preferential (100) orientation of poly-Si films is about 60% on both substrates. No
significant change of a preferential (100) orientation with increasing annealing
temperature was observed. A (100) orientation of the seed layer is very advantageous
for subsequent epitaxial thickening. In previous publications [Kim02, Gal06] it was
shown that the preferential (100) orientation decreases with increasing annealing
temperature. In this study Borofloat 33 glass (from Schott) was used instead of Corning
1737 glass that was used in other studies. The thermal expansion coefficient of
Borofloat 33 glass, crystalline silicon, and Corning 1737 glass is about 3.25×10
-6
K
-1
,
2.6×10
-6
K
-1
, and 3.76×10
-6
K
-1
, respectively. The thermal expansion coefficient of
Borofloat33 glass is close to the thermal expansion coefficient of crystalline silicon. The
thermal stress applied on the glass substrate during the ALILE process with different
temperatures is changed due to the change of thermal expansion coefficient of the glass
substrate. The change of the thermal stress at different temperatures could lead to the
change of an orientation of the crystallized silicon thin film. The preferential (100)
orientation of the poly-Si films on ZnO:Al coated glass can make these layers act as
favorable seed layers for epitaxial growth of silicon absorber layers.
POLY-Si FILMS ON ZnO:Al COATED GLASS 72
Glass/poly-Si Glass/ZnO:Al /poly-Si
425 °C
450 °C
475 °C
See the next page for the caption
POLY-Si FILMS ON ZnO:Al COATED GLASS 73
Glass/poly-Si Glass/ZnO:Al /poly-Si
500 °C
525 °C
Figure 5. 13: EBSD orientation maps for poly-Si thin films on glass and on ZnO:Al
coated glass annealed at 425 °C, 450 °C, 475 °C, 500 °C and 525 °C. Color scale for the
orientation maps: Red color is (100) orientation, blue color is (111) orientation, and
green color is (101) orientation.
POLY-Si FILMS ON ZnO:Al COATED GLASS 74
425 450 475 500 525
0
20
40
60
80
100
R
(100)
[%]
Annealing Temperature
[
[[
[
o
C
]
]]
]
Glass/poly-Si
Glass/ZnO:Al/poly-Si
Figure 5. 14: Preferential (100) orientation R
(100)
of poly-Si films prepared on ZnO:Al
coated glass (solid squares) and on glass (open circles) as a function of annealing
temperature.
Even though the preferential (100) orientation of poly-Si films on glass and ZnO:Al
coated glass is similar as shown in Fig 5.14, the different behavior of grain orientations
of poly-Si film formed on glass and on ZnO:Al coated glass is shown in Fig 5.15. As
can be seen in Fig 5.15(a) many grains of poly-Si film on glass are oriented to (111)
orientation even though few grains of poly-Si film on ZnO:Al coated glass have a (111)
orientation. Also the tilted angles of grains from a perfect (100) orientation show a
different behavior between the glass/poly-Si stack and the glass/ZnO:Al/poly-Si stack.
In case of glass/poly-Si stack most of grains are located in a range of 10 ~ 20° tilted
from a perfect (100) orientation, but in case of glass/ZnO:Al/poly-Si stack most of
grains are located in a range of 0 ~ 10° tilted from a perfect (100) orientation.
POLY-Si FILMS ON ZnO:Al COATED GLASS 75
100 110
111
(a)
100 110
111
(b)
Figure 5. 15: Inverse pole figure of an electron back scatter diffraction (EBSD)
orientation map of a poly-Si film (a) on glass and (b) on ZnO:Al coated glass prepared
at 425 °C for 16 hours as annealing temperature of the ALILE process. In both cases
about 60% of the area under investigation is tilted by less than 20° with respect to the
(100) orientation (the corresponding region is indicated by a dashed line).
POLY-Si FILMS ON ZnO:Al COATED GLASS 76
Both EBSD and X-ray diffraction (XRD) method are powerful techniques for
determining orientation of poly-Si films, but a very flat sample surface is required in
case of EBSD measurements. Due to this fact, XRD was measured to compare with the
orientation of poly-Si films extracted from EBSD measurements. The XRD spectra of
poly-Si films on ZnO:Al coated glass and on glass are shown in Fig. 5.16. Due to the
high texture of the underlying ZnO:Al film only the (002) and the (004) reflections of
zinc oxide are seen in the spectrum. Apart from these peaks, the positions of all silicon
reflexes expected in the examined diffraction angle range are indicated. For a real
investigation of film texture, i.e. preferential crystal growth in certain crystallographic
directions, the measured peak areas first have to be corrected for the X-ray absorption
for different incident angles and the structure factors of the individual peaks. The latter
can be done by comparing the relative peak intensities to the ones of powder diffraction
standards [ICDD]. In our case it was found that only orientations along (111), (100) and
mixtures thereof play a significant role, while (110) orientation could not be found.
Based on this finding a tentative calculation of the fraction of grains oriented along the
four crystal directions indicated in Fig 5.16 were carried out. The result is shown in Fig
5.17. The open circles indicate the fraction of each orientation of the
glass/ZnO:Al/poly-Si stack and the open squares indicate the fraction of each
orientation of the glass/poly-Si stack. The fraction of (400) orientation of poly-Si films
on glass and on ZnO:Al coated glass was determined to 31% and 48%, respectively.
And the fraction of (511) orientation of poly-Si films on glass and on ZnO:Al coated
glass was determined to 27% and 23%, respectively. These results suggest that the
preferential (400) orientation of poly-Si film on ZnO:Al coated glass is stronger than the
preferential (400) orientation of poly-Si film on glass. For the calculation of a
preferential (400) orientation both the fraction of (400) and (511) orientation were
considered. As can be seen in Table 5.1, (411) and (511) orientations are tilted within
20° of the perfect (400) direction. Nevertheless, (411) orientation was not detected in
the poly-Si films. The preferential (400) orientation R
(400)
(i.e. the sum of (400) and
(511) peak) of poly-Si films on glass and on ZnO:Al coated glass is 58% and 71%,
respectively. The numbers are similar to those determined using EBSD on samples
prepared in the same way, but the difference between glass and ZnO:Al was not
POLY-Si FILMS ON ZnO:Al COATED GLASS 77
observed previously. Grigorov et al. [Gri07] showed XRD results of poly-Si film on
ZnO:Al coated glass. According to their investigation poly-Si films on ZnO:Al coated
glass have a preferential (111) orientation. This result is different from our results where
a preferential (100) orientation is observed. There are two differences: (a) the different
orientation of ZnO:Al films and (b) a different preferential orientation of poly-Si films
on glass. ZnO:Al films have a (101) orientation for their experiment and a (100)
orientation in our experiments. Poly-Si films on glass have a preferential (111)
orientation in their experiment [Dim06] and a preferential (100) orientation in this
experiment as shown in Fig 5. 13.
30 40 50 60 70 80 90 100
ZnO <004>
Glass/ZnO:Al/poly-Si
2
θ
θθ
θ
[deg.]
Si <111>
ZnO <002>
Si <311>
Si <400>
Si <511>
Glass/poly-Si
Count [cps, log scale]
Figure 5. 16: XRD spectra of poly-Si films on glass (upper) and on ZnO:Al coated
glass (below).
POLY-Si FILMS ON ZnO:Al COATED GLASS 78
<111> <311> <400> <511> <400>+<511>
0
20
40
60
80
100
Glass/poly-Si
Glass/ZnO:Al/poly-Si
Fraction of orientation [%]
Orientation <hkl>
Figure 5. 17: The percentages of grain orientation of poly-Si films formed on glass and
on ZnO:Al coated glass at 425 °C for 16 hours calculated from XRD measurement.
TABLE 5. 1: Tilted Angles between (400) orientation and different orientations.
Angle
(111) (211) (311) (411) (511)
(400) 54.7 35.2 25.2 19.47 15.8
POLY-Si FILMS ON ZnO:Al COATED GLASS 79
5.2.3 GRAIN SIZE
To investigate the grain size of poly-Si films on glass and on ZnO:Al coated glass
prepared at different annealing temperatures EBSD measurements and in-situ optical
microscope measurements were used.
Fig 5.18 shows (a) the maximum grain size and (b) the average grain size of the poly-Si
thin films on glass (open circles) and on ZnO:Al coated glass (solid squares) as a
function of the annealing temperature. The maximum and average grain size of the
poly-Si thin films was extracted from orientation maps measured by EBSD data (see Fig
5.13). The maximum grain size and the average grain size of poly-Si films on ZnO:Al
coated glass are about 16 µm and 5 µm at 425 °C, respectively. The maximum grain
size (solid squares) and the average grain size (solid circles) of the poly-Si thin films on
glass are 18 µm and 7 µm at the same annealing temperature, respectively. The grain
size decreases with increasing annealing temperature on ZnO:Al coated glass as well as
on glass. For all temperatures, the grain size of the poly-Si films on ZnO:Al coated glass
is slightly smaller than that of the poly-Si films on glass. This might be due to the
surface roughness of ZnO:Al film. The surface of ZnO:Al layer is rougher than the
surface of glass substrates. Due to this reason, the ALILE process time on ZnO:Al
coated glass is shorter than that on glass and more nucleation sites of silicon and hence
smaller grains lead to the faster crystallization process. As mentioned before, the time
between the time needed for first nucleation and the time needed for the last grain
nucleation has an impact to determine average grain size. The small time gap leads to
similar grain sizes of the sample. Even though the maximum grain size of poly-Si films
on glass is bigger than the maximum grain size of poly-Si films on ZnO:Al coated glass,
the average grain size of poly-Si films on glass is slightly bigger than the average grain
size of poly-Si films on ZnO:Al coated glass.
POLY-Si FILMS ON ZnO:Al COATED GLASS 80
425 450 475 500 525
5
10
15
20 Maximum grain size
Glass/poly-Si
Glass/ZnO:Al/poly-Si
Grain size
[
[
[
[
µ
µ
µ
µ
m
]
]
]
]
Annealing temperature
[
[[
[
o
C
]
]]
]
(a)
425 450 475 500 525
0
2
4
6
8
10 Average grain size
Glass/poly-Si
Glass/ZnO:Al/poly-Si
Grain size
[
[
[
[
µ
µ
µ
µ
m
]
]
]
]
Annealing temperature
[
[[
[
o
C
]
]]
]
(b)
Figure 5. 18: (a) maximum grain size and (b) average grain size extracted from EBSD
data of poly-Si films formed on glass (open circles) and on ZnO:Al coated glass (solid
squares) as a function of the annealing temperature.
POLY-Si FILMS ON ZnO:Al COATED GLASS 81
The grain size distribution as a function of the annealing temperature for the poly-Si
films formed on glass and on ZnO:Al coated glass are shown in Fig 5.19. The data were
obtained from electron backscatter diffraction (EBSD) measurements. The lines are
Gaussian fitting curves. When the poly-Si films are formed at high temperatures (475 ~
525 °C) by the ALILE process, the distribution of grain size is quite compact with a
maximum at a grain size of about 3 µm. On the other hand, a very large distribution of
the poly-Si grain diameter is obtained for low annealing temperatures (425 ~ 450 °C)
and no real maximum of the grain size can be found. Although in this regime most of
the surface is determined by the large grains it is expected that the quality of the
resulting poly-Si layer will be determined by the low-quality (small grains) regions.
Fig 5.20 shows the grain density on ZnO:Al coated glass versus annealing time for
different annealing temperatures. Si grains larger than 0.5 µm were considered due to
the resolution of the optical microscope. At 425 and 450 °C, a different behavior is
observed: the grain density is found to increase slightly and both nucleation and grain
growth are taking place at the same time. Above 475 °C, the nucleation stops at an early
stage as shown in Fig 5. 20. The formed Si grains continue to grow until they coalesce
without significant additional nucleation and it ends with the formation of a continuous
poly-Si layer. An interpretation of this behavior is that suppression of nucleation can
take place over a large distance at high temperatures. Information on the grain size
distribution will be greatly helpful to validate such assumption.
In Fig 5.21 the average grain size of poly-Si films extracted by EBSD (squares) is
compared with the estimated average grain size of poly-Si films calculated by in-situ
optical microscopy (circles) measurements. The estimated average grain size of poly-Si
films observed by in-situ optical microscopy (OM) is calculated from the maximum
nucleation density N
G
shown in Fig 5.20. The maximum of the nucleation density N
G
was used to determine the final estimated grain size d
G
of grains (
GG
N/Ad =
). A is the
image size of the optical micrographs. The average grain size of poly-Si films
calculated by in-situ optical microscopy is slightly larger than the average grain size of
poly-Si films calculated by EBSD. A simple technique like optical microscopy can give
good information locally.
POLY-Si FILMS ON ZnO:Al COATED GLASS 82
0
10
20
0
10
20
30
0
10
20
30
40
2 4 6 8 10 12 14 16 18
0
10
20
30
40
50
0
10
525
o
C
500
o
C
475
o
C
450
o
C
Number of grain
425
o
C
Glass/poly-Si
Grain size [µm]
(a)
0
10
20
0
10
20
30
0
10
20
30
40
2 4 6 8 10 12 14 16 18
0
10
20
30
40
50
0
10 Glass/ZnO:Al/poly-Si
525
o
C
500
o
C
475
o
C
450
o
C
Number of grains
425
o
C
Grain size [
µ
m]
(b)
Figure 5. 19: Grain size distribution as a function of the annealing temperature for the
poly-Si films formed (a) on glass and (b) on ZnO:Al coated glass. The data were
extracted from electron backscatter diffraction (EBSD) data.
POLY-Si FILMS ON ZnO:Al COATED GLASS 83
0 10 20 30 40 50
0
2
4
6
8
10
12 Annealing temperature
525
o
C
500
o
C
475
o
C
450
o
C
425
o
C
Grain density
[
[
[
[
x
10
6
/cm
2
]
]
]
]
Annealing time [min.]
Figure 5. 20: Grain density vs. annealing time for different annealing temperatures. The
substrate is ZnO:Al coated glass. Si crystals larger than 0.5 µm were considered due to
the resolution of optical microscope. The size of images from in-situ optical microscope
is 350×260 µm
2
.
425 450 475 500 525
0
2
4
6
8
10 Glass/ZnO:Al/poly-Si
from EBSD
from in-situ optical microscope
Grain size
[
[
[
[
µ
µ
µ
µ
m
]
]
]
]
Annealing temperature
[
[[
[
o
C
]
]]
]
Figure 5. 21: The average grain size of poly-Si films on ZnO:Al coated glass extracted
by EBSD (squares) and calculated by in-situ optical microscopy (circles) measurements.
POLY-Si FILMS ON ZnO:Al COATED GLASS 84
5.2.4
DEFECT ANALYSIS
As can be seen in Fig 5.22 (a), there are numerous twin grain boundaries within the
poly-Si film formed on ZnO:Al coated glass at 425 °C. These twin grain boundaries can
be identified using EBSD analysis. Coincident site lattice (CSL) boundaries are shown
by color. CSL boundaries are mainly twin boundaries of the first order (Σ3), second
order (Σ9) and third order (Σ27). It is a highly symmetrical interface; also, atoms are
shared by the two crystals at regular intervals. This is also a much lower-energy
interface than the grain boundaries that form when crystals of arbitrary orientation grow
together. Higher order twins are formed by the subsequent influence of twinning or by
the reaction of lower order twins [Pih07]. For example, the Σ27 boundary is formed by
Σ3 and Σ9 boundaries. CSL boundaries generally present low-energy configurations
compared to random grain boundaries [Pih07, Bro81].
Knowledge of the density of these twins is very important in order to understand the
formation of these twins and to improve the crystalline quality of the poly-Si layer. As
shown in Fig. 5.22 (b), twin boundaries of Σ3 and Σ9 are present for poly-Si film
formed on ZnO:Al coated glass at 425 °C. A very high distribution of Σ3 boundaries is
observed while a very low proportion of third order (Σ27) twin boundaries are present.
CSL Boundaries for poly-Si film formed on glass at 425
o
C are shown in Fig 5.23. Twin
boundaries of Σ3, Σ9 and Σ27 are present. The boundaries of the twins are mainly Σ3
for both cases.
In order to study the extended defects Secco etching [Sec72] was used. Secco etching
can not be applied directly to the poly-Si film prepared by the ALILE process because
they are highly doped with Al. Therefore, epitaxially thickened poly-Si films were used
to study extended defects. For the epitaxial thickening of the poly-Si seed layer the
surface of poly-Si layers was polished by the chemical mechanical polishing (CMP)
process followed by a RCA cleaning and an oxide layer removal by a 1% HF dip. Then
the thin poly-Si seed layers have been epitaxially thickened. The thickness of layers
grown at 600 °C by e-beam evaporation was around 2 µm and the doping concentration
of Boron was about 4×10
16
cm
-3
. To make extended defects visible, on the sample
POLY-Si FILMS ON ZnO:Al COATED GLASS 85
surface Secco etching solution was applied. The Secco solution is HF/K
2
Cr
2
O
7
/H
2
O
obtained by mixing 2 parts of HF (49%) with 1 part of 0.15 Mol K
2
Cr
2
O
7
/H
2
O. The
samples were etched at room temperature for 7 seconds and the resulting etch pits were
analyzed by scanning electron microscopy (SEM).
Fig 5.24 (a) and (b) show the SEM images of Secco-etched poly-Si seed layer/absorber
layer stacks (epitaxial poly-Si films) on bare glass and on ZnO:Al coated glass,
respectively. From the images it is seen that while some of the grains exhibit better
quality epitaxial growth, some other grains exhibit more defective regions. The area of
the more defectively grown grains corresponds to about 35% of the total area of the
sample surface analyzed by SEM images. We know from EBSD and XRD analysis that
about 60% of the grains feature a (100) orientation and about 48% of the grains feature
a (400) orientation and 23% of the grains feature a (511) orientation for the poly-Si film
on ZnO:Al coated glass, which is favorable for epitaxial thickening at low temperatures.
This means that the more defective regions (35% of the total area) could be related to
underlying grains with orientations towards (110) and (111) orientation.
Fig 5.25 shows intragrain defects of poly-Si films (a) on glass and (b) on ZnO:Al coated
glass. It can be also seen that even less defectively grown grains exhibit dislocations
(circular pits in the figure). Comparison of Fig 5.25 (a) and (b) shows that underlying
ZnO:Al layer has no significant influence on the intra grain defect density. These
circular etch pits are characteristic for dislocations in Si (100) grains [Dog08]. A cross
sectional TEM image of a glass/poly-Si/epi-Si sample is shown in Fig 5.26. The glass
substrate is located at the bottom. The sample was shortly defect etched before sample
preparation for TEM to facilitate the selection of interesting regions for TEM analysis.
The traces of this short defect etching are visible at the top of the sample. The left and
the right part of the image show a clear difference of the surface. The absorber layer on
the right hand side contains a high density of twins within the cone that is visible and
twin and dislocations on the right side of this cone. The extended defects on the left
hand side are mainly dislocations. Because dislocations can not stop suddenly but end
by forming a loop, merging together or reaching a surface, the dislocations in the left
POLY-Si FILMS ON ZnO:Al COATED GLASS 86
20 mm
(a)
(b)
Figure 5. 22: EBSD analysis of a poly-Si film formed on ZnO:Al coated glass at
425 °C: (a) map of the twins (red = Σ3, violet = Σ9, green = Σ27), and (b) distribution of
coincident site lattice boundaries.
POLY-Si FILMS ON ZnO:Al COATED GLASS 87
20 mm
(a)
(b)
Figure 5. 23: EBSD analysis for poly-Si film formed on bare glass at 425 °C: (a) map
of the twins (red = Σ3, violet = Σ9, green = Σ27), and (b) distribution of coincident site
lattice boundaries.
POLY-Si FILMS ON ZnO:Al COATED GLASS 88
part of the image are in a direction not parallel to the image plane. Many of stacking
faults are present in the seed layer as can be seen in Fig 5.26.
(a)
(b)
Figure 5. 24: SEM images of Secco-etched thickened poly-Si films (~ 2µm) (a) on
glass and (b) on ZnO:Al coated glass.
POLY-Si FILMS ON ZnO:Al COATED GLASS 89
(a) (b)
Figure 5. 25: SEM images of Secco-etched epitaxial poly-Si films (a) on glass and (b)
on ZnO:Al coated glass.
cone
Seedlayer
Glass
Epi-Si
Figure 5. 26: Cross section TEM image of a glass/poly-Si/epi-Si sample. The glass
substrate is located at the bottom [Ges08b].
5.2.5
CONCENTRATION OF IMPURITIES
Fig 5.27 shows the secondary ion mass spectroscopy (SIMS) depth profile for
Aluminum and Boron in the poly-Si seed layer grown on ZnO:Al coated glass and on
glass at 425 °C annealing temperature, respectively. The calibration is only valid for the
poly-Si film. The concentration of Boron in the glass/ZnO:Al/poly-Si stack is one order
of magnitude lower than that in poly-Si on glass. This means that ZnO:Al films act as a
POLY-Si FILMS ON ZnO:Al COATED GLASS 90
diffusion barrier for Boron coming from the glass substrate. The concentration of Al in
the poly-Si film grown on ZnO:Al is similar to the Al concentration of poly-Si film
grown on glass.
Nast et al. [Nas01] showed by Hall Effect measurement that roughly 10% of the Al
content in poly-Si film on glass is electrically active at room temperature. As can be
seen in Fig. 5.27, the SIMS results revealed that the average Al concentration in the
poly-Si film formed on glass at 425 °C is about 9×10
19
cm
-3
. This means that the Al
concentration in poly-Si on glass is 3 times higher than 3×10
19
cm
-3
published by Nast
et al. [Nas01]. However, the carrier concentration of 3×10
17
cm
-3
in poly-Si film on bare
glass was measured at room temperature by the Hall measurements in our case. This
carrier concentration is one order lower than the carrier concentration measured by Nast
et al.. In case of glass/ZnO:Al/poly-Si stacks, the concentration of electrically active Al
in poly-Si films could not be deduced by Hall measurements. The resistivity of as-
deposited ZnO:Al layers and poly-Si films is 2.3×10
-4
and 2×10
-1
cm, respectively.
Therefore the carrier concentration and the Hall mobility of the ZnO:Al layer itself can
be measured. But the concentration of electrically active Al in poly-Si film on ZnO:Al
coated glass can be inferred from the result of poly-Si on glass. From SIMS data in Fig.
5.27 the average Al concentration in poly-Si film formed on ZnO:Al coated glass at
425 °C is about 3×10
20
cm
-3
. The 1% of the Al content in poly-Si film on ZnO:Al
coated glass would be assumed, the concentration of electrically active Al would be
3×10
18
cm
-3
.
During heat treatment steps (epitaxial growth, RTA, and H-passivation) boron atoms
can diffuse from the glass substrate to the thickened poly-Si films. The SIMS profile of
the boron concentration in the glass/SiN
X
/ZnO:Al/p+poly-Si seed layer/p-epi-Si stack
and the glass/p+poly-Si seed layer/p-epi-Si stack after RTA and H-passivation is shown
in Fig 5.28. The main boron source in thickened poly-Si seed layers is the glass
substrates. The boron concentration in the ALILE poly-Si film on glass is higher than
the boron concentration in the ALILE poly-Si film on SiN
X
/ZnO:Al coated glass. The
SiN
X
/ZnO:Al stack seems to act as a barrier layer for a diffusion of boron from the glass
substrates. Although the poly-Si seed layer has a high boron concentration, the B
POLY-Si FILMS ON ZnO:Al COATED GLASS 91
concentration in the epi-Si film is not influenced. It seems that the poly-Si seed layer act
as a barrier for a diffusion of boron. As can be seen in SIMS profile of Fig 5.28 the
boron concentrations in the epi-Si layer have similar values. This shows that boron
impurities could not diffuse from the glass substrate to the epi-Si layer during RTA and
H-passivation treatments.
0.1 1
10
16
10
17
10
18
10
19
10
20
10
21
10
22
10
23
10
24
Concentration
[
[
[
[
atoms/cm
3
]
]
]
]
Sputtering time [a.u]
B in poly-Si/ZnO:Al
B in poly-Si/glass
Al in poly-Si/ZnO:Al
Al in poly-Si/glass
Poly-Si ZnO:Al
Figure 5. 27: SIMS depth profile of the Al and the B concentration in the poly-Si seed
layer grown on ZnO:Al coated glass and bare glass (annealed at 425 °C for 16 hours).
POLY-Si FILMS ON ZnO:Al COATED GLASS 92
0.0 0.5 1.0
1015
1016
1017
1018
1019
1020
1021
1022
1023
Concentration
[
[
[
[
cm-3
]
]
]
]
Depth
Glass/poly-Si/Epi-Si
Glass/SiNx/ZnO:Al/poly-Si/Epi-Si
Epi-Si Seed layer
Figure 5. 28: SIMS profile of boron concentration in the thickened poly-Si seed layer
on glass and on SiN
x
/ZnO:Al coated glass after RTA and H-passivation treatment.
5.3 CONCLUSION
Studying the kinetics of the crystallization revealed that the activation energy for the
nucleation on ZnO:Al coated glass and on bare glass has similar values. Also the
activation energies for the grain growth on ZnO:Al coated glass and on bare glass have
similar values (1.8~1.9 eV). But the crystallization time on ZnO:Al coated glass is
shorter than the crystallization time on bare glass due to differnet value of the pre-factor
(see Fig 5.5) for the grain growth.
Continuous poly-Si films were formed on ZnO:Al coated glass by the ALILE process
and it was measured by Raman spectroscopy and UV-VIS spectroscopy method that the
crystalline quality of poly-Si films on ZnO:Al coated glass is similar to the crystalline
quality of poly-Si films on bare glass. The FWHM values of 4.1 cm
-1
and 3.8 cm
-1
have
been calculated for the poly-Si films formed on ZnO:Al coated glass and on bare glass
POLY-Si FILMS ON ZnO:Al COATED GLASS 93
at 425 °C as a ALILE process temperature, respectively. The crystalline quality
calculated from UV-VIS results (reflectance) is in the range of 98.5 ~ 99.4% (on both
ZnO:Al coated glass and bare glass).
The average grain size and maximum grain size of poly-Si films on ZnO:Al coated
glass is smaller than the average grain size and maximum grain size of poly-Si films on
bare glass. Although the preferential (100) orientation of poly-Si films on both ZnO:Al
coated glass and bare glass is about 60%, in case of poly-Si film on ZnO:Al coated glass
most of grains are located in a range of 0 ~ 10° (in a range of 10 ~ 20° in case of poly-Si
films on bare glass). No significant change of a preferential (100) orientation with
increasing annealing temperature was observed.
The surface of the thickened poly-Si films on ZnO:Al coated glass and on bare glass
observed with Secco etching shows only circular etch pits which are characteristic for Si
(100) dislocations in grains and it was observed that the ZnO:Al layer has no significant
influence.
POLY-Si FILMS ON ZnO:Al COATED GLASS 94
POLY-Si THIN-FILM SOLAR CELL 95
CHAPTER 6
P
OLY-Si THIN-FILM SOLAR CELLS
Polycrystalline silicon (poly-Si) thin-film solar cells on low-cost substrates can
significantly reduce manufacturing cost due to their potential for high efficiencies. In
practical efficiencies almost as high as far wafer based silicon solar cells should be
provided, however, the realisation of these efficiencies exceeding 15% remains
challenging. The application of ZnO:Al coated glass substrates would enable an
integrated series connection of solar cells simultaneously processed on large substrates,
and could provide an effective incoupling light and an efficient light trapping when
surface textured ZnO:Al films are applied [Rec06, Ber08].
The previous chapters described the electrical, optical, and structural properties of poly-
Si films on ZnO:Al coated glass. It has been shown that ZnO:Al films prepared by
sputtering and over coated by a thin Si film exhibit excellent electrical and optical
properties. Moreover, high structural quality of poly-Si films could be demonstrated.
The ZnO:Al coated glass is a promising alternative route for the preparation of low cost
thin film solar cells on nonconductive substrates. This chapter describes on the
realization of poly-Si thin film solar cells on ZnO:Al coated glass by using the
aluminum-induced layer exchange (ALILE) process described in the previous chapter.
6.1 PREPARATION AND STRUCTURE
The solar cell concept investigated here is based on the preparation of a thin, active
silicon base layer on a low-cost substrate covered with a conductive layer. Typical
features of a thin-film solar cell on a substrate covered with a conductive layer are the
following:
POLY-Si THIN-FILM SOLAR CELL 96
(i) The implementation of diffusion barriers for impurities from the glass and/or
highly conductive layers between substrate and silicon film for simplifying
the structure of solar cells, and
(ii) The application of ALILE process to enlarge the grain size of polycrystalline
silicon layers.
The preparation of ALILE poly-Si seed layers has been described in detail in section 3.1.
For the solar cells on ZnO:Al coated glass, the thin poly-Si seed layers produced by the
ALILE process at 450
o
C on ZnO:Al coated glass have been epitaxially thickened to
form the absorber layer of the solar cell. The epitaxial absorbers are ~ 2 µm thick grown
at 600 °C by e-beam evaporation and the doping concentration of boron is about 4×10
16
cm
-3
. To prevent crystallization of the a-Si:H emitter, defect passivation is realized at
600 °C by exposing the samples to a hydrogen plasma prior to the deposition of n
+
-type
a-Si:H hetero-emitter with TCO (Al doped ZnO) on top. For the defect passivation of
the poly-Si seed layer/absorber layer stacks the hydrogen plasma was applied for 10
minutes at a substrate temperature of 600 °C. It features a hollow-cathode RF high
density plasma source (13.56 MHz) that was used in a simple diode configuration
[Gor07c]. A defect annealing process typically performed at 900-1000 °C [Rau08] was
not applied. A hetero-junction emitter layer of about 10 nm n
+
-type a-Si:H has been
deposited by using plasma enhanced chemical vapor deposition (PECVD) using
conditions optimized for wafer based Si solar cells [Kor07]. Interdigitated contacting
using Al is done by photolithography. The sample area is 4×4 mm
2
with an emitter area
of 8.6 mm
2
. The schematics of the solar cell on ZnO:Al coated glass is shown in Fig 6.1.
The solar cells on ZnO:Al coated glass were designed in the so-called “substrate-type”
(the light enters the cell through the air side) configuration. The structure of solar cells
is glass/ZnO:Al/p
+
seed layer/p
-
epi-Si/n
+
a-Si/ZnO/Al. The total thickness of the solar
cells is ~ 2.3 m. During the current-voltage measurements, a white reflector (several
sheets of white papers) was used on the backside while the light was irradiated on the
layer side. The efficiency and short-current density was calculated using the emitter area.
POLY-Si THIN-FILM SOLAR CELL 97
n+ a-Si:H
Borofloat33glass
Light
ZnO
a-Si:H,n+
Poly-Si,p
Poly-Si,p+
Interdigitatedcontacts
ZnO:Al
Figure 6. 1: Schematics of the solar cell on ZnO:Al coated glass: “substrate-type” (the
light enters the cell through the air side) configuration and interdigitated contacts.
6.2 SOLAR CELL RESULTS
In this section solar cell the results achieved on ZnO:Al coated glass obtained in this
study are described (section 6.2.1) and compared to previous solar cell results on bare
glass which are reviewed and discussed (section 6.2.2).
6.2.1 SOLAR CELL ON ZnO:Al COATED GLASS
The current-voltage (I-V) curve of the first solar cell on ZnO:Al coated glass is plotted
in Fig 6.2 (for standard test conditions: AM1.5, 100 mW/cm
2
, 25 °C). The thin poly-Si
seed layer was produced on ZnO:Al coated glass at 450
o
C for 16 hours by ALILE
process. The solar cell on ZnO:Al coated glass showed an open-circuit voltage (V
OC
) of
POLY-Si THIN-FILM SOLAR CELL 98
389 mV, a short-current density (J
SC
) of 9.1 mA/cm
2
and an efficiency of 2%. The
emitter area (8.6 mm
2
) of the solar cell was considered for calculating the efficiency of
the solar cell. The series resistance of the solar cell on ZnO:Al coated glass is 7 cm
2
.
As a first test of the feasibility of poly-Si rear junction cells based on ALILE process,
the illuminated I-V parameters of an existing poly-Si cell on ZnO:Al coated glass were
measured in both substrate and superstrate configuration (see Table 6.1). We note that
the structure of this cell was designed for the substrate configuration (see Fig. 6.1). The
absorber layer was grown by e-beam evaporation method and its thickness was around 2
µm. Due to the fact that the as-processed devices are essentially bifacial, white paper
was used as reflector in both configurations. In superstrate configuration, the short-
current density of the cell is slightly lower than in substrate configuration. Gordon et al.
[Gor07a] asserted that J
SC
is decreased due to the absence of an anti-reflective coating
(ARC) in superstrate configuration. In the case of solar cells illiuminated throughthe
glass, the thickness of the ZnO:Al layer is about 700 nm and does not act as an ARC
layer. The difference in V
OC
between both configurations corresponds to what is
expected from the decrease in short-circuit current density [= n × (kT/q) × ln(9.1/7.7)]
taken from the simple diode equation for solar cell as a rough estimate. Here, n is the
ideality factor, k is the Boltzmann’s constant, T is the temperature, and q is the Planck’s
constant.
The distribution of the efficiency and the open-circuit voltage of solar cells on a single
ZnO:Al coated glass substrate is shown in Fig 6.3. The entire size of the ZnO:Al coated
glass substrate is 2.54 by 1.25 cm
2
. Most of the measured open-circuit voltages are quite
similar. This means that most of the areas feature a similar material quality. The open-
circuit voltage of solar cells on ZnO:Al coated glass with poly-Si seed layers formed at
425 and 450 °C is compared in Table 6.2. The V
OC
of solar cells with poly-Si seed layer
formed at 425 °C is lower than that of solar cells with poly-Si seed layer formed at
450 °C. This reason could be that a poly-Si film formed on ZnO:Al coated glass at
425 °C has a lot of very small grains (less than 1 µm) which were not observed at higher
temperature (as shown in Fig 5.16 (b)). Although large grains cover on most of the
POLY-Si THIN-FILM SOLAR CELL 99
surface it is expected that the performance of the resulting poly-Si layers will be mainly
limited by the low-quality (small grains) regions.
0 100 200 300 400
0
2
4
6
8
10
Emitter area
H-passivation/ No RTA
White relflector
J
sc
= 9.1 mA/cm
2
V
oc
= 389 mV
FF = 57 %
η
ηη
η
= 2 %
Voltage [mV]
Current Density
[
[
[
[
mA/cm
2
]
]
]
]
Figure 6. 2: Current density versus voltage (I-V) of the solar cell structure of glass/
ZnO:Al/p+ seed layer/p- epi-Si/n+ a-Si/ZnO/Al stacks.
TABLE 6. 1: Comparison of substrate and superstrate structure of solar cell on ZnO:Al
coated glass.
Structure J
SC
(mA/cm
2
) V
OC
(mV) FF
(%)
η
(%)
Substrate 9.1 389 57 2.0
Superstrate
7.7 376 57.4 1.5
POLY-Si THIN-FILM SOLAR CELL 100
1 2 3 4 5 6 7
0.0
0.5
1.0
1.5
2.0
2.5
0.35
0.40
0.45
0.50
Cell No.
Efficiency
Efficency [%]
VOC
VOC [V]
Figure 6. 3: Distribution of efficiency and open-circuit voltage of solar cells on ZnO:Al
coated glass on a single substrate.
TABLE 6. 2: Comparison of open-circuit voltage (V
OC
) of solar cells on ZnO:Al coated
glass for poly-Si seed layers formed at 425 and 450 °C.
T
A
(°C) V
OC
(mV)
425 316
450 372
6.2.2 SOLAR CELL ON GLASS
In the section 6.2.2 the solar cell results on Borofloat33 glass reported recently are
reviewed and compared with the solar cell result on ZnO:Al coated glass. These solar
cells on bare glass were also based on the seed layer formed at 450 °C for 16 hours by
the ALILE process.
POLY-Si THIN-FILM SOLAR CELL 101
The current-voltage characteristic of the poly-Si thin-film solar cell on bare glass
obtained by Dogan et al. [Dog08] is shown in Fig 6.4 (for standard test conditions:
AM1.5, 100 mW/cm
2
, 25 °C). The solar cell on bare glass was measured with white
reflector. For the measurements a white paper was placed below the glass substrate. The
solar cell has interdigitated contacts and an area of 16 mm
2
. The short-current density
was calculated using the emitter area of the solar cell: 8.6 mm
2
. Together with an open-
circuit voltage V
OC
of 406 mV, a short circuit current density J
SC
of 10.3 mA/cm
2
and a
fill factor FF of 67.4% this had led to an emitter area efficiency of 2.8%. For
comparison the current-voltage characteristics of the thin-film solar cell prepared so far
on an ‘ideal seed layer’ (a p
+
-type mono-crystalline Si(100) wafer) was measured (see
ref. [Dog08]). The wafer-based thin-film solar cell showed an emitter area short circuit
current density J
SC
of 15.9 mA/cm
2
. The emitter area efficiency of this wafer-based
thin-film solar cell was 7.3% (V
OC
= 583 mV, FF = 78.7%).
0 100 200 300 400 500
0
2
4
6
8
10
12
Current Density
[
[
[
[
mA/cm2
]
]
]
]
Voltage [mV]
Emitter area
H-passivation / No RTA
White reflector
JSC = 10.3 mA/cm2
VOC = 406 mV
FF = 67.4 %
η
= 2.8 %
Figure 6. 4: Emitter area current density versus voltage for the thin-film solar cell on
bare glass. The 2 µm thickness of absorber layer was grown at 600 °C. The solar cell
was measured with a white reflector.
POLY-Si THIN-FILM SOLAR CELL 102
Current-voltage curves of the solar cell on bare glass before (black line) and after H-
passivation (red line) are shown in Fig 6.5. The influence of H-passivation on solar cells
was investigated by Gorka et al. [Gor08]. Current-voltage measurements of the solar
cell test structures were carried out under standard test conditions (AM1.5 illumination,
100 mW/cm
2
, 25 °C). The emitter area was taken for the calculations and an underlying
black reflector was used during the measurements to discard any reflections. After H-
passivation the V
OC
increased strongly for the solar cell on glass from 223 mV (as-
grown) to 351 mV. It seems that mostly the V
OC
is affected by the passivation. The
short circuit current J
SC
and the fill factor (FF) were improved by a factor of 1.2 so that
the overall efficiency is more than doubled after H-passivation. In this study the solar
cell was passivated by Hydrogen plasma at 520°C for 10 min. Fig 6.6 shows the
dependence of the average open-circuit voltage (red circles) and the short circuit current
density (black squares) on the substrate temperature during H-passivation. The V
OC
increased notably with the temperature. The highest average value of 385 mV was
achieved for hydrogenation at 620 °C. The short circuit current densities were not
influenced significantly by different temperatures.
The influence of the rapid thermal annealing process (RTA) on solar cells was
investigated by B. Rau [Rau08]. In this study a condition of the RTA process was
900 °C for 200 seconds. Improved solar cell performances were achieved by applying
the RTA process. And the external quantum efficiency (EQE) of solar cells applied only
H-passivation and both RTA and H-passivation was discussed. In the shorter
wavelength region, i.e. close to the pn-junction, both spectra are almost identical. The
additional RTA treatment increased the EQE for wavelengths above 450 nm indicating
the improved material quality. The EQE maximum shifts slightly to longer wavelengths.
The improved material quality can also be seen on the solar cell parameters: V
OC
= 405
mV, J
SC
= 8.3 mA/cm², Fill factor (FF) = 63% and efficiency (η) = 2.1% for the sample
with both treatments and V
OC
= 360 mV, J
SC
= 7.4 mA/cm², FF = 56% and η= 1.5% for
the sample with hydrogenation only.
POLY-Si THIN-FILM SOLAR CELL 103
0 100 200 300 400
0
2
4
6
8
10
JSC = 6.5 mA/cm2
VOC = 351 mV
FF = 60 %
η
= 1.4 %
H-passivation
Current Density
[
[
[
[
mA/cm2
]
]
]
]
Voltage [mV]
as-grown
Emitter area
Black reflector
JSC = 5.3 mA/cm2
VOC = 223 mV
FF = 52 %
η
= 0.6 %
Figure 6. 5: Current density versus voltage characteristics of solar cells on glass before
(black line) and after H-passivation (red line).
450 500 550 600 650
4
6
8
10
320
340
360
380
400
Current Density
[
[
[
[
mA/cm2
]
]
]
]
Process temperature
[
[[
[
oC
]
]]
]
Voltage [mV]
JSC
VOC
Figure 6. 6: Dependence of the average open-circuit voltage (red circles) and the short
circuit current density (black squares) on the substrate temperature during H-passivation.
POLY-Si THIN-FILM SOLAR CELL 104
The solar cell performances on bare glass reviewed above are summerized in Table 6.3.
The best efficieny of the solar cell on bare glass is 2.8% realized by by P. Dogan
[Dog08]. Although this efficency is a higher than an efficiency of 2% of the solar cell
on ZnO:Al coated glass achieved in this study, it can not be compared directly. The
deposition parameters of the absorber layer were changed during the fabrication of the
best solar cell on bare glass. Nevertheless, the solar cell on ZnO:Al coated glass shows a
possibility to be improved.
The solar cell on ZnO:Al coated glass shows a better performance compared to the solar
cell results on bare glass measured by B. Gorka [Gor08] and B. Rau [Rau08]. These
results measured by B. Gorka and B. Rau show the lower values compared to the results
achieved in this study. A short circuit current density (J
SC
) of the solar cell on ZnO:Al
coated glass measured with black reflector and with white reflector in this study is 7.8
and 9.1 mA/cm², respectively. A J
SC
of the solar cell on bare glass with black reflector
and with white reflector was 6.5 from [Gor08] and 7.4 mA/cm² from [Rau08],
respectively. An open-circuit voltage (V
OC
) of the solar cell on ZnO:Al coated glass is a
slightly higher than a V
OC
of the solar cells on bare glass. But a Fill Factor (FF) of the
solar cells on bare glass and on ZnO:Al coated glass are similar. An improved J
SC
and
V
OC
on the solar cell on ZnO:Al coated glass were achieved.
TABLE 6. 3: Recent results in our laboratory.
Ref. η
[%] J
SC
[mA/cm
2
] V
OC
[mV] FF
[%]
[Dog08] 2.8 10.3 406 67.4
[Gor08] 1.4 6.5 351 60
[Rau08] 1.5 7.4 360 56
This study 2 9.1 389 57
POLY-Si THIN-FILM SOLAR CELL 105
6.3 CONCLUSION AND OUTLOOK
The best poly-Si solar cell on ZnO:Al coated glass has achieved an open-circuit voltage
(V
OC
) of 389 mV, a current density (J
SC
) of 9.1 mA/cm
2
and an efficiency of 2%. V
OC
and J
SC
results on ZnO:Al coated glass are only slightly lower than results recently
published for the solar cell on bare glass [Dog08]. One of the reasons for the difference
in the efficiency should be the series resistance of the solar cell.
These world-wide first large-grained poly-Si solar cells realized on ZnO:Al coated glass
may serve as a first proof of concept. However, there are still many open questions
which have to be addressed and solved on the path to a potentially highly efficient solar
cell.
1) High temperature treatments for defect annealing have to be developed which
are applicable to the glass/ZnO:Al/Si device structure
2) Adapted barrier coatings have to be developed to avoid cross contamnations
3) Chemical processes which are compatible with the ZnO:Al coated glass
substrate have to be developed.
POLY-Si THIN-FILM SOLAR CELL 106
CONCLUSIONS 107
CHAPTER 7
C
ONCLUSIONS
In this study the formation of large-grained poly-Si films on transparent conductive
oxide (TCO) coated glass is addressed in order to open up new possibilities for the
fabrication of photovoltaic devices. Being ahead of the objective the realization of
temperature stable Al doped ZnO and the successful formation of poly-Si layers on the
ZnO:Al layer should be preceded for the above purpose. The investigated key aspects
have been (i) the temperature stability of the ZnO:Al films which are capped with a
poly-Si layer, (ii) the study of poly-Si thin films formed on ZnO:Al coated glass and ,for
comparison, on bare glass by the aluminium-induced layer exchange (ALILE) process,
and (iii) the fabrication of first poly-Si thin film solar cells on ZnO:Al coated glass.
The resistivity of ZnO:Al layers covered with poly-Si films formed by ALILE or SPC is
improved even though the resistivity of uncapped ZnO:Al layer is decreased by heat
treatment in nitrogen ambient. The resulting sheet resistance of the glass/ZnO:Al/poly-
Si samples formed by ALILE and by SPC is almost independent of the annealing
temperature.
On both ZnO:Al coated glass substrates and bare glass substrates the activation energy
for the nucleation and the grain growth during the ALILE process is 1.1~1.4 eV and
1.7~1.9 eV, respectively. Nevertheless, the ALILE process time on ZnO:Al coated glass
is shorter than the ALILE process time on bare glass.
Poly-Si films formed on ZnO:Al coated glass and on bare glass have the same
crystalline quality as determined from the FWHM values deduced from the Raman
results, thereof it could be concluded that ZnO:Al films did not determine the change of
the crystal quality of poly-Si films. The quality factors calculated from the optical
reflectance spectra of poly-Si films on ZnO:Al coated glass and on bare glass show the
same values (~ 99%) and are not affected by the annealing temperature. Hence we
CONCLUSIONS 108
conclude that ZnO:Al layer and the ALILE process temperature cut down the
crystallization time without a falling-off in the crystalline quality of poly-Si films .
The preferential (100) orientation of poly-Si films on both ZnO:Al coated glass and bare
glass is about 60% and no significant change of the preferential (100) orientation with
increasing annealing temperature was observed using EBSD. Also the total fractions of
the (400) and (511) orientations, tilted within 20° of the perfect (400) orientation,
calculated from XRD resuls and measured on larger areas have similar values as the
values extracted by EBSD results which give information on a small area.
Even though few holes which are a typical feature of the ALILE process are revealed at
grain boundaries by SEM investigations, it can be stated that the poly-Si film formed on
ZnO:Al coated glass by ALILE process is continuous in general.
The electrical properties of ZnO:Al/poly-Si stacks and the Si-Si TO-LO phonon band
position of poly-Si film on ZnO:Al coated glass are not significantly influenced by H-
passivation while the electrical properties (mobility and resistivity) of ZnO:Al/poly-Si
stacks are improved and the Si-Si TO-LO phonon band position of poly-Si film on
ZnO:Al coated glass is shifted by RTA treatment.
The solar cell on ZnO:Al coated glass has achieved a open-circuit voltage (V
OC
) of 389
mV, a short-current density (J
SC
) of 9.1 mA/cm
2
and an efficiency of 2%.
In this thesis large-grained poly-Si films and solar cells on ZnO:Al coated glass were
realized for the first time and it was shown that the utilization of the ZnO:Al is a
promising option for the preparation of high efficiency poly-Si thin-film soalr cells.
109
ABBREVIATIONS, SYMBOLS AND UNITS
ABBREVIATION MEANING
AIC Aluminium-induced crystallization
ALILE Aluminum-induced layer exchange process
APCVD Atmospheric pressure chemical vapor deposition
a-Si:H Hydrogenated amorphous silicon
CIS Copper-indium-selenide (sulfite)
CIGS Copper indium gallium selenide
CdTe Cadmium telluride
CMP Chemical mechanical polishing
CSG Crystalline silicon on glass
EBSD Electron back scatter diffraction
ECRCVD Electron cyclotron resonance chemical vapor deposition
EPIA European photovoltaic industry association
FOx Flowable oxide
HR-TEM High resolution transmission electron microscope
IAD Ion assisted deposition
LC Laser crystallization
MW Megawatts
nc-Si Nano-crystalline silicon
PECVD Plasma enhanced chemical vapor deposition
Poly-Si Large-grained polycrystalline silicon
PV Photovoltaic
PVD Physical vapor deposition
RTA Rapid thermal annealing
SAC Selected area channeling
SPE Solid phase epitaxy
SPC Solid phase crystallization
TEM Transmission electron microscope
TCO Transparent conductive oxide
c-Si:H Hydrogenated microcrystalline silicon
W
p
Watt peak
ZnO:Al Al doped ZnO
110
SYMBOL UNIT MEANING
A µm
2
Area of optical microscope image
d
G
µm Grain size
E
A
eV Activation energy
FF % Fill Factor
I
SC
mA/cm
2
Short circuit current
k eV/K Boltzmann constant (8.6215 ×10
-5
eV/K)
n ideality factor (1 < n < 2)
N
G
mm
-2
Nucleation density
Q % Quality factor
q C Absolute value of electron charge
R
e1
% Reflectance at e1 peak
R
e2
% Reflectance at e2 peak
T
A
°C Annealing temperature
T
ALILE
°C Process temperature of ALILE
T
RTA
°C Process temperature of RTA
t min Time
t
N
min Nucleation time
V
OC
mV Open-circuit voltage
η % Efficiency of solar cell
111
R
EFERENCES
[CHAPTER 1]
[And06] G. Andra, J. Bergmann, A. Bochmann, F. Falk, A. Gawlik, E. Ose, J. Plentz, S.
Dauwe, and T. Kieliba, Conference Record of the 2006 IEEE 4th World
Conference on Photovoltaic Energy Conversion. Waikoloa, HI, USA, (2006) 4
[Deb00] Satyen K. Deb, NREL/CP-590-28060 (2000)
[Dim07] D. Dimova-Malinovska, O. Angelov, M. Kamenova, A. Vaseashta, and J. C.
Pivin, J. Optoelectron Adv Mater. 9 (2007) 355
[Ear07] http://www.earthpolicy.org/Indicators/Solar/2007.htm (Dec. 27, 2007)
[EPIA] www.epia.org/
[EPI08] EPIA, Solar Generation V 2008
[Fuh04] W. Fuhs, S. Gall, B. Rau, M. Schmidt, and J. Schneider, Solar Energy 77
(2004) 961
[Ges06] D. Van Gestel, I. Gordon, L. Carnel, K. Van Nieuwenhuysen, J. D´Haen, J.
Irigoyen, G. Beaucarne, and J. Poortmans, Thin Solid Films 511/512 (2006) 35
[Gre04] M.A. Green, P.A. Basore, N. Chang, D. Clugston, R. Egan, R. Evans, D. Hogg,
S. Jarnason, M. Keevers, P. Lasswell, J. O’Sullivan, U. Schubert, A. Turner, S.R.
Wenham, and T. Young, Solar Energy 77 (2004) 857
[Im93] J. S. Im, H. J. Kim, and M. O. Thompson, Appl. Phys. Lett. 63 (1993) 1969
[Kee07] M.J. Keevers, T.L. Young, U. Schubert, and M.A. Green, Proc. of the 22nd
European Photovoltaic Solar Energy Conference, Milan, Italy (2007) 1783
[Kim02] H. Kim, D. Kim, G. Lee, D. Kim, and S.H. Lee, Sol. Energy Mater. Sol. Cells
74 (2002) 323
[Klu03] O. Kluth, G. Schöpe, J. Hüpkes, C. Agashe, J. Mueller, and B. Rech, Thin Solid
Films 442 (2003) 80
[Mat01] Y.Matsumoto and Z. Yu, Jpn. J. Appl. Phys. 40 (2001) 2110
112
[Mat90] T. Matsuyama, K. Wakisaka, M. Kameda, M. Tanaka, T. Matsuoka, S. Tsuda,
S. Nakano, Y. Kishi, and Y. Kuwano, Jpn. J. Appl. Phys. 29 (1990) 2327
[Mul04] J. Muller, B. Rech, J. Springer, and M. Vanecek, Solar Energy 77 (2004) 917
[Nas00] O. Nast and A. J. Hartmann, J. Appl. Phys. 88 (2000) 124
[Nas98] O. Nast, T. Puzzer, L.M. Koschier, A.B. Sproul, and S.R. Wenham, Appl. Phys.
Lett. 73 (1998) 3214
[Rau04] B. Rau, I. Sieber, J. Schneider, M. Muske, M. Stöger-Pollach, P.
Schattschneider, S. Gall, and W. Fuhs, J. Cryst. Growth 270 (2004) 396
[Rec06] B. Rech, T. Repmann, M.N. van den Donker, M. Berginski, T. Kilper, J.
Hüpkes, S. Calnan, H. Stiebig, and S. Wieder, Thin Solid Films 511/512 (2006)
548
[Sch05] J. Schneider, J. Klein, M. Muske, S. Gall, and W. Fuhs, Appl. Phys. Lett. 87
(2005) 031905
[She08] S. Sheng, Y.K. Chae, L. Li, J. Su, and X. Yang, Semiconductor International
(Nov. 1, 2008)
[Sol08] http://www.solarbuzz.com/Marketbuzz2008-intro.htm (March 17, 2008)
[Sta77] D. L. Staebler and C. R. Wronski, Appl. Phys. Lett. 31 (1977) 292
[Tan03] M. Tanaka, S. Okamoto, S. Tsuge, and S. Kiyama, Proc. 3rd world conference
(2003)
[Vou03] A.T. Voutsas, Appl. Surf. Sc. 208/209 (2003) 250
113
[CHAPTER 2]
[Abe05] A.G. Aberle, P.I. Widenborg, D. Song, A. Straub, M.L. Terry, T. Walhs, A.
Sproul, P. Campbell, D. Inns, and B. Beilby, In Proc. of the 31st IEEE PV
Conference (2005)
[Ada05] S. Adachi, Properties of Group-IV, III-V and II-VI Semiconductors, John
Wiley and Sons, Ltd,West Sussex, England, 2005.
[Add99] M.L. Addonizio, A. Antonaia, G. Cantele, C. Privato, Thin Solid Films 349
(1999) 93
[Aga04] C. Agashe, O. Kluth, J. Hüpkes, U. Zastrow, B. Rech and M. Wuttig, J. Appl.
Phys. 95 (2004) 1911
[Al05] M. Al-Barghouti, H. Abu-Safe, H. Naseem, W.D. Brown, and M. Al-Jassim, J.
Electrochem. Soc. 152 (2005) G354
[Boa76] R.L. Boatright and J.O. McCaldin, J. Appl. Phys. 47 (1976) 2260
[Boh96] E. Böhmer, F. Siebke, B. Rech, C. Beneking and H. Wagner, Mater. Res. Soc.
Symp. Proc. 426 (1996) 519
[Bos96] S. Bose, S. Ray, A.K. Barua, J. Appl. Phys. 29 (1996) 1659
[Bre83] M.J. Brett, R.W. McMahon, J.D. Affinito and R. Parsons, J. Vac. Sci. Technol.
A 1 (1983) 352
[Bre99] S. Brehme, F. Fenske, W. Fuhs, E.Nebauer, M. Poschenrieder, B. Selle, I.
Sieber, Thin Solid Films 342 (1999) 167
[Cai92] W. Cai, and D. Wan, Thin Solid Films 219 (1992) 1
[Ceb98] R. Cebulla, R. Wendt, K. Ellmer, J. Appl. Phys. 83 (1998) 1087
[Cha80] B. Chapman, Glow Discharge Processes. John Wiley & Sons, 1980
[Cho83] K. L. Chopra, S. Major and D. K. Pandya, Thin Solid Films 102 (1983) 1
[Con50] E. Conwell and V.F. Weisskopf, Phys. Rev. 77 (1950) 388
[Dim06] D. Dimova-Malinovska, V. Grigorov, M. Nikolaeva-Dimitrova, O. Angelov,
and N. Peev, Thin Solid Films 501 (2006) 358
[Dim07] D. Dimova-Malinovska, O. Angelov, M. Kamenova, A. Vaseashta and J. C.
Pivin, J. Optoelectron Adv Mater. 9 (2007) 355
114
[Dog08] P. Dogan, E. Rudigier, F. Fenske, K.Y. Lee, B. Gorka, B. Rau, E. Conrad, and
S. Gall, Thin Solid Films 516 (2008) 6989
[Eka03] G. Ekanayake, S. Summers, and H. S. Reehal, Proc. of the 3rd WCPEC,
volume B (2003) 1867
[Eka04] G. Ekanayake, Z. Lu, and H. S. Reehal, Proc. of the EPVSEC 19th (2004)
[Ell01] K. Ellmer, J. Phys. D: Applied Physics 34 (2001) 3097
[Ell94] K. Ellmer, F. Kudella, R. Mientus, R. Schieck and S. Fiechter, Thin Solid Films
247 (1994) 15
[Fay00] S. Fay, S. Dubail, U. Kroll, J. Meier, Y. Ziegler and A. Shah, Proceedings of
the 16th European Photovoltaic Solar Energy Conference. Glasgow, UK, 2000
[Gal06] S. Gall, J. Schneider, J. Klein, K. Hübener, M. Muske, B. Rau, E. Conrad, I.
Sieber, K. Petter, K. Lips, M. Stöger-Pollach, P. Schattschneider, and W. Fuhs,
Thin Solid Films 511/512 (2006) 7
[Ges08a] D.V. Gestel, I. Gordon, A. Verbist, L. Carnel, G. Beaucarne, and J. Poortmans,
Thin Solid Films 516 (2008) 6907
[Gju04] M. Gjukic, M. Buschbeck, R. Lechner, and M. Stutzmann, Appl. Phys. Lett. 85
(2004) 2134
[Gju05] M. Gjukic, R. Lechner, M. Buschbeck, and M. Stutzmann, Appl. Phys. Lett. 86
(2005) 3
[Gla00] H. J. Gläser, Large Area Glass Coating. Bd. 1. von Ardenne Anlagentechnik,
2000
[Gor07] I. Gordon, L. Carnel, D. Van Gestel, G. Beaucarne, and J. Poortmans, Progress
in Photovoltaics: Research and Applications 15 (20077) 575
[Gre76] J. E. Greene, and L. Mei, Thin Solid Films 37 (1976) 429
[Gri06] V. Grigorov, O. Angelov, M. Sendova-Vassileva, and D. Dimova-Malinovska,
Thin Solid Films 511-512 (2006) 381
[Gro01] R. Groenen, J. Loeffler, P.M. Sommeling, J.L. Linden, E.A.G. Hamers, R.E.I.
Schroppand M.C.M. van de Sanden, Thin Solid Films 392 (2001) 226
[Har00] N. P. Harder, J. A. Xia, S. Oelting, O. Nast, P. Widenborg, and A. G. Aberle,
Conference Record of the Twenty-Eighth IEEE Photovoltaic Specialists
Conference (2000)
115
[Har77] J. M. Harris, R. J. Blattner, I. D. Ward, Jr C. A. Evans, H. L. Fraser, M. A.
Nicolet, and C. L. Ramiller, J. App. Phys. 48 (1977) 2897
[Har91] G.L. Harding, B. Window, E.C. Horrigan, Sol. Energ. Mater. 22 (1991) 69
[Hau01] F.J. Haug, Z. Geller, H. Zogg, A.N. Tiwari, C. Vignali, J. Vac. Sci. Technol. A
19 (2001) 171
[Hel82] K.-H. Hellwege, Semiconductors. Bd. 17 b: Landolt-Börnstein Numerical Data
and Functional Relationships. Springer-Verlag, 1982
[Hic73] F.S Hickernell, J. Appl. Phys 44 (1973) 1061
[Hsu03] C. M. Hsu, I.F. Chen, and M.C. Yu, Jpn. J. Appl. Phys. 42 (2003) 4928
[Hup06] J. Hüpkes, B. Rech, S. Calnan, O. Kluth, U. Zastrow, H. Siekmann and M.
Wuttig, Thin Solid Films 502 (2006) 286
[Iba03] H. Ibach and H. Lüth, Solid State Physics. Bd. 3. Springer-Verlag, 2003
[Iga88] Y. Igasaki, M. Ishikawa, G. Shimaoka, Appl. Surf. Sci. 33/34 (1988) 926
[Iga91] Y. Igasaki, H. Saito, J. Appl. Phys. 29 (1991) 3613
[Ito03] T. Ito, H. Fukushima, and M. Yamaguchi, Jpn. J. Appl. Phys. 42 (2003) 1526
[Ito04] T. Ito, H. Fukushima, and M. Yamaguchi, Sol. Energy Mater. Sol. Cells, 83
(2004) 91
[Iza96] M. Izaki, T. Omi, J. Appl. Phys. 68 (1996) 2439
[Jae08] C. Jaeger, T. Antesberger, and M. Stutzmann, J. Non-Cryst. Solids 354 (2008)
2314
[Jin88] Z.C. Jin, I. Hamberg, C.G. Granqvist, Thin Solid Films 164 (1988) 381
[Kim02] H. Kim, D. Kim, G. Lee, D. Kim, and S. H. Lee, Sol. Energy Mater. Sol. Cells,
74 (2002) 323
[Kim96] J.H. Kim and J.Y. Lee, Jpn. J. Appl. Phys. 35 (1996) 2052
[Kis03] R. Kishore, C. Hotz, H. A. Naseem, and W. D. Brown, J. App. Crystal. 36
(2003) 1236
[Kle04] J. Klein, J. Schneider, M. Muske, S. Gall, and W. Fuhs, Thin Solid Films 451-
452 (2004) 481
116
[Klu03] O. Kluth, G. Schöpe, J. Hüpkes, C. Agashe, J. Müller and B. Rech, Thin Solid
Films 442 (2003) 80
[Kon03] M. Kon, P.K. Song, Y. Shigesato, P. Frach, S. Ohno, K. Suzuki, Jpn. J. Appl.
Phys. 42 (2003) 263
[Kon92] T. J. Konno, and R. Sinclair, Philosophical Magazine B 66 (1992) 749
[Kon93] T. J. Konno, R. Sinclair, A. G. Cullis, A. E. Staton-Bevan, and J. L. Hutchison,
Proceedings of the Royal Microscopical Society Conference (1993)
[Kon94] T. J. Konno, and R. Sinclair, Mater. Sci. Eng. A 179/180 (1994) 426
[Kub96] M. Kubon, E. Böhmer, F. Siebke, B. Rech, C. Beneking and H. Wagner, Sol.
Energ. Mater. Sol. Cell. 41-42 (1996) 485
[Lor03] M. Lorenz, E.M. Kaidashev, H.von Wenckstern, V. Riede, C. Bundesmann, D.
Spermann, G. Benndorf, H. Hochmuth, A.R. Semmelhack and M. Grundmann,
Solid-State Electronics 47 (2003) 2205
[Maj77] G. Majni and G. Ottaviani, Appl. Phys. Lett. 31 (1977) 125
[Maj78] G. Majni and G. Ottaviani, Thin Solid Films 55 (1978) 235
[Maj79] G. Majni and G. Ottaviani, J. Cryst. Growth 46 (1979) 119
[Mal00] N. Malkomes, T. Yamamoto, T. Miyata, Thin Solid Films 366 (2000) 63
[Man81] J. C. Manifacier, J. P. Fillard and J. M. Bind, Thin Solid Films 77 (1981) 67
[Men98] R. Menner, R. Schäffler, B. Sprecher and B. Dimmler, in Proceedings of the
2nd World Conference on Photovoltaic Energy Conversion, Vienna, 1998
[Min84] T. Minami, H. Nanto and S. Takata, Jpn. J. Appl. Phys. 23 (1984) L280
[Min90] T. Minami, K. Oohashi, S. Takata, T. Mouri, and N. Ogawa, Thin Solid Films
193/194 (1990) 721
[Mor09] H. Morkoç and Ü. Özgür, Zinc Oxide: Fundamentals, Materials and Device
Technology, 2009
[Mul01] J. Müller, O. Kluth, S. Wieder, H. Siekmann, G. Schöpe, W. Reetz, O. Vetterl,
D. Lundszien, A Lambertz, F. Finger, B. Rech and H. Wagner, Sol. Energ.
Mater. Sol. Cell. 66 (2001) 275
[Mul03] J. Müller, G. Schöpe, O. Kluth, B. Rech, V. Sittnger, B. Szyszka, R. Geyer, P.
Lechner, H. Schade, M. Ruske, G. Dittmar and H.-P. Bochem, Thin Solid Films
442 (2003) 158
117
[Nas00] O. Nast and A. J. Hartmann, J. Appl. Phys., 88 (2000) 716
[Nas00S] O. Nast and S.R. Wenham, J. Appl. Phys. 88 (2000) 124
[Nas01] O. Nast, S. Brehme, S. Pritchard, A. G. Aberle, and S. R. Wenham, Sol. Energy
Mater. Sol. Cells, 65 (2001) 385
[Nas98] O. Nast, T. L. M. Putter, Koschier, A. B. Sproul, and S. R.Wenham, App. Phys.
Lett. 73 (1998) 3214
[Ott79] G. Ottaviani, and G. Majni, J. Appl. Phys. 50 (1979) 6865
[Par97] K.C. Park, D.Y. Ma, K.H. Kim, Thin Solid Films 305 (1997) 201
[Par99] D.F. Paraguay, L.W. Estrada, N.D. R. Acosta and E. Andrade, Thin Solid Films
350 (1999) 192
[Pen08] S. Peng, X. Shen, Z. Tang, and D. He, Thin Solid Films 516 (2008) 2276
[Pih04] E. Pihan, A. Slaoui, P. R. I. Cabarrocas, and A. Focsa, Thin Solid Films 451
/452 (2004) 328
[Pul84] H.K. Pulker, Coatings on Glass. Elsevier, 1984
[Qin82] H. Qingheng, E.S. Yang, and H. Izmirliyan, Solid-State Electronics 25 (1982)
1187
[Rau04] B. Rau, I. Sieber, J. Schneider, M. Muske, M. Stöger-Pollach, P.
Schattschneider, S. Gall, and W. Fuhs, J. Cryst. Growth 270 (2004) 396
[Ree06] H.S. Reehal and G. Ekanayake, Vacuum 81 (2006) 272
[Rei02] P. Reinig, F. Fenske, W. Fuhs, and B. Selle, J. Non-Cryst. Solids, 299/302
(2002) 128
[Sch05] J. Schneider, PhD thesis (2005)
[Scr78] R. A. Scranton and J. O. McCaldin, J. Vac. Sci. Technol. 15 (1978) 1358
[Sie03] I. Sieber, R. Schneider, I. Doerfel, P. Schubert-Bischoff, S. Gall, and W. Fuhs,
Thin Solid Films 427 (2003) 298
[Sla06] A. Slaoui, E. Pihan, A. Focsa, Sol. Energy Mater. Sol. Cells 90 (2006) 1542
[Sug05] Y. Sugimoto, N. Takata, T. Hirota, K. Ikeda, F. Yoshida, H. Nakashima and H.
Nakashima, Jpn. J. Appl. Phys. 44 (2005) 4770
[Szy99] B. Szyszka, Thin Solid Films 351 (1999) 164
118
[Tho58] D.G. Thomas, J. Phys. Chem. Solids 9 (1958) 31
[Tsa81] B-Y. Tsaur, G.W. Turner and J.C.C Fan, Appl. Phys. Lett. 39 (1981) 749
[Tuz08] Ö. Tüzün, A. Slaoui, I. Gordon, A. Focsa, D. Ballutaud, G. Beaucarne and J.
Poortmans, Thin Solid Films 516 (2008) 6892
[Vos76] J. L. Vossen, Transparent conducting films, J. Vac. Sci. and Tech. 13 (1976)
116
[Wal03] T. Wallendorf, S. Marke, C. May, J. Strümpfel, Surf. Coat. Technol. 174-175
(2003) 222
[Wid02] P. Widenborg and A. G. Aberle, J. Cryst. Growth 242 (2002) 270
[Wid03] P. Widenborg, A.G. Aberle, A. Straub, NP. Harder, DH. Neuhaus and O. Nast.,
International PCT patent application PCT/AU03/01313 (filed on 7 October
2003).
[Wid07] P. Widenborg and A.G. Aberle, J. Cryst. Growth 306 (2007) 177
[Wu00] H.Z. Wu, K.M. He, D.J. Qiu, D.M. Huang, J. Crys. Growth 217 (2000) 131
[Zou05] M. Zou, L. Cai, and W. Brown, Electrochemical and Solid-State Letters, 8
(2005) G103
[Zou07] M. Zou, S. Dorey, D. Cai, Y. Song, R. Premachandran-Nair, L. Cai, and W.
Brown, Electrochemical and Solid-State Letters, 10 (2007) K7
119
[CHAPTER 3]
[Aga04] C. Agashe, O. Kluth, J. Hüpkes, U. Zastrow, B. Rech, and M. Wuttig, J. Appl.
Phys. 95 (2004) 1911
[Blo92] P. Blood and J. W. Orton, The electrical characterization of Semiconductors:
Majority Carriers and Electron States (1992)
[Bo02] X. Bo, N. Yao, S.R. Shieh, T.S. Duffy and J.C. Sturm, J. Appl. Phys. 91 (2002)
2910
[Cou00] T.J. Coutts, D.L. Young, and X. Li, MRS Bull. 25 (2000) 58
[Day01] A. P. Day et al., Channel 5 User Manual, HKL Technology A/S, Hobro,
Denmark (2001)
[Gal02a] S. Gall, J. Schneider, M. Muske, I. Sieber, O. Nast, and W. Fuhs, Proc. of PV
in Europe -From PV Technology to Energy Solutions (2002) 87
[Hul99] R. Hull, Properties of Crystalline Silicon, INSPEC, London (1999)
[Kam88] T. Kamins, Polycrystalline Silicon for Integrated Circuit Applications.
Kluwer: Boston, 1988.
[Klu74] H.P. Klug and L.E. Alexander, X-ray diffraction procedures for polycrystalline
and amorphous materials (2nd edition) (John Wiley & Sons, New York, 1974)
120
[Mar93] P. Marcus, C. Hinnen, and I. Olefjord, Surface and Interface Analysis 20(1993)
923
[Nas00] O. Nast and A. J. Hartmann, J.Appl. Phys. 88 (2000) 716
[Nic00] N. H. Nickel, P. Lengsfeld, and I. Sieber, Phys. rev. B 61 (2000) 15 558
[Nyl94] A. Nylund and I. Olefjord, Surface and interface analysis 21 (1994) 283
[Owe03] J. Owens, D. Han, B. Yan, J. Yang, K. Lord, and S. Guha, NCPV and Solar
Program Review Meeting (2003) 776
[Pau58] L.J. Van der Pauw, Philips Research Reports 13 (1958) 1
[Sch02] J. Schneider, Master thesis, Technische Universität Berlin (2002)
[Sch05] J. Schneider, PhD thesis (2005)
120
[Sch90] D.K. Schroder, Semiconductor material and device characterization (John
Wiley & Sons, New York, 1990)
[Sch96] M. Schuster and H. Göbel, Adv. X-Ray Anal. 39 (1996) 1
[Tro93] K.Z. Troost. P. van der Sluis, and D.J. Gravesteijn, Appl. Phys. Lett. 62 (1993)
1110
[Wat04] H. Watanabe, N. Yamada, and M. Okaji, International J.Thermophys., 25
(2004) 221
[Wu96] X.L. Wu, G.G. Siu, S. Tong, X.N. Liu, F. Yan, S.S. Jiang, X.K. Zhang, and D.
Feng, Appl. Phys. Lett. 69 (1996) 22
[Zho07] W. Zhou and Z.L. Wang, Scanning Microscopy for Nanotechnology (2007)
121
[CHAPTER 4]
[Abe06] A.G. Aberle, J. Cryst. Growth 287 (2006) 386
[Aga04] C. Agashe, O. Kluth, J. Hüpkes, U. Zastrow, B. Rech, and M. Wuttig, J. Appl.
Phys. 95 (2004) 1911
[Bae07] J. W. Bae, S. W. Lee, and G. Y. Yeom, J. Electro. Soc. 154 (2007) 34
[Bel92] J.R. Bellingham, W.A. Phillips, and C.J. Adkins, J. Mat. Sci. Lett. 11 (1992)
263
[Ber06] M. Berginski, B. Rech, J. Hüpkes, G. Schöpe, M.N. van den Donker, W. Reetz,
T. Kilper, and M. Wuttig, 21st European Photovoltaic Solar Energy Conference
(2006)
[Ber08] M. Berginski, PhD thesis (2008)
[Bre03] S. Brehme, F. Fenske,W. Fuhs, E. Neubauer,M. Poschenrieder, B. Selle, and I.
Sieber, Thin Solid Films 342 (1999) 167
[Cha01] J.F. Chang, W.C. Lin, M.H. Hon, Applied Surface Science 183 18 (2001).
[Dim07] D. Dimova-Malinovska, O. Angelov, M. Kamenova, A. Vaseashta and J. C.
Pivin, J. Optoelectronics and Advanced Materials 9 (2007) 355
[Due08] J.N. Duenow, T.A. Gessert, D.M. Wood, A.C. Dillon, and T.J. Coutts, J. Vac.
Sci. Tech. A 26 (2008) 692
[Ell01] K. Ellmer, J. Phys. D : Appl. Phys. 34 (2001) 3097
[Ell05] K. Ellmer and G. Vollweiler, Thin Solid Films 496 (2006) 104
[Ell06] K. Ellmer and G. Vollweiler, Thin Solid Films 496 (2006) 104.
[Ell08] K. Ellmer, R. Mientus, Thin Solid Films 516 (2008) 4620
[Ham05] O. Hamad, G. Braunstein, H. Patil, and N. Dhere, Thin Solid Films 489 (2005)
303
[Iga88] Y. Igasaki, M. Ishikawa, G. Shimaoka, Applied Surface Science 33/34 (1988)
926
[Kaw03] T. Kawashima, H. Matsui, and N. Tanabe, Thin Solid Films 445 (2003) 241
122
[Kim05] K.-K. Kim, S. Niki, J.-Y. Oh, J.-O Song, T.-Y. Seong, S.-J. Park, S. Fujita, and
S.-W. Kim, J. Appl. Phys. 97 (2005) 066103
[Kim05b] K.K. Kim, H. Tampo, J.O. Song, T.Y. Seong, S.J. Park, J.M. Lee, S.W. Kim,
S. Fujita, and S. Niki, Jpn. J. Appl. Phys. 44 (2005) 4776
[Kon03] M. Kon, P.K. Song, Y. Shigesato, P. Frach, S. Ohno, and K. Suzuki, Jpn. J.
Appl. Phys. 42 (2003) 263
[Lec03] R. Lechner, Master thesis (2003)
[Li05] X. Li, B. Keyes, S. Asher, S. B. Zhang, S.H. Wei, and T. J. Coutts, S.
Limpijumnong, and C.G. Van de Walle, Appl. Phys. Lett. 86 (2005) 122107
[Lor03] M. Lorenz, E.M. Kaidashev, H. von Wenckstern, V. Riede, C. Bundesmann, D.
Spemann, G. Benndorf, H. Hochmuth, A. Rahm, H.-C. Semmelhack, and M.
Grundmann, Solid-State Electron. 47 (2003) 2205
[Mas83] G. Masetti, M. Severi and S. Solmi, IEEE Trans. Electron Devices 30 (1983)
764
[Min00] T. Minami, MRS Bulltin 25 (2000) 38
[Min89] T. Minami, H. Sato, H. Nanto, and S. Takata, Thin Solid Films 176 (1989) 277
[Min90] T. Minami, K. Oohashi, S. Takata, T. Mouri, and N. Ogawa, Thin Solid Films
193/194 (1990) 721
[Min92] T. Minami, H. Sato, K. Ohashi, T. Tomofuji, and S. Takata, J. Cryst. Growth
117 (1992) 370
[Min99] T. Minami, T. Miyata, and T. Yamamoto, J. Vac. Sci. Technol. A 17(4) (1999)
1822
[Mue04] J. Müller, B. Rech, J. Springer, and M. Vanecek, Solar Energy 77 (2004) 917
[Myo03] S.Y. Myong and K.S. Lim, Appl. Phys. Lett. 82 (2003) 3026
[Oh05] B. Oh, M. Jeong, D. Kim, W. Lee, and J. Myoung, J. Cryst. Growth 281 (2005)
475
[Rec06] B. Rech, T. Repmann, M.N. van den Donker, M. Berginski, T. Kilper, J.
Hüpkes, S. Calnan, H. Stiebig, and S. Wieder, Thin Solid Films 511/512 (2006)
548
[Suz96] A. Suzuki, T. Matsushita, N. Wada, Y. Sakamoto, and M. Okuda, Jpn. J. Appl.
Phys. 35 (1996) L56
123
[Tho56] D.G. Thomas and J.J. Lander, J. Chem. Phys. 25 (1956) 1136
[Wal03] C. G. Van de Walle and J. Neugebauer, Nature 423 (2003) 626–628
[Yu05] X. Yu, J. Ma, F. Ji, Y. Wang, X. Zhang, and H. Ma, Thin Solid Films 483
(2005) 296
124
[CHAPTER 5]
[Avr39] M. Avrami, J. Phys. Chem., 7 (1939) 1103.
[Avr40] M. Avrami, J. Phys. Chem., 8 (1940) 212.
[Bro81] A. Brokman and R. Balluffi, Acta Metall. 29 (1981) 1703
[Cer72] F. Cerdeira, C. J. Buchenauer, F. H. Pollak and M. Cardona, Phys. Rev. B 5
(1972) 580.
[Cse75] L. Csepregi, J.W. Mayer and T.W. Sigmont, Physics Letter A 54 (1975) 157
[Cse78] L. Csepregi, E.F. Kennedy, J.W. Mayer and T.W. Sigmont, J. Appl. phys. 49
(1978) 3906
[Dim06] D. Dimova-Malinovska, V. Grigorov, M. Nikolaeva-Dimitrova, O. Angelov
and N. Peev, Thin Solid Films 501(1-2) (2006) 358
[Dog08] P. Dogan, E. Rudigier, F. Fenske, K.Y. Lee, B. Gorka, B. Rau, E. Conrad and
S. Gall, Thin Solid Films 516 (2008) 6989
[Gal02b] S. Gall, M. Muske, I. Sieber, O. Nast and W. Fuhs, J. Non-Crystal. Solids
299–302 (2002) 741
[Ges08b] D. Van Gestel, P. Dogan, I. Gordon, H. Bender, K. Y. Lee, G. Beaucarne, S.
Gall and J. Poortmans, Material science & engineering B, in press (2008)
[Gor07] B. Gorka, B. Rau, K.Y. Lee, P. Dogan, F. Fenske, E. Conrad, S. Gall and B.
Rech, 22nd European Photovoltaic Solar Energy Conference (2007) 2024
[Gri07] V. Grigorov, O. Angelov, M. Kamenova and D. Dimova-Malinovska, NATO
Science Series II: Mathematics, Physics and Chemistry 223 (2007) 333
[Hir79] J. Hirvonen and A. Anttila, Appl. Phys. Lett. 35 (1979) 703
[Hsu03] C.M. HSU, I.F. CHEN and M.C. YU, Jpn. J. Appl. Phys. 42 (2003) 4928
[Hul98] R. Hull (Ed.), Properties of Crystalline Silicon—Technology, 1998, pp. 600
[Hwa80] J.C.M. Hwang, P.S. Ho, J.E. Lewis and D.R. Campbell, J. Appl. phys. 51
(1980) 1576
[ICDD] International Center for Diffraction Data (ICDD) database, PDF-card 00-027-
1402
125
[Jen92] AK Jena,MC Chaturvedi (1992). Phase Transformations in Materials. Prentice
Hall. p. 247.
[Joh39] W.A. Johnson and R.F. Mehl, Trans. Amer. Inst. Min. Metall. Eng. 135 (1939)
416.
[Kal79] L. Kalinovski and R. Seguin, Appl. Phys. Lett. 35 (1979) 211
[Kam98] T.I. Kamins, Polycrystalline Silicon for Integrated Circuit and Displays, 2nd
ed., Kluwer Academic Publishers (1998)
[Kim02] H. Kim, G. Lee, D. Kim and S.-H. Lee, Curr. Appl. Phys. 2 (2002) 129
[Kis03] R. Kishore, C. Hotz, H. A. Naseem and W. D. Brown, Applied Crystallography
36(5) (2003) 1236
[Kit02] K. Kitahara, R. Yamazaki, T. Kurosawa, K. Nakajima, and A. Moritani, Jpn. J.
Appl. Phys. 41 (2002) 5055
[Kon92] T. J. Konno and R. Sinclair, Philos Mag-B 66 (1992) 749
[Kon95] T.J. Konno and R. Sinclair, Philos. Mag. B 71 (1995) 163
[Lee97] J.N. Lee, B.J. Lee, D.G. Moon and B.T. Ahn, Jpn. J. Appl. phys. 36 (1997) 686
[Nas00b] O. Nast, Conference Record of the Twenty-Eighth IEEE Photovoltaic
Specialists Conference (2000)
[Man78] G. Manjni and G. Ottaviani, J. Cryst. Growth 45 (1978) 132
[McC71] J. McCaldin and H. Sankur, Appl. Phys. Lett. 19 (1971) 524
[Nas00] O. Nast, PhD thesis, (2000)
[Nas00b] O. Nast and A. Hartmann, J. Appl. Phys. 88 (2000) 716
[Nas00S] O. Nast and S. R. Wenham, J. Appl. Phys. 88 (2000) 124
[Nas01] O. Nast, S. Brehme, S. Pritchard, A.G. Aberle and S.R. Wenham, Sol. Energy
Mater. Sol. Cells 65 (2001) 385
[Pih07] E. Pihan, A. Slaoui and C. Maurice, J. Cryst. Growth 305 (2007) 88
[Rau06] B. Rau, E. Conrad and S. Gall, Twentyfirst European Photovoltaic Solar
Energy Conference (2006)
[Rob04] J.D. Robson, Acta Mater. 52 (2004) 4669
126
[Sch04] J. Schneider, J. Klein, M. Muske, S. Gall and W. Fuhs, Journal of Non-
Crystalline Solids 338–340 (2004) 127
[Sch05] J. Schneider, PhD thesis, (2005)
[Sch06] J. Schneider, A. Schneider, A. Sarikov, J. Klein, M. Muske, S. Gall and W.
Fuhs, Journal of Non-Crystalline Solids 352 (2006) 972
[Sch98] R. E. I. Schropp and M. Zeman, Amorphous and Microcrystalline Silicon Solar
Cells: Modelling, Materials and Device technology (Kluwer Acadamic, Boston,
1998)
[Sec72] F. Secco d'Aragona, J. Electrochem. Soc. 119 (1972) 948
[Spi98] C. Spinella, S. Lombardo and F. Priolo, J. Appl. phys. 84 (1998) 5383
[Str04] A. Straub, P.I. Widenborg, A.B. Sproul, Y. Huang, N.-P. Harder and A.G.
Aberle, J. Cryst. Growth 265 (2004) 168
[Ter07] M.L. Terry, D. Inns and A.G. Aberle, Advances in OptoElectronics 2007
(2007) 11 Article ID 83657
[Vou95] A.T. Voutsas, M.K. Hatalis, J. Boyce and A. Chiang, J. Appl. Phys. 78 (12)
(1995) 6999
[Wid02] P. Widenborg and A.G. Aberle, J. Cryst. Growth 242 (2002) 270
[Wid05] P.I. Widenborg, A. Straub and A.G. Aberle, J. Cryst. Growth 276 (2005) 19
[Wid07] P.I. Widenborg and A.G. Aberle, J. Cryst. Growth 306 (2007) 177
[Yu97] G. Yu, Y.K.L. Lai and W. Zhang, J. Appl. Phys. 82 (1997) 4270
127
[CHAPTER 6]
[Abe06] A.G. Aberle, J. Cryst. Growth 287 (2006) 386
[Ber06] M. Berginski, B. Rech, J. Hüpkes, G. Schöpe, M.N. van den Donker, W. Reetz,
T. Kilper, and M. Wuttig, 21st European Photovoltaic Solar Energy Conference
(2006)
[Gal08] S. Gall, C. Becker, E. Conrad, P. Dogan, F. Fenske, B. Gorka, K.Y. Lee, B.
Rau, F. Ruske and B. Rech, Sol. Energy Mater. Sol. Cells, in press (2008)
[Gor07a] I. Gordon, D. Van Gestel, L. Carnel, G. Beaucarne, J. Poortmans, K.Y. Lee, P.
Dogan, B. Gorka, C. Becker, F. Fenske, B. Rau, S. Gall, B. Rech, J. Plentz, F.
Falk and D. Le Bellac, 22nd European Photovoltaic Solar Energy Conference
(2007) 1890
[Gor07b] I. Gordon, L. Carnel, D. Van Gestel, G. Beaucarne and J. Poortmans, paper
accepted for publication in Progress in Photovoltaics (2007) 765
[Kor07] L. Korte, E. Conrad, H. Angermann, R. Stangl and M. Schmidt, 22nd European
Photovoltaic Solar Energy Conference (2007) 859
[Pih07] E. Pihan, A. Slaoui and C. Maurice, J. Cryst. Growth 305 (2007) 88
[Rau08] B. Rau, T. Weber, B. Gorka, P. Dogan, F. Fenske, K.Y. Lee, S. Gall and B.
Rech, Material science & engineering B, in press (2008)
[Rec06] B. Rech, T. Repmann, M.N. van den Donker, M. Berginski, T. Kilper, J.
Hüpkes, S. Calnan, H. Stiebig, and S. Wieder, Thin Solid Films 511/512 (2006)
548
128
129
L
IST OF PUBLICATIONS
K.Y. Lee, P. Dogan, F. Ruske, B. Gorka, J. Hüpkes, S. Gall and B. Rech; "The
properties of poly-Si films grown on ZnO:Al coated borofloat glass by aluminium-
induced layer exchange (ALILE) process", Proceeding of 23rd European
Photovoltaic Solar Energy Conference: 2008 p. 2261-2265
K.Y. Lee, M. Muske, I. Gordon, M. Berginski, J. D'Haen, J. Hüpkes, S. Gall, B.
Rech: “Large-grained poly-Si films on ZnO:Al coated glass substrates”. Thin Solid
Films 516 (20) (2008), p. 6869-6872
P. Dogan, E. Rudigier, F. Fenske, K.Y. Lee, B. Gorka, B. Rau, E. Conrad, S. Gall:
“Structural and electrical properties of epitaxial Si layers prepared by E-beam
evaporation”. in press, Thin Solid Films 516 (20) (2008), p. 6989-6993
B. Rau, T. Weber, B. Gorka, P. Dogan, F. Fenske, K.Y. Lee, S. Gall, B. Rech ;
“Development of a rapid thermal annealing process for polycrystalline silicon thin-
film solar cells on glass” : in press, Materials Science and Engineering: B (2008)
D. Van Gestel, P. Dogan, I. Gordon, H. Bender, K.Y. Lee, G. Beaucarne, S. Gall,
J. Poortmans ; “Investigation of intragrain defects in pc-Si layers obtained by
aluminum-induced crystallization: comparison of layers made by low and high
temperature epitaxy”, in press, Thin Solid Film (2008)
K.Y. Lee, C. Becker, M. Muske, F. Ruske, S. Gall, B. Rech, M. Berginski, and J.
Hüpkes: “Temperature stability of ZnO:Al film properties for poly-Si thin-film
devices”. Applied Physics Letter 91 (2007), p. 241911/1-3
S. Gall, C. Becker, E. Conrad, P. Dogan, F. Fenske, B. Gorka, K.Y. Lee, B. Rau, F.
Ruske, and B. Rech: “Polycrystalline silicon thin-film solar cells on glass”.
Technical digest / 17th International Photovoltaic Science and Engineering
Conference (PVSEC-17) : 2007, p. 343-344
C. Becker, T. Hänel, B. Gorka, K.Y. Lee, P. Dogan, F. Fenske, M. Berginski, J.
Hüpkes, S. Gall, and B. Rech: “Solid phase crystallization of amorphous silicon on
ZnO:Al for thin film solar cells”. Technical digest / 17th International Photovoltaic
Science and Engineering Conference (PVSEC-17) : 2007, p. 1122-1123
K.Y. Lee, C. Becker, M. Muske, S. Gall, B. Rech, I. Gordon, J. D'Haen, M.
Berginski, and J. Hüpkes: “Poly-Si films grown on ZnO:Al coated glass by
aluminum-induced layer exchange process”. Proceedings of 22nd European
Photovoltaic Solar Energy Conference : 2007, p. 2028-2031
S. Gall, K.Y. Lee, P. Dogan, B. Gorka, C. Becker, F. Fenske, B. Rau, E. Conrad,
B. Rech,: “Large-grained polycrystalline silicon thin-film solar cells on glass”.
Proceedings of 22nd European Photovoltaic Solar Energy Conference : 2007, p.
2005-2009
130
I. Gordon, D. Van Gestel, L. Carnel, G. Beaucarne, J. Poortmans, K.Y. Lee, P.
Dogan, B. Gorka, C. Becker, F. Fenske, B. Rau, S. Gall, B. Rech, J. Plentz, F. Falk,
and D. Le Bellac: “Advanced concepts for thin-film polycrystalline-silicon solar
cells”. Proceedings of 22nd European Photovoltaic Solar Energy Conference :
2007, p. 1890-1894
B. Gorka, B. Rau, K.Y. Lee, P. Dogan, F. Fenske, E. Conrad, S. Gall, and B. Rech:
“Hydrogen passivation of polycrystalline Si thin films by plasma treatment”.
Proceedings of 22nd European Photovoltaic Solar Energy Conference : 2007, p.
2024-2027
P. Dogan, F. Fenske, L.-P. Scheller, K.Y. Lee, B. Gorka, B. Rau, E. Conrad, S.
Gall, and B. Rech: “Structural and electrical properties of epitaxial Si layers
prepared by e-beam evaporation”. Proceedings of 22nd European Photovoltaic
Solar Energy Conference : 2007, p. 2019-2023
B. Rau, K.Y. Lee, P. Dogan, F. Fenske, E. Conrad, and S. Gall: “Improvement of
epitaxially grown poly-Si thin-film solar cells on glass by rapid thermal annealing”.
Proceedings of 22nd European Photovoltaic Solar Energy Conference : 2007, p.
2010-2014
C
ONFERENCES
K.Y. Lee, P. Dogan, F. Ruske, B. Gorka, J. Hüpkes, S. Gall and B. Rech; "The
properties of poly-Si films grown on ZnO:Al coated borofloat glass by aluminium-
induced layer exchange (ALILE) process", 23rd European Photovoltaic Solar
Energy Conference , Valencia, Spain, 1-5 September (2008)
K.Y. Lee, C. Becker, M. Muske, S. Gall, B. Rech, I. Gordon, J. D'Haen, M.
Berginski, and J. Hüpkes: “Poly-Si films grown on ZnO:Al coated glass by
aluminum-induced layer exchange process”. 22nd European Photovoltaic Solar
Energy Conference Milan, Italy, 03.09.2007 - 07.09.2007 (2007)
K.Y. Lee, M. Muske, I. Gordon, M. Berginski, D'Haen, J. Hüpkes, S. Gall, and B.
Rech: “Large-grained poly-Si films on ZnO:Al coated glass substrates”. E-MRS
SPRING MEETING 2007 Strasbourg, France, 28.05.2007 - 01.06.2007 (2007)
131
A
CKNOWLEDGEMENT
I would like to thank Prof. Dr. B. Rech for his fantastic support of my study from idea
to finish.
I would like to thank Dr. Stefan Gall for supervising my work and being supportive and
giving me all these opportunities to present my results to the scientific community and
also accepting my application which I could start this work at HZB on an almost blind
date.
I would like to thank Dr. Florian Ruske as a TCO professionalist for all the on and off
topic discussions and for being the first proof reader of this thesis.
I would like to thank Dr. Christiane Becker for discussion of results and being the first
proof reader of this thesis.
I would like to thank Dr. Björn Rau for the RTA treatments.
I would like to thank Pinar Dogan for growing absorber layers.
I would like to thank Benjamin Gorka for working on hydrogen passivation treatments.
I would like to thank Martin Muske as a room mate and a best friend for working
together in order to make tons of samples.
I would like to thank Stefan Common for all the hilarious times, especially coffee time
on afternoon.
I would like to thank Kerstin Jacob and Anja Scheu for tons of sample and substrate
treatments.
I would like to thank Carola Klimm for performing SEM measurements.
132
I would like to thank Marion Krusche and Dr. Gabriele Lampert for helping me with
administrative matters, which allowed me to focus on my scientific work.
I would like to thank all the other members of silicon thin film solar cell group for a
pleasant work atmosphere.
I would like to thank all the other members of E-I1 in HZB for a pleasant work
atmosphere.
I would like to thank Dr. Ivan Gordon for EBSD measurements and discussions.
I would like to thank Dr. Jürgen Hüpkes and Dr. Michael Berginski for preparing lots of
ZnO:Al samples.
I would like to thank my wife and my children for filling with joy and happiness.
