Final_Doc.

Final_Doc.

Cells by

Oyediran David Oludare

Of

Master of Science

In

2009

Growth of Thin Film Microcrystalline Silicon Solar Cells

Oyediran David Oludare

DELFT UNIVERSITY OF TECHNOLOGY

FACULTY OF ELECTRICAL ENGINEERING, MATHEMATICS AND

COMPUTER SCIENCE.

MSc Thesis Committee

Prof. Dr. M. Zeman (Full Professor, Electrical Sustainable Energy)

Dr. R.A.C.M.M. Van Swaaij (Associate Professor, Electrical Sustainable Energy)

Dr. ir. Martin D. Verweij (Associate Professor, Telecommunication Department)

Dr. Sergiy Dobrovolsky (Post Doc. Electrical Sustainable Energy)

This project was done under the supervission of Dr. R.A.C.M.M. Van Swaaij, Associate

Professor, Department of Electrical Sustainable Energy, Laboratory of Photovoltaic

Materials and Devices, Delft University of Technology – TUDelft.

Copyright © 2009 the Laboratory of Photovoltaic Materials and Devices.

All rights reserved. ii

Dedication

To God Almighty, the giver of life, wisdom and strength,

........ to my wife Ogechukwu,

........ my son Light of God,

........ my daughter the Delight of God,

....... for your love, patience and understanding.

iii

ACKNOWLEDGEMENTS

I wish to appreciate the God of heaven and the earth for His unfailling love and mercies in my life. He created the opportunity for me to be in TUDelft and also granted the grace for successful completion. Though the road was long and rough but He was there with me from the beginning to the end. I will forever remain thankful to you.

My heart felt appreciation goes to my project supervisor, Dr. René Van Swaaij, for his constant and unconditional availability throughout this research. I am very grateful for the time invested in reading through the manuscript and for your guidiance and help. I also appreciate your wealth of knowledge and authority in semiconductor physics.

The chair of the Laboratory of Photovoltaic Materials and Devices, Prof. dr. M. Zeman is specially appreciated for accepting me and the opportunity to do my thesis in the group and for his continous encouragement. I appreciate your simplicity and leadership style. I also thank Dr. Ir. Martin D.Verweij and Dr. Sergiy Dobrovolskiy for reading through my work and for accepting to be a members of the thesis committee.

My daily supervisor, S.N Agbo deserves my full appreciation. I thank you for providing the samples used for this study and also for your guidiance and useful discussions in helping me to have better understanding of this research. My special thanks to Kasper

Zwetsloot for teaching me how to use the measurement setups and Martijn Tijssen for keeping the deposition setup running always. Thanks also for your prompt responses and technical assistance. My sincere appreciation goes to all the Post Docs. and PhD students in the PVMD group for their help and support. I thank Ir. Micheal Wank for the assistance on FTIR analysis. All the MSc students are equally thanked, Natalie for useful discussion and many trials on the possibility of determinning R

*

from the Raman spectra,

Stern for your assistance in software related issues and for quick Dutch translations.

Folkert and Chimdi (CDF group), Aster, Arash, Kehinde, Dr. R. Liang, thanks for being there always. iv

I was opportuned to have a close contact with my teacher in design challenge, Dr Marnix

Ten Kortenaar, a Dutch and practising Christian. I was amazed to know that people like you exist in a country believed to be far from God. Thanks for your encouragement and kindness.

I will also like to specially thank my friend and colleague, Sam Ofordile and family for your loving kindness and constant encouragement. Thanks for helping me to come this far. Pastor and Pastor Mrs. Nwosu and family, Deacon Alex Dawotola and family are deeply appreciated for your prayers. Thanks to Engr. and Mrs L.M Obi for your kind interest and favour towards me and my family.

Studying in a great University like the TUDelft would have been a mirage but for the financial support from the government of The Netherlands (NFP). Thanks so much for this wounderful and life changing opportunity. I thank the management of TUDelft and

University of Nigeria, Nsukka for the collaboration that provided this openning. Prof.

A.N Nzeako and Sandra Irobi are specially acknowledged for keeping the ball of the collaboration rolling.

Finally, i thank my parents Mr Oyediran Emmanuel Adegboyega and Mrs Comfort

Oyelola for your parental love and care. To all of you who have made so much contributions to my life i say thank you all and may the blessings of our father in heaven abide with you forever, amen. v

TABLE OF CONTENTS

Dedication iii

List of figures viii

Abstract xi

1. INTRODUCTION

1

1.1 Solar energy

1.2 Photovoltaic technology

2

3

1.3 Thin-film silicon solar cells

1.3.1 Amorphous silicon

1.3.2 Microcrystalline silicon

1.4 Basic solar cell configurations (pin and nip)

1.4.1 Transparent conductive oxide (front contact)

1.4.2 The back reflectors

6

7

8

9

9

9

1.4.3 The p-layer and its properties

1.4.4 The i-layer and its properties

1.4.5 The n-layer of a solar cell

1.5 Micromorph tandem silicon solar cells

1.6 Sensitivity of deposition parameters for microcrystalline solar cell

1.7 Purpose of the project

1.8 Project Organisation

2. MEASUREMENT AND CHARACTERIZATION

2.0 Introduction

2.1 Deposition of microcrystalline silicon

2.1.1 The AMOR deposition system

2.2 Material and Device characterization

2.2.1. Electrical characterization

19

19

21

22

17

19

10

11

12

12

14

17

2.2.1.1. Dark and photoconductivity measurement

2.2.2. Optical Properties Estimation

2.2.2.1 Fourier Transform Photocurrent Spectrocopy

2.2.2.2 Measurement of the sub-gap absorption coefficient

2.2.2.3 Reflection and Transmission Measurements

2.2.2.4 The mini RT measurement setup

2.2.3 Structural properties characterization

2.2.3.1 Fourier Transform Infrared Spectrometer

2.2.3.2 Raman Spectroscopy

2.2.3.2.1Evaluation of crystallinity fraction

2.2.4 Device Characterization

2.2.4.1 Measurement of external parameters of solar cells

2.2.4.2 Quantum efficiency measurement (QEM)

28

29

29

31

32

32

34

22

22

23

24

26

27 vi

3. INFLUENCE OF DEPOSITION PARAMETERS ON µc-Si:H PLAYER

3.0 Abstract

3.1 Introduction

3.2 Experimental Details

3.3 Results and Discussion

3.3.1 Substrate material (TCO)

3.3.2 Microcrystalline P-layer

36

36

36

37

38

38

39

3.4 Application to solar cells

3.5 Effect of µc-Si:H p-layer deposition power on the performance of µc-

Si:H solar cell

3.6 Conclusions

4. SENSITIVITY STUDY OF µc-Si:H I-LAYER

4.0 Abstract

4.1 Introduction

4.2 Experimental Details

4.3 Results and Discussion

4.3.1 Electrical Properties

4.3.2 Material Properties

4.3.3 Absorption Coefficient and Defect Density

4.3.4 Hydrogen Content and Microstructure

4.4 Application to solar cells

4.5 Effect of µc-Si:H i-layer thickness reduction on hydrogen content and

microstructure

4.6 Conclusions

5. CONCLUSIONS AND RECOMMENDATION

References

64

66

68

69

70

56

57

57

58

61

63

46

48

54

55

55

55 vii

LIST OF FIGURES

Figure 1.1 Solar radiation spectrum

Figure 1.2 Structure and working principles of solar cell

Figure 1.3 Typical structure of thin-film silicon solar cell

Figure 1.4 Typical p-i-n structure for µ c-Si:H solar cell

Figure 1.5 The “micromorph” tandem cell, a-Si:H top and a bottom cell

Figure1. 6 Semi-empirical upper limit of the efficiency as a function of the

energy gap E g

of the bottom and top cells of a tandem solar cell

based on an empirical minimum value (according to Green) for the

reverse saturation current of the diode [12]

Figure 2.1 Schematic diagram of the PECVD reactor for deposition of

microcrystalline silicon [26]

Figure 2.2 Top view of AMOR deposition system

Figure 2.3 Fourier transform photocurrent spectrocopy setup

Figure 2.4 Absorption spectra of microcrystalline, amorphous and

crystalline silicon measured by CPM [45]

Figure 2.5 Schematic diagram of the Lambda system

Figure 2.6 I-V characteristics of a p-n junction in the dark and under illumination

Figure 2.7 Quantum efficiency measurement setup

27

28

33

35

20

21

24

8

13

14

3

5

6

Figure 3.1 Transmittances of etched and flat ZnO:Al and ZnO:Al with P-layers

Figure 3.2 Transmittance of different thicknesses for varying diborane flow

Figure 3.3 Transmittance in p – layers at different diborane flow

Figure 3.4 Crystallinity volume fraction, conductivity and activation energy

against diborane flow

Figure 3.5 Crystallinity volume fraction, conductivity and activation energy

against deposition time

Figure 3.6 Thickness and deposition rate of µc-Si p-layer at different deposition

time and doping

Figure 3.7 Spectra response of solar cells as a function of p-layer diborane

concentration

Figure 3.8 µc-Si:H solar cells external parameters versus diborane

concentrations of p-layers at different deposition time

Figure 3.9 Transmittance in µc-Si:H p-layer deposited at varying deposition

power

Figure 3.10 Thickness series of µc-Si:H p-layer deposited at 0.2 sccm diborane

flow and 300 seconds deposition time.

Figure 3.11 µc-Si:H p-layer structural properties dependence on deposition

power

Figure 3.12 Spectra response of solar cells deposited at a fixed diborane flow

(0.2 sccm) and deposition time (300 seconds)

Figure 3.13 µc-Si:H solar cells external parameters versus power at 0.2 sccm

diborane flow and 300 s deposition time.

51

52

49

50

53

45

46

47

38

40

41

43

44

Figure 4.2 Photoresponse of µc-Si:H i-layer at different deposition power

Figure 4.3 Crystallinity of µc-Si:H i-layer at different deposition power

Figure 4.4 Silane concentrations at different deposition power and pressure

58

59

60 viii

Figure 4.5 Deposition rate and thickness series of µc-Si:H i-layer versus deposition power and pressure

Figure 4.6 Optical absorption of µc-Si:H i-layer

Figure 4.7 Spectra response of solar cells as a function of i-layer deposition

power (W), pressure (mba) and silane concentration (SC) in

percentage

Figure 4.8 Photoresponse of µc-Si:H i-layer at 60 Watt deposition power

60

63

66

67 ix

LIST OF TABLES

Table 4.1

Solar cells parameters obtained from the best cells

Table 4.2 A verage of best ten cells

Table 4.3 R

*

and c

H

values at 60 Watt deposition power

63

63

65 x

ABSTRACT

The plasma-enhanced chemical vapour deposition (PECVD) method is widely used compared to other methods to deposit µc-Si:H because of the high potential to prepare high quality material uniformily on a large area substrate at low temperature. This method was used to grow µc-Si:H p- and i-layers. The effect of p– layer deposition parameters on the short– wavelength response of µc-Si:H solar cells is investigated.

We also investigated the influence of deposition parameters on the properties of the µc-Si:H absorber layer deposited at the a-Si:H/ µc-Si:H transition. Parameters such as

RF power, silane concentration, and deposition pressure were studied. The effect of these parameters on the material properties of intrinsic µc-Si:H layers and the device performance of single junction µc-Si:H solar cells is presented.

The results show that p-layer deposited at 300 seconds with 0.2 sccm diborane flow has the optimum value with respect to transparent and conductive nature. It gave a high FF and V oc when applied in a single junction p-i-n type µc-Si:H solar cells with efficiency of 5.4%. Significant gain in quantum efficiency of the solar cell was observed especially in the short-wavelength region. With the optimized p-layer and at 80 W deposition power, the quantum efficiency increased to about 65% at 400 nm when compared to the obtained value of about 35% with the same optimized p-layer deposited at 60 W. The overall results show that the spectral response is highly sensitive to diborane flow at short wavelength.

The result of i-layer sensitivity study reveals µc-Si:H i-layer deposited at a low power but higher pressure has high photoresponse. The structural properties of these layers shows defects which may be related to the grain boundries and material contamination due to stress. This was evident as the film oxidizes immediately it is brought out of the deposition system for FTIR analysis, leaving the substrate with little or no films. xi

Chapter One

1.0 Introduction

There is demand for new sources of energy all over the world because of numerous enviromental degradation caused by the excessive releases of carbon dioxide to the atmosphere.

This carbon dioxide is the by product of fossil fuel, a major supplier of world energy. There are three main types of fossil fuels, namely, oil, coal and natural gas. After food, the next source of energy for human activities is fossil fuel. It is needed for the generation of electricity, heating, transport and running of machineries.

However, apart from the enviromental concern associated to these sources of energy supply, there is already global shortages in supply. Increasing world population is another factor contributing to the shortage in energy supply. This reqiures urgent steps to finding a way of conserving this non-renewable energy sources. Over the last 25 years, there has been various predictions about the depleting global crude oil reserve. According to figures, oil should have run out by now [1]. Though different strategies like energy-saving measures are being employed today at prolonging the lifespan of these sources but one day, the world will have to look for a replacement when eventually these sources are no more available. In the light of these challenges and the need to have a more sustainable and enviromentally-friendly energy sources, renewable energy options are imperative. Such renewable energy sources as solar energy, wind power, geothermal, biomas, tidal power, wave power, hydro power and biofuels.

1

Solar energy is one of the renewable sources of power. With its abundance and accessibility, it can enormously have potential to supply electricity to all people. Aside from the developed countries, the use of solar energy can be perfect to the developing countries especially to those located in the equatorial regions. As developing countries, it is more advisable to begin to introduce the use of renewable sources of electricity than spending so much on fossil fuel. In issues of the environment, solar energy is favorable since it does not emit harmful greenhouse gases that may deplete the ozone layer. There are no waste products and there are no gases emitted that can be harmful to the animals or people. Solar energy is considered the cleanest source of electricity. Solar radiation represents such an infinite source of energy for the Earth.

1.1 Solar energy

The term solar energy means energy from the sun in form of heat and light. It refers to the utilization of the radiant energy from the sun. The Sun delivers 1.2 × 10

14

kW (i.e 4.32 × 10

20

J/h) energy to the earth, which is about 10,000 times more than the present energy consumption

[2]. The energy that the Earth receives from the Sun in just one hour is equal to the total amount of energy consumed by humans in one year [2]. About 30% of the radiated light is reflected back to the space while the rest are absorbed by cloud, ocean and land masses [3].

Solar radiation spectrum (figure 1.1), describes the visible and near-visible radiation emitted from the sun. Their wavelength range within the broadband range of 0.20 to 4.0 µm describes the different regions.

Infrared radiation emitted from the atmosphere is termed as terrestrial radiation. Components of solar and terrestrial radiation and their approximate wavelength ranges are shown as follows:

Ultraviolet: 0.20 - 0.39 µm

Visible: 0.39 - 0.78 µm

Near-Infrared: 0.78 - 4.00 µm

Infrared: 4.00 - 100.00 µ m

2

Approximately 99% of solar, or short wave, radiation at the earth's surface is contained in the region from 0.3 to 3.0 µm while most of terrestrial, or long-wave, radiation is contained in the region from 3.5 to 50 µm. Figure 1.1 below shows solar radiation spectrum.

Figure 1.1 Solar radiation spectrum

The distance the sunlight has to travel through the atmosphere under clear sky condition is very important because it determines the solar irradiance. This distance is the shortest when the sun is at the zenith, i.e. directly overhead. The ratio of an actual path length of the sunlight to this minimal distance is known as the optical air mass. When the sun is at its zenith, the optical air mass is unity and the radiation is described as air mass one (AM1) radiation. AM1.5 is the standard value to measure solar cell performances [2]. The energy received from the sun in form of light (photons) is then converted into electricity by the process called photovoltaic technology

(PV). Details of this process is discussed in the next section.

1.2 Photovoltaic technology

Photovoltaic is the direct conversion of solar radiation into electricity. Photovoltaics (PV) literally means ‘light-electricity’ and this technology is possible with advanced semiconductor devices and the main areas of applications are terrestrial and space [4]. Photovoltaic as an energy technology has numerous environmental benefits. As domestic source of electricity, it

3

contributes to the nation’s energy security. It is reliable and needs little maintenance. However, it’s major drawback is the high cost when compared to other electricity sources [5].

Some materials exhibit a property known as the photoelectric effect that causes them to absorb photons of light and release charge carriers (electrons and holes). When these free electrons are collected, electric current results. Solar cells are built using this effect. Incident photons, which enter the lattice of solar cell structure, have a not null probability to create an electron-hole pair due to their energy. In fact, if their energy (hν) is greater than the energy band-gap (E g

) of the material used in the solar cell (h is the Planck constant and the frequency is v = c/λ with c the speed of light and λ the wavelength), then they have enough energy to break covalent bonds that tie valence electrons to the nucleus of the single atom. Furthermore, this break results in the movement of electron over the lattice in the covalence band of energy [4, 5]. Basically, photovoltaic effect consist of three processes namely, generation of charge carriers due to the absorption of photons in the materials that form a junction, separation of the photo-generated charge carriers in the junction and the collection of the photo-generated charge carriers at the terminals of the junction. Figure 1.2 below shows the structure and working principles of solar cells. When photons strike a photovoltaic cell, they may be reflected, pass right through, or be absorbed depending on the energy, band-gap and the wavelength of this photon. However, only the absorbed photons provide energy to generate electricity.

When a semiconductor material is exposed to sunlight, energy is absorbed and this leads to the creation of electron-hole pairs. The negatively charged electrons are diplaced from their parent atom and are collected at the front contact with holes left behind. These results in imbalance of charge between the material's front and back contact hence, creating a voltage potential. When the contacts are connected to external load, then electricity flows [6]. However, because of the limitation in output voltages, approximately 0.5 and currents to approximately 7 A of individual cells [4], the overall output power is low and may not support adequately the intended load. In order to increase the output power, cells are connected together in series and parallel to obtain higher voltages and currents in a packaged weather-tight module. The photovoltaic cell efficiency is higher than the module efficiency unless the cells are identical electrically.

Moreover, modules can further be connected to form an array. Array refers to the entire generating plant, irrespective of the number of integrated modules, which is largely dependent on the amount of power output needed.

4

Figure 1.2 Structure and working principles of solar cell

[http://micro.magnet.fsu.edu/primer/java/solarcell]

The first practical use of solar cells was the generation of electricity on the orbiting satellite

Vanguard 1 in 1958, made from single crystal silicon wafers. This solar cell had efficiency of 6%

[2]. Because of search for alternative energy sources, this solar cell also became the first solar cells to be used for terrestrial generation of electricity. This eventually resulted in efforts at increasing its efficiency and lowering their price. The crystalline silicon solar cell technology has improved and it is the dominant solar cell technology today.

Over the last few years, the PV industry has experienced a strong growth, which is expected to continue. As this technology has matured, cost has become increasingly dominated by material costs, namely those of the silicon wafer, the glass cover sheet and encapsulants. Crystalline silicon (c-Si) technology presently dominate the phovoltaic industry contributing more than 95 % through the cells and modules based on mono- and multicrystalline wafer technology. However, the contionous increase in demand has made cost reduction a major challenging. Currently research effort is towards reducing the material cost by developing thin film technology.

5

1.3 Thin-film silicon solar cells

Thin-film silicon based solar cells have the potential of a sustantial cost reduction when compared with bulk wafer based silicon solar cells. Thin films for solar cells can be deposited by a variety of processes that lead to different deposition rates and widely varying material quality.

Thin film technology generally provides high production capacity at reduced material consumption and energy input in the fabrication process and integrated module structures with deposition processes which normally are suitable for mass production. The practical efficiency of different thin films becomes the crucial determinant of their competitive potential. Typical thin film solar cells example as in figure 1.3 are only a few microns thick compared to hundreds of microns for a wafer based cell. Thin film cells consist of a light absorbing silicon layer sandwiched between thin conductive films serving as electrodes on a glass substrate. One electrode will be transparent and serves as the front contact while the second electrode usually consists of metals for the back contact.

Since silicon is abundantly available, thin-film silicon solar cells have the advantage of constituting an industrially mature technology. However, researchers are trying to improve their efficiencies and stability. It is expected that the efficiency of this class of solar cells will likely reach 15 % as research progresses in this field.

Figure 1.3 Typical structure of thin-film silicon solar cell

6

Thin-film silicon material in the form of hydrogenated amorphous silicon (a-Si:H) and hydrogenated microcrystalline silicon (µc-Si:H) are commonly used in thin-film silicon solar cells.

1.3.1 Amorphous silicon

Thin film solar cells based on hydrogenated amourphous silicon have received adequate attention and have emerged as a mature technology for large scale photovoltaic energy conversion. However, the efficiencies achieved in production are still significantly lower than for solar modules based on mono- or polycrystalline silicon wafers [8,9].

Silicon is a four-fold coordinated atom that is normally bonded to four neighboring silicon atoms in a tetrahedron shape. In crystalline silicon, this bond structure is continued over a large range, forming a well-ordered lattice (crystal). In amorphous silicon, this long-range order is not present and the atoms form a continuous random network. Not all the atoms within amorphous silicon are four-fold coordinated. Due to the disordered nature of the material, some atoms have a dangling bond. These are defects in the continuous random network, which cause undesired

(electrical) behavior. The material can be passivated by hydrogen, which bonds to the dangling bonds and can reduce the dangling bonds density by several orders of magnitude.

Thin film silicon can be made by different methods one of which is by using the process of radio frequency plasma-enhanced chemical vapor deposition (RF-PECVD) with silicon radicals plus hydrogen as making up the growing film. This is the most commonly used deposition method to produce “device quality’’ a-Si:H, both on the laboratory and industrial scale. The advantages of this method include: low deposition temperature (100

0 c < T s

< 400

0 c), large area deposition (more than 1 m

2

), effective p- and n-type dopping and alloying, deposition of multilayer structures by control of gas mixtures in a continuous process, easy patterning and integration technology, good mass-producibility and low cost [2,9]. The rate of deposition however is low resulting in long deposition time but this could be reduced by the use of newer deposition techniques, such as VHF-PECVD [2].

It is actually very easy to dope a-Si:H layers, therefore the building of actual semiconductor devices with a-Si:H layers is physically achievable but not advisable using p-n-type diodes as solar cell structure but instead the p-i-n diode the reason for this will be discussed in the subsequent section.

7

1.3.2 Microcrystalline silicon

These layers are deposited under high hydrogen dilution of silane. The crystallinity of the deposited layer depends on the silane concentration and on other deposition conditions.

Microcrystalline silicon material is reported to be quite complex, consisting of an amourphous matrix with embeded crystallites plus grain boundaries [9,10].

Figure 1.4 Typical p-i-n structure for µ c-Si:H solar cell

Its optical properties have a marked crystalline characterristics with an optical gap at 1.12 eV like c-Si [11,13,16]. More information on this and the benefits arising from it is discussed in detail in section 1.5.

Thin film microcrystalline silicon has the advantage that it shows less degradation on prolonged illumination when compared to amourphous [2]. Another interesting characteritic of

µc-Si:H solar cell is the ability to absorb photons within the long wavelength range when compared to a-Si:H solar cells. These are the main reasons why it is receiving research attention.

8

Combinning the two solar cells (tandem arrangements) will ensure effective harvesting of light required for energy conversion and generation within the solar cells.

The p-i-n-type configuration just like in the a-Si:H is also used here because of the low diffusion lengths of minority carriers observed in these two forms of silicon used for constituting these types of solar cell. This configuration type ensures drift assisted collection of the electronhole pairs. The diagram above shows a typical p-i-n thin-film silicon solar cell as used in both amorphous silicon and the microcrystalline silicon with light rays falling on glass substrates through a transparent conductive oxide (ZnO or SnO

2

) down to the back reflector.

1.4 Basic solar cell configurations (pin and nip)

The two basic configurations are the p-i-n or superstrate configuration and the n-i-p or substrate configuration. The major difference between the two is the choice of substrates, in the p-i-n type, the substrates is glass and it serves as the topmost layer through which the light passes but in the n-i-p type, the substrates is plastic or stainless steel which serves as the back layer. The superstrate technology is mostly used in the production of thin-film silicon solar cell [2, 49]. In both configurations, the solar cell consists of the absorber layer sandwiched in between the two doped layers with contacts for carrier extraction.

1.4.1 Transparent conductive oxide (front contact)

Transparent conductive oxides (TCO's) are an integral part of silicon thin film solar cells.

Applied as front contact or in combination with metal films as back contact, TCO films play a central role for the cell efficiency. As this electrode layer is situated at the front side of the solar cell, where the light enters the device, in addition to a high lateral electrical conductivity, this layer has to exhibit a high optical transmittance to the sunlight. Metal oxides belonging to the class of transparent conductive oxides (TCOs) combine these two properties.

1.4.2 The back reflectors

The back reflector should perform two important functions. It must be highly reflecting and it should also scatter light at an angle higher than the critical angle for total internal reflection.

Silver (Ag) is usually used to obtain high reflectivity. The interface between Ag and Si, however,

9

is not highly reflecting because of intermixing of the two elements, and a buffer layer of ZnO is deposited in between to prevent intermixing. The required texture for optimum scattering is obtained by depositing Ag and ZnO at a high temperature in the range between 100 and 400

0

C

[12, 14]. Back reflectors composed of two metals stacked together (Al and Ag) and must possess some physical and electrical properties in other to be qualified for use as part of the layers making up the thin-film solar cell.

The TCO (ZnO) must have good electrical and optical properties, which of course are interrelated. This means that efficiency considerations will imply a trade-off between these two

TCO qualities: the sheet resistance and the optical behavior. Low sheet resistance of the back reflector is an important feature as it also serves as a contact to the solar cell. Increasing the carrier mobility, rather than raising the carrier concentration by increasing the thickness of the reflector reduces the parasitic absorption here which is very important for the solar cell performance [12, 13, 14].

1.4.3 The p-layer and its properties

Light is normally incident on the p-side, and thus the junction characteristics of the TCO/p-layer interface must exhibit loss-free behavior [15] .The p-layer is deposited as the first semiconductor layer on top of the glass - TCO combination. The incoming light first crosses the player before getting into the intrinsic layer where absorption and photogeneration of the electron- hole pairs occurs. This layer should be relatively thin (about 20 nm in a microcrystalline silicon cell) and have a low absorption coefficient. This is important since it is required that most of the incoming photons gets to the i-layer without being absorbed in the p-layer.

A careful optimized p-layer is necessary to achieve both low- ohmic ZnO/p- contact and a high

V oc

[12,15]. Other feature of a good p-layer of a thin film solar cell includes:

• high activation energy

• high optical transmission on a specific wavelength range

• high crystallinity fraction.

High activation energy refers to the ability of the p-layer to offer low resistance path for the generated holes to be collected on the top electrode. One have to take into account two major

10

design facts, which are small thickness for enhancing transmittance but low sheet resistance for consideration as a good contact layer. However, the sheet resistance decreases as thickness increases, these two facts must be carefully resolved in the design of the p-layer.

When light penetrates into a material, it will be absorbed as it propagates through that material. The absorption of light in the material depends on its absorption coefficient which is related to the extinction coefficient. For a good p-layer, a high optical transmission on a specific wavelenght range is required in order to minimize optical and collection losses in the solar cell.

This high optical transmission is required for the light in the region where the photovoltaic absorber layer is active thereby ensuring that most of the incoming photons are collected at the i-layer and contributes to the photogenerated current. Moreover, a good crystalline volume fraction of the p-layer could have a positive influence on the nucleation of the i-layer for better performance. It is reported that increasing the diborane flow , the crystallinity of the layer decreases because boron suppresses nucleation [17, 18]. However, if the crystallinity is not affected, the conductivity increases as p-doping amount increases.

1.4.4 The i-layer

The layer is very important in the p-i-n structure of both the amorphous and microcrystalline silicon cells as it plays the role of the main photovoltaic active layer. Since photogeneration or absorption takes place here, the layer must possess some properties to be usable within the p-i-n structure. These include a high optical absorption coefficient in the useful spectral range of solar radiation, a sufficiently high mobility × lifetime (µ

τ

) products for the minority charge carriers.

Also important is the ability to allow for high and uniform internal electric field E(x), so that the corresponding drift lengths are sufficiently long for both carriers. However, this aspect should be improved, because, for dangling bonds in the lattices, there is a problem of non-uniformity of the electric field in the i-layers of both the a-Si:H and µc-Si:H solar cells. The thickness of the various i-layers is also a factor contributing to the enhancement of the photogeneration processes. For the growth of thin film microcrystalline solar cell, the best crystalline volume fraction value in the case of i-layers is between 50% and 60% [18, 19].

11

1.4.5 The n-layer of a solar cell

The n-layer of a thin film solar cell is also important in the p-i-n structure as it plays the role of collecting the photogenerated-electrons from the i-layer. This layer is another semi-conductor layer with a thickness of ~20 nm [2]. Separations of the electron hole- pairs takes place in the ilayer and due to the internal electric field already generated by the sandwiching of the i-layer in between the two doped layers, electrons drift towards the n-layer and holes to the p-layer and are eventually collected by the electrodes. The n-layer is expected to possess the ability to allow some of the photon not absorbed by i-layer from the back reflector (~300nm) reflected into the ilayer for further absorbtion. This is important for efficient light trapping.

1.5 Micromorph tandem solar cells

One of the strategy that can be used to improve the efficiency is the idea of stacking more than one single junction solar cells (multi-junctions arrangements). This structure can achieve higher conversion efficiency by capturing a larger portion of the solar spectrum. As shown in figure1.5, the maximal achievable value in terms of efficiency is roughly 39 % and corresponds to particular values of band-gap energies for used material to build the top and the bottom cell. In a micromorph, a-Si:H top cell and µc-Si:H bottom cells are joined. In 300 nm - 800 nm wavelength range, a-Si:H has a higher absorption coefficient than µc- Si:H but suffers from light induced degradation due to Staebler-Wronski effect. The Staebler-Wronski effect states that after some hours of light soaking, a-Si:H solar cells decrease their efficiencies, because of light energy which can break some bonds and create further on some defect states in the lattice. This will result in decrease of performance. This degradation depends also on the thickness of the layer: the thicker is the layer, the higher will be the degradation value. On the other hand, µc-Si:H solar cell exhibits higher absorption within the long wavelength range and doesn’t suffer from lightinduced degradation (already mentioned in section 1.2.2), therefore, stacking them together maximizes absorption through the entire solar spectrum. Besides, it will be possible to deposit thinner layers of amorphous silicon [2, 19, and 20].

From an operational point of view, photons not absorbed in the first cell, transparent to the lower-energy photons (i.e. higher wavelength) because of its low thickness, are transmitted to the second cell, which then absorbs the lower-energy portion of the remaining solar radiation. These

12

selective absorption processes continue through to the final cell if the structure was made up of more than two stacked solar cells (multi-junctions arrangement) as shown in figure 1.6.

Figure1. 5 Semi-empirical upper limit of the efficiency as a function of the energy gap E g

of the bottom and top cells of a tandem solar cell based on an empirical minimum value

(according to Green) for the reverse saturation current of the diode. [12 ]

An important design problem in such structures is the current matching between the top and the bottom cell. In fact, a good design should provide the same value of current that flows throughout the entire solar cell. Optimal matching means that both cells are operating in their maximum power point for MPP operation of the stacked cell. This is realized if the current densities of the top and bottom cell are equal at the MPP [10, 12].

13

Figure 1.6 The "micromorph" tandem cell, a Si:H top cell and a bottom cell

1.6 Sensitivity of microcrystalline silicon solar cell performance to deposition parameters

The growth of hydrogenated microcrystalline silicon or hydrogenated amorphous silicon can be achieved by different methods. They includes reactive sputtering of crystalline-silicon targets with Ar+H

2 plasma [22], mercury sentitized photo chemical vapour deposistion utilizing decomposistion reaction of silane molecules with photo-excited Hg [22,23]. Other methods are by direct photo chemical vapour deposition method. This is a method where high energy photons from Xe-resonance lamp or low presure Hg lamp are used for direct exitation of silane molecules to their electronic excited states [24]. Also the hot-wire chemical vapour deposition method to decompose silane and hydrogen by means of catalytic reactions on heated metal surface [25].

Plasma enhanced chemical vapour deposition method ( PECVD) is also another method used to grow either hydrogenated microcrystalline silicon (µc-Si:H) or hydrogenated amorphous silicon

(a-Si:H). The most commonly used method to produce ‘device quality’ a-Si:H, both on the laboratory and industrial scale is the radio frequency (13.56 MHz) plasma decomposition of

14

silane (SiH

4

)

, known as the plasma enhanced CVD (rf PECVD) method . This is prefered because of its large area deposistion ( more than 1m

2

) and low deposition temperatures ( 100 o

C <

T s

< 400 o

C) [2]. The growth mechanisms of µc-Si:H thin film and its characteristic properties has remained a major research subject partly due to its complex structure and the interdependence of the effects of the deposistion parameters. Without an appreciable light – induced degradation, a fabrication of microcrystalline solar cell of 8.5% efficiency has been reported using undoped microcrystalline sillicon as an active layer [18].

Hydrogenated µc-Si:H evolved as an off-shut of research in amorphous silicon by increasing the hydrogen dilution of silane, (silane-SiH

4 is used as an original source gas and is strongly diluted with hydrogen). When introduced into the plasma, during deposition, the thin film growths evolve first from amorphous incubation layer into a mixed phase of amorphous and crystalline phase and finally evolve into microcrystalline film. The transition from the amorphous to crystalline phase occurs when the hydrogen dilution is increased. The process of growing µc-Si:H material starts with the introduction of hydrogen to silane (SiH

4

) plasma, when this is done, the layer quality which was amorphous is altered. Increasing the hydrogen dilution further changes the state of the material until a threshold concentration known as amorphouscrystalline transition is reached [29, 30]. The best microcrystalline solar cells are deposited near this threshold concentration. However, for a very high value of hydrogen dilution beyond this region, voids are created within the material and leads to rapid oxidation of the sample [30, 42].

So at present, efforts are on by researchers to focus on the transition materials (materials deposited at the a-Si:H to µc-Si:H transition) in order to understand the transformation network because the best material is obtained within the transition. In order to determine this transition, the Raman spectroscopy is used. This measurement setup reveals such structural transformations and marks the transition. Other measurement setups used to accertain the material quality includes the transmission electron microscopy, the Fourier transform photocurrent spectroscopy

(FTPS) and the Fourier transform infrared (FTIR) spectroscopy [18,25,32].

The crystallization of thin film µc-Si:H from amorphous to microcrystalline regime occurs within a narrow process window which is affected by the following deposition parameters:

• rf power,

• gas pressure,

15

• electrode distance,

• silane concentration

• Substrate temperature.

The sensitivity study done by [18,19] revealed that by varying the substrate temperature, T s

,

( beween 250 - 550

0

C ) and silane concentration ( 1.5 – 10 %) while the chamber pressure and rf power were kept at optimal level and constant throughout deposition period, a good transistion from amorphous to microcrystalline silicon was obtained at a high temperature ( ~ 450

0

C ) and

10% silane concentration. The transition from amorphous to microcrystalline silicon was also achieved at low temperature (~250

0

C) with silane concentration of 4.5% by the same group. It was reported that the most convinient way is to adjust the silane concentration [SiH

4

]/[SiH

4

+H

2

]

[16,28,30]. In trying to study the effect of hydrogen dilution on layer crystallinity and microstructure, it is also reported that the best microcrystalline solar cells are deposited near the threshold concentration. At this point, further increase in hydrogen concentration results in the deposistion of crystallites, with a rapid increase in crystalline volume fraction until microcrystalline layer is obtained [30]. It is also stated that by increasing the proccess pressure in combination with high discharge power, a high deposition rate of µc-Si:H can be achived.

[16,17,31]. With this method, high solar cell of efficiency greater than 8% was attained using conventional rf (13.56 MHz) or very high excitation frequency [27]. This high efficiency of µc-

Si:H cell is obtainnable in the µc-Si:H growth regime close to the a-Si:H/ µc-Si:H transition [9].

Also by increasing the plasma excitation frequency, f exc,

from 13.56 MHz to around 70MHz, the deposition rate of device-quality amorphous silicon layers can be increased significantly by a factor of about 5 [27], although, this results in the reduction of the maximum energy of ions impinging on the growing surface in the reactor but it was reported that the flux is augumented at the same time. The variation of crystallite size within µc-Si:H layers was also invesigated and results shows an increase in crystallite size as f exc is increased [27].

In order to achive high deposition rate, the use of higher pressures (up to 10 Torr) i.e high- pressure depletion regime was reported [28]. The high pressure depletion and silane depletion conditions both in connection with RF-plasmas and with VHF-plasmas are required. Though high gas consumption and high plasma excitation power are needed but high-quality µc-Si:H cells of efficiencies between 7% and 9% have been reported [27]. It was reported [16,22,31] that

16

the variation of deposition pressure and power during i-layer growth shows that at deposition power = 0.4W/cm

2

and 0.5W/cm

2

, respectively with [SiH

4

]/[ H

2

] = 1%, the efficiency increases linearly until a value of 7.2% at 8 Torr and 7.4% at 9 Torr, respectively, are reached. When the pressure is increased, efficiency drops down to a significantly lower value [18,19,30]. So, we can conclude that all the deposition parameters already listed above are very important in the deposition of µc-Si:H solar cells with improve efficiency.

1.7 Purpose of the project

The aim of this research is in two major headings namely:

I. To understand the effect of p–layer deposition parameters on the short–wavelenght region of the solar cells

II. Investigate a-Si:H/ µc-Si:H transition materials deposited under different deposition conditions.

We hope to optimize the deposition parameters for making device quality µc-Si:H layer and implementing same to obtain high efficiency µc-Si:H solar cells. This will be done by investigating and optimizing the influence of deposition parameters on the µc-Si:H material properties. Parameters such as RF power, silane cocentration, deposition pressure would be studied. The research will include the effect of these parameters on the material properties of the intrinsic µc-Si:H layers and on the performance of single junction solar cells and the a-Si:H/µc-

Si:H transition film are developed and characterized. Also included in the task is p-layer optimization, fabrication of optimal µc-Si:H and device characterrization.

1.8 Project Organisation

The project will be organised in different sections. The first chapter will discuss theory and general knowledge on solar cells down to thin films solar cells and sensitivity of deposition parameters. The second chapter will be on the measurement setups and equipment used in solar cells characterization while the third chapter will dwell on microcrystalline p-layer optimization.

17

Furthermore, the sensitivity study of microcrystalline i-layer deposition will take the fourth chapter. Chapter five of this report dicusses the conclusion and proposal for further work based on the obtained results.

18

Chapter Two

Measurement and Characterization.

2.0 Introduction

This chapter is dedicated to deposition techniques and measurement setups used in the growth and characterization of µc-Si:H materials and devices. The principles of operation of the different set ups used are described in details.

2.1

Deposition of microcrystalline silicon

The methods that are used for depositing thin films of hydrogenated amorphous silicon (a-

Si:H) and hydrogenated microcrystalline silicon (µc-Si:H) can be divided in two groups. The first group includes methods that form a-Si:H and µc-Si:H from a gas phase by decomposition of silicon– bearing gas, usually silane, SiH

4

known as chemical vapour deposition (CVD) methods.

19

The second method is by physical deposition in which silicon atoms for a-Si:H are obtained by sputtering a silicon target [2].

As already indicated in section 1.2.1 about the mostly used method of deposition, the schematic diagram of a PECVD reactor for microcrystalline deposition is shown in figure 2.1 below.

Figure 2.1 Schematic diagram of the PECVD reactor for deposition of microcrystalline silicon

[26].

It comprises an anode and a cathode electrode. As can be seen from the diagram, the substrate is attached to the anode. The cathode is the powered electrode while the anode electrode is connected to the ground. Dissociation of silicon bearing gas i.e silane is done by means of electronic collision with plasma that plays the role of a source of accelerated electrons. When the substrate and the electrodes are slightly negative with respect to the plasma bulk, negative ions are trapped in the plasma and this creates dust by reacting with silane under high-pressure.

The positive ions and the radicals reach the substrate by drift and diffusion respectively. This deposition method however results in a more complex phenomenon regulated by deposition parameters like the power density, substrate temperature, gas composition and flow and deposition pressure [30, 31].

20

2.1.1 The AMOR deposition system

The AMOR deposition system is used to deposit rf PECVD-based p-i-n solar cell in the

Laboratory of Photovoltaic Materials and Devices, TU Delft- Dimes. This system is computer controlled and was recently automated for better and accurate performance. The control system is used to manipulate the robotic arm transport mechanism and the deposition parameters. It comprises four reaction chambers and a load lock system installed around a centrally located transport chamber (ITZ). Each of the chambers is dedicated to the deposition of one or two types of material. This is done in other to avoid contamination of the samples. The first chamber, MPZ1, is used for p-type µc-Si:H and a-Si:H deposition, MPZ2 for n-type a-Si:H,

MPZ3 for intrinsic a-Si:H layer and the last chamber, MPZ4 is specially dedicated to deposition of µc-Si:H and silicon based multilayer / supperlatices. The remaining parts of the AMOR deposition system constitute the vacuum pump, gas handling system, rf power generator and substrate heating compartment . Shown in figure 2.2 is the top view of the system with a complete details of the above described part composition.

.

Figure 2.2 Top view of AMOR deposition system

21

2.2 Material and Device characterization

The methods used in characterising µc-Si:H layers and the microcrystalline silicon solar cells can be divided into two headings, namely, material and device characterization methods. Under material characterization, the electrical, optical and structural material properties are investigated. For the device characterization, I-V and quantum efficiency measurements are done to accertain the overall performance of the solar cells.

2.2.1 Electrical characterization

2.2.1.1 Dark and photoconductivity measurement

These measurements show the conductivity of the layers under illumination and in the dark.

Before carrying out this measurement using the setup, the sample with contact (electrodes) is annealed for 30 minutes at 130

0

C. To determine dark conductivity (σ d

), a bias voltage in the range of -100 to 100V (for layers) is applied while 0.1 to 1V is used for solar cell in the solar cell group of TU-Delft. When this is done, then the current through the sample can then be measured.

The dark conductivity is determined as:

σ

=

d d I t l V

(1)

Where I is the measured current, V is the applied voltage, d is the gap between the two coplanar electrodes, l is the length of the electrodes and t is the thickness of the sample. The dark conductivity is temperature dependent σ d

(T) and this is expressed by Arrhenius equation as

σ

d

=

σ

o

e

E a c t k T

(2)

ܧ

௔௖௧

is the activation energy, ߪ

is a conductivity prefactor, T is the absolute temperature and

K

the Boltzman’s constant. In equation 2 above,

ܧ

௔௖௧

is the quantity to be determined and it indicates approximately the position of the Fermi level in the semiconductor material. The

22

activation energy is a good measure to evaluate the presence of impurities in the film often act as semiconductor material dopants. The same setup is used to determine both the dark conductivity and activation energy of the material. The difference is in the use of a temperature control device (Temptronic) for activation energy measurement.

Photoconductivity on other hand is the sensitivity of a material to conduct electricity when illuminated. The AM 1.5 light spectrum with an incident power of 100 mWcm

-2

is used for this measurement using the solar simulator setup in the measurement room. The IV responses of both the i-layer and the solar cells can be obtained with these set up.

2.2.2 Optical Properties Estimation

Optical properties of a-Si:H are usually characterized by the absorption coefficient (α), the complex refractive index, ñ = n– jk (where n is the real part of ñ, j is the imaginary part and k is the extinction coefficient) and the value of the optical band-gap (E opt

). The complex refractive index describes how efficiently the photons of a particular energy are absorbed in the material.

The optical band-gap determines which part of the solar spectrum is absorbed. Most effective material for solar energy conversion have a band-gap in the range of 1.0 to 1.8 eV, whereas the entire solar spectrum of sunlight from infrared to ultraviolet, covers a range of about 0.5 eV to about 4.0 eV [2, 49].

Optical absorption measurements are used to determine the density of states distribution in a-

Si:H because the optical absorption coefficient, α(E), is determined by optical transitions involving all pairs of occupied and unoccupied electronic states that are separated by the same photon energy E. The absorption coefficient generally depends on the wavelength of the incoming photons [49]. Figure 2.4 below shows the absorption spectra of microcrystalline, amorphous and crystalline silicon.

In order to obtain a more accurate optical property measurements of the materials, the solar cell group of TU Delft uses a program known as OPTA, which uses the optical data to calculates α

(E) by modeling the sample in this format, air-glass-fill-air. This program is used to estimate the extinction coefficient, k values from the refractive index, n values obtained using the mini RT set up. The Dual Beam Photoconductivity (DPB) results obtained from the Fourier Transform

Photoconductivity Spectroscopy (FTPS) measurement is uploaded and fit together for more

23

accurate data. From this integration, parameters like the optical gap, E

04

, Ubach energy, Tauc gap and the thickness measurement are obtained. These data are very important in quality estimation of the material.

2.2.2.1 Fourier Transform Photocurrent Spectroscopy

The Fourier transform photocurrent spectroscopy (FTPS) is used to determine the sub-band gap absorption of photon in layers and solar cells. This technique is important because it contains useful information about the quality of a material. It can be used to determine the

Urbach energy and defect density of the layer. FTPS is a sensitive and accurate measurement technique that can be used to calculate the absorption coefficient spectrum of a material over a wide energy range (typically 0.5 eV – 3 eV) by calculating the spectrum of the photocurrent using an interferometer. The measurement setup consists of Fourier transform infra-red (FTIR) spectroscopy setup with a mounted optical filters, current amplifier, and voltage source and aluminum chamber including 3D displacement stage with sample holder. This measurement setup is used for both FTIR and FTPS measurements but with different methods (beam splitter).

Shown in figure 2.3 below is FTIR/ FTPS measurement setup in The Laboratory of Photovoltaic

Materials and Devices, TU Delft- Dimes.

Figure 2.3 Fourier transform photocurrent spectroscopy setup

24

The interferometer, Thermo Electron Nicolet 5700 Fourier Transform Infrared spectrometer type is used. It consists of two perpendicular plane mirrors, one is fixed and the second mirror moves along the axis perpendicular to its plane. The beamsplitter (CaCl

2

) at the middle of the interferometer splits the light from the source into two parts with one part of the light reflecting towards the fixed mirror while the transmitted part strikes the movable mirror. A beam is reflected from the two mirrors which returns to the beamsplitter and then causes interference to occur. Only the the output beam which propagates in the direction perpendicular to the input beam is used while the beam returning to the source is ignored because of the difficulty associated with separating it from the source light. The intensity of the output beam depends on the optical path difference between the two interfering beams [21].

The measurement is done by applying a voltage of 1kV on the sample, the induced photocurrent is then amplified by a Keithley model 428-PROG current preamplifier. The amplified signal is then converted into a digital signal by an analogue-to-digital converter and fed to a computer. The converted signal is then transformed to the photocurrent spectrum by performing the inverse cosine Fourier transform. Mounted on the filter wheels are five filters inside the spectrometer. They are used to cut off all other wavelengths below the laser wavelength in order to avoid aliasing. The filters have cut-off wavelength of 645 nm, 695 nm, 780 nm, 850 nm, and 1050 nm respectively and they are named differently for easy identification. These filters measures different parts of the spectrum and they are controlled by OMNIC program (sofware), on the computer. Proper callibration and alignment with ZnO is done in order to ensure that the detector inside the FTIR spectrometer is in the wavenumber range of interest (4000 cm

-1

– 15700 cm

-1

). After this step, the background measurement is done using each of the five filters and the obtained spectra from the filters are stored on the computer.

The stored data are processed in a MATLAB script file [21]. The result obtained from

Matlab is a relative photocurrent spectrum for all the samples defined in the input matrix of the script. The result obtained from the FTPS measurements is fitted together with the RT files from the mini RT or LAMBDA (CRT) in OPTA program developed in the group in which the thicknesses of the samples are adjusted so that there is smooth connection of the spectrum form

FTPS and spectrum from RT or CRT measurement.

25

2.2.2.2 Measurement of the sub-gap absorption coefficient

Sub band-gap absorption is commonly measured with very sensitive techniques like Photo

Deflection Spectroscopy (PDS), Constant Photocurrent Measurement (CPM) and Fourier-

Transform Photo Spectroscopy (FTPS). The absorbance spectrum of interest extends roughly from 0.7 eV to 2 eV. In addition to the measurement of the optical band-gap, the absorption spectrum yields two important parameters related to the disorder and the defect density of the probed material. These parameters are related to the gap states that are undesirable for solar cell applications. Indeed, if there are states in the centre of the gaps, they act as recombination centers, thereby limiting the electrical performances.

In the region just below the gap (1.12 eV for µc-Si:H and 1.75 eV for a-Si:H), the absorption coefficient (and the FTPS spectrum) stems from optical transitions involving band tail states.

The spectrum in that region increases exponentially with the photon energy, forming the socalled Urbach tail. The exponential slope E

Urbach

of the absorption spectrum characterizes it. It is determined by fitting the absorption curve using the following formula:

ߙ = ߙ

݁

ா/ா

ೆೝ್ೌ೎೓

(3) where

ߙ

is the exponential pre-factor.

In microcrystalline silicon, the Urbach tail before the bend edge extends from 1.12 eV. The slope at that edge is described by the so called E

Urbach

, e.g. 36 to 40 meV [49] are measured for device grade µc-Si:H material. E

Urbach can be related to the material disorder.

The defect density is estimated from the value of the absorption coefficient at a photon energy of 0.8 eV for µc-Si:H silicon (α

0

.

8eV

). The figure 2.4 below shows the absorption spectra of different material obtained from constant photocurrent method (CPM).

26

Figure 2.4 Absorption spectra of microcrystalline, amorphous and crystalline silicon measured by CPM [45].

2.2.2.3 Reflection and Transmission Measurements

In other to obtain the total transmitted and reflected light of a sample, RT measurement incorporating the Total Integrated Sphere (TIS) in the Perkin-Elmer photospectrometer, Lambda

900 is used. The TIS comprises a light source for producing an incident beam of light at a known wavelength range; source optics for directing the incident beam at an incident angle (θ i

). In addition, a hollow sphere having a radius (R s

) and configured with an input aperture, a sampling aperture and an output aperture. It has light source, source optics, and sphere positioned such that the incident beam is directed through the input aperture, through the sampling aperture, and onto the surface. This is done such that the specular beam reflected off the surface is directed out of the sphere through the output aperture. The interior surface of the sphere includes an absorption region surrounded by the sampling aperture while the interior surface of the sphere outside the absorption region comprises of reflective region. The absorption region having a reflectance less than the reflectance of the reflective region over the wavelength range of the light source. To detect the intensity of light within the sphere, a scatter detector is incorporated while a specular detector detects the intensity of the reflected specular beam [50].

The schematic of this setup is shown in figure 2.5 below.

27

Figure 2.5 Schematic diagram of Perkin-Elmer photospectrometer, Lambda 900 system

The mini RT as will be mentioned in the next section can also be used to carry out R&T measurements. The Lambda system is also suitable for the determination of the specular and the diffused components of both the transmitted and reflected light from where haze parameters can also be determined.

2.2.2.4 The mini RT measurement setup

The mini RT measurement setup as already indicated above is used for the measurement of reflection and transmission of light in a material and it is one of the methods used for obtaining the absorption coefficient of a material. It is suitable when measurements on rough surfaces is required. The reason is because, in rough surfaces, significant portion of the light is scattered and this cannot be detected by the respective sensors unlike the Lambda system. This same setup can also be used to estimate the thickness, band gap and Urbach energy measurement of the deposited films. Before the actual measurements, the detectors are calibrated by silicon and germanium photodiodes using light generated by a 100 Watt halogen lamp that illuminates the sample perpendicularly. The silicon wafer was used in our own case. To measure reflection and transmission as a function of wavelength, a Spex 1680B double grating monochromator is used that scans through the spectrum from 2.60 eV 0.70 eV in steps of 0.02 eV. Because of the above reasons, we only used this setup to measure other parameters like the thickness, band gap and

Urbach energy of our layers and not R&T.

28

2.2.3 Stuctural Properties Characterization

2.2.3.1

Fourier Transform Infrared Spectrometer

The Fourier transform infrared (FTIR) spectrometer measures the hydrogen bounding in material i.e. checks hydrogen incorporation on a layer. The sample to be measured is connected to an electrical circuit with a low noise voltage source and a current preamplifier for the amplification of its photocurrent.

The system is equipped with an A/D converter, which digitizes the output of the preamplifier, and the signal is translated to frequency domain from time domain by Fourier transformation. The FTIR signal from the sample is normalized to the FTIR signal from a spectrally independent detector. The measurement of the background (i.e. the chamber before and after deposition for i-layer and crystalline silicon wafer) is done. The sample is then measured and the obtained spectra are analyzed by fitting with ORIGIN program. From the spectra, information on hydrogen-bonding configuration of the material is obtained. These are in three modes in amorphous network, the wagging mode at 640 cm

-1 of Si-H x

, and the two stretching modes, which are the low stretching mode (LSM) at 1980 – 2000 cm

-1

and the high stretching mode (HSM) at

2070 – 2100 cm

-1

. HSM absorption is observed to be dominated in a-Si:H that has inferior electrooptical properties [47].

The microstructure factor,

*

R , is an important parameter and a figure of merit for a material structure and is defined as

R

*

=

Ι

HSM

(

Ι + Ι

LSM HSM

)

(4) where

Ι

HSM

is the HSM integrated absorption strength and

Ι

LSM

is the LSM integrated absorption strength. The integrated absorption of the wagging mode (640cm

-1

), the low stretching mode (~2000cm

- 1

) and the high stretching mode (~2100cm

-1

) are normally used to determine the total hydrogen content in a-Si: H, the hydrogen incorporation into both vacancies and voids respectively.

29

C

H

=

N

H

N

SI

+

N

H

(5)

Where N

H

and N

SI

are hydrogen and silicon densities respectively and N

SI

+ N

H

is approximated to be

×

22

5 10 cm

− 3

[47].

The density N x of the Si-H x

mode is obtained from the integrated absorption strength and it is defined as

Ν = Α

x x

ω

d

ω

(6) where A x

is the proportionality constant and

( )

is the absorption coefficient [47].

Unlike the crystalline and amorphous silicon, the plasma deposited hydrogenated microcrystalline silicon is not a unique material in terms of its microstructure and content of its constituent phases. This is known to be a complex material consisting of crystalline and amorphous silicon phases plus grain boundaries. Its crystallite may have a disribution in shapes, sizes and orientation. It may be very difficult to arrive at a qualitative analysis of the material microstructure. Therefore, in order to have a comprehensive and detailed analysis of the material microstructure, consistent microstructural characterization program or tools would be required.

Another important factor is the fact that hydrogenated microcrystalline silicon exhibits a wide range of microstructures that depend both on the deposition conditions [44] and on the substrate material. The microstructure of µc-Si:H varies not only with silane concentration but also while growth of the film proceeds. It was observed in [34, 38] that a fully amorphous incubation layer has frequently been observed at the bottom of films that are grown close to the µc-Si:H /a-Si:H transition.

The microstructure factor

*

R is defined as the ratio of the areas of the multihydrides in the bulk to monohydrides and multihydrides in the bulk. In hydrogenated silicon, the microstructure is defined as the ratio of the integral intensities of the LSM and HSM absorption bands, it expresses the ratio of hydrogen bonded in the HSM to the total bonded hydrogen [ 47, 48 ] .

30

R

=

׬ ୍

ౄ౏౉

׬ ୍

ౄ౏౉

ሺ஝തሻୢ஝ത

ሺ஝തሻୢ஝തା ׬ ୍

ై౏౉

ሺ஝തሻୢ஝ത

(7)

The number of hydrogen atoms bonded to silicon is N = A x

׬ሺ

஑ሺ஝തሻ

ሻdνത, where A

஝ത x

is the proportionality constant for a specific vibrational mode, α

ሺνതሻ is the wave number-dependent absorption coefficient of the vibrational mode determined from absorbance spectra [48]. The hydrogen concentration in atomic percent is expressed as c

H

= (A x

/ N int

)

஑ሺ஝തሻ

ሻdνത , where N

஝ത int

is the total atomic concentration of the films (5

× 10

22

cm

-3

for c-Si). The hydrogen content is preferably calculated using the wagging vibrational mode at 640 cm

_1

, because the proportionality constant A x

for this vibrational mode is generally accepted for both a-Si:H and

µc-Si:H [47,52].

According to [47], monohydrides in vacancies contribute to the LSM and hydrides on the void surfaces to the HSM. The peak at 2000 cm

-1

is often attributed to the isolated hydrogen in monohydrides bonding configuration. Hydrogen manifested in the LSMs is bonded to silicon in monohydrides in small volumes of monovacancies, divacancies or polyvacancies. The HSM peak at 2090 cm

-1

is assigned to clustered hydrogen in monohydrides, dihydrides or trihydrides at the internal surfaces of voids or at the boundaries between the crystalline grains. The presence of Si–Hx while x > 1 is common for materials with microvoids [47, 48 and 50]. To determine the microstructures of the microcrystalline layers, we adopted the methods used for both the a-Si:H and the c-Si as sugested in [48]. The hydrogen content in material was calculated using the wagging vibrational mode at 640 cm

_1

.

2.2.3.2 Raman Spectroscopy

Raman spectroscopy is a light scattering technique. It is a process where a photon of light interacts with a sample to produce scattered radiation of different wavelengths [51]. The measurement setup is very useful in the characterization of the crystallinity properties of the layers of solar cells. It can be used to ascertain the structural properties of the material

(crystallinity ratio or fraction) that gives information on the volumetric percentage of the crystalline phase in materials. This is a very useful means in determining the microcrystalline – amorphous transition especially with respect to i-layer structural characterization. Raman spectroscopy is used to measure the amorphous or crystalline volume fractions of the µc-Si:H

31

materials. The evaluation of the crystallinity fraction of a material like the microcrystalline silicon is very important because it consists of both the crystalline and amorphous part.

2.2.3.2.1 Evaluation of crystallinity fraction

The evaluation of the crystalline volume fraction is done by monitoring the changes in the transverse optical (TO) peak. TO is a vibrational mode of the phonons. The crystalline volume fraction evaluation is done from the Raman peaks. Gaussian deconvolution of Raman scattering profile of hydrogenated microcrystalline silicon is performed using two Raman peaks contributing to the Raman spectrum of µc-Si:H silicon. The first is the crystalline peak at 520 cm

-1

corresponding to transverse optical mode (TO) in crystalline silicon and a broad peak centered at 480 cm

-1 for the amorphous silicon (a-Si:H TO mode) [53]. We subtract a scaled

Raman spectrum that was obtained from an the amorphous silicon film from the Raman spectrum of the microcrystalline silicon film. The result of the subtraction is the Raman spectrum of the crystalline part of the microcrystalline film and the crystalline fraction can be determined.

This method as proposed by [53] is used in the determination of the crystalline fractions in the

Laboratory of Photovoltaic Materials and Devices, TU Delft- Dimes. This procedure is implemented in a matlab code and is used for the crystallinity volume fraction analysis of the deposited samples

2.2.4 Device Characterization

2.2.4.1 Measurement of external parameters of solar cells

The determination of the performance of a good solar cell is done by investigating its external parameters namely the short-circuit current density, J sc

, open-circuit voltage, V oc

, fill factor, FF, and the conversion efficiency, η. These parameters are used to characterize the performance of solar cells and they are determined from the illuminated I-V characteristic as shown in figure 2.8.

The short-circuit current density (J sc

) is the current flowing through the cell when the voltage is zero (V = 0). In the ideal case, J sc

is equal to the photo-generated current density J ph

, while the open-circuit voltage (V oc

) is the voltage measured on the solar cell when no current flows through the device and it depends on the photo-generated current density.

32

The fill factor (FF) is the ratio between the maximum power deliverable by a solar cell and the products of V oc

and J sc

while the conversion efficiency (η) is calculated as the ratio between the generated maximum power and the incident power.

Figure 2.6 I-V characteristics of a p-n junction in the dark and under illumination [2]

The V oc is given as

V o c

=

k T q

l n

J p h

J o

+ 1

(8)

This is obtained by setting the net current in diode equation to zero. J o

the saturation current density, q is the elementary charge, k the Boltzmann constant and T is the temperature.

The fill factor, FF is given by:

F F

=

J m p

V m p

J s c

V o c

(9)

The conversion efficiency,

η

, is

33

η

=

P

max

P in

=

J V mp mp

P in

=

(10)

P in

The irradiance value, P in

, of 1000 W/m

2

of AM1.5 spectrum is a standard for measuring the conversion efficiency of solar cells.

In determining these parameters, the measurement is done by illuminating the sample with a continuous solar simulator providing an irradiance of 1000 W/m

2

under AM1.5 condition while sensing the photo-generated current flowing into the probes connected to the front and back contact of the solar cell providing a voltage between them.

2.2.4.2 Quantum efficiency measurement (QEM)

The quantum efficiency measurement is performed on solar cells to determine the spectral response. It gives the percentage of photons hitting the photoreactive surface that will produce the electron-hole pair. It also gives information on the current that a given cell will produce when illuminated. The setup comprises a light source used to illuminate the solar, chopper monochromator for illuminating the solar cell at a standard frequency of 123 Hz (depending on the set up). Also included is the lock-in amplifier designed to measure small signals in noise. The internal oscillator controls the chopping frequency while the power supply unit used to apply potential over the solar cell. Hamamatsu S1337-1010BQ silicon photo diode is used as reference cell (calibration) and a computer providing the software written in Lab View for the control of the QEM setup. Figure 2.7 is a picture of the QEM setup.

The quantum efficiency relates the response of solar cell to the wavelength of light. This measurement is important because it gives a record of how much of the generated charge carriers are collected at the electrodes within the solar spectrum.

The external QE is the ratio of the number of collected photo-generated carriers to the number of photons incident on the solar cell while the internal QE is the ratio of the number of photogenerated carriers collected by the solar cell to the number of incident photons.

34

Figure 2.7 Quantum efficiency measurement setup.

The QE at a particular wavelength is one if all photons of energy are absorbed and the generated photo carriers are collected. If the band gap of solar cells is greater than the band gap of the incident photons, then it could be said that charge carriers are not generated and thereby, the eternal QE is zero.

35

Chapter Three

Influence of deposition parameters on µc-Si:H P-layer

3.0 Abstract

The influence of deposition parameters of microcrystalline silicon layer on the performance of

µc-Si:H silicon solar cell was investigated in this report. Series of p-layer depositions were made for p-layer optimization using a PECVD multi-chamber system at a frequency of 13.56 MHz. In the first deposition series, SiH

4

= 1.3 sccm, deposition pressure (2.5 mbar), plasma power (60

W), and hydrogen flow of 200 sccm were used while diborane flow was varied from 0.2 – 0.5 sccm. The rf power was varied from 30 - 90 W in the second series with deposition time fixed at

300 s while keeping the diborane at 0.2 sccm. The other deposition parameters were kept constant as in the first series. The results results indicates the deposition of high conductive and crystalline fraction thin film µc-Si:H p-layers. These layers were implemented in p-i-n type solar cells. The spectral responses of the solar cells show a relatively higher quantum efficiency value in the short wavelength region for lower diborane flow. This is an indication that the spectral response is highly sensitive to diborane flow at short–wavelength region.

36

3.1 Introduction

The p-layer is the window layer of a solar cell. Looking at the structure of a p-i-n solar cell, solar radiation first passes through this layer and then further into the solar cell. A good design of this layer is very important for the optimal performance of the entire solar cell.

In this research, our attention is on the optimization of µc-Si:H p-layers in p-i-n of a µc-Si:H solar cell. As already stated in section 1.4, our target is to achieve µc-Si:H p-layer with high conductivity, offering low resistance path for the holes to be collected at the respective electrode, high crystalline fraction sufficient enough to promote the nucleation of the microcrystalline ilayer and high transparency of light. We investigated the effect of diborane flow, p - layer thickness and deposition power on the solar cell performance. All other deposition parameters were kept constant. In summary, the deposition series are as follows: a) Diborane flow series b) Thickness series c) Power series

3.2 Experimental Details

The p-layers were deposited on 40 seconds etched ZnO:Al which was etched by dipping in 0.5 %

HCl. In order to determine the electrical parameters of the p-layers, each deposition run had a

Corning Eagle 2000 glass upon which the p-layer is deposited. Three series of depositions for the optimization of the microcrystalline p-layer deposition parameters were carried out. Diborane was used as doping gas and was varied from 0.2 to 0.5 sccm. The deposition time was varied from 200 to 500 seconds resulting in layer thicknesses between 12 and 40 nm. The layers were characterized by measuring the following: RT using the Perkin-Elmer spectrometer, Lambda 900 and crystalline volume fraction (Raman spectroscopy). The dark conductivity and activation energy measurements were carried out using their respective measurement setups. Using the same deposition parameters, the p–layers were implemented in µc-Si:H solar cells. Under AM

1.5 at 25 o

C, the I-V characteristics of the solar cells were determined. Using the solar simulator from quantum efficiency set up, the spectral responses of the solar cells was equally determined.

37

3.3 Results and Discussion

3.3.1 Substrate material (TCO)

From figure 3.1, it can be seen that in TCO with flat surfaces (0 s etched), there is not much scattering of light, what we can notice is interference fringes. Here light is partly transmitted and reflected with respect to the normal of the interface. In the case of TCO etched at 40 s, one can see a clear distinction; this shows the scattering of light as a result of the textured surfaces of the

TCO. Here greater percentage of the light is transmitted in the visible range of the solar spectrum. Comparing transmittances of the etched TCO and TCO with p-layer, a drastic reduction in the transmittance of the ZnO/p-layer substrate. At 400 nm, the transmittance reduces from about 60 % to 15 %. This is an indication of enormous absorption of the incident light by the p-layer.

60

40

20

100

80

ZnO + P-layer (0 s)

ZnO (0 s)

ZnO + P-layer (40 s)

ZnO (40 s)

0

400 600 800

Wavelenght (nm)

1000

Figure 3.1 Transmittance of etched and flat ZnO:Al and ZnO:Al with P-layers.

38

Surfaces provided with rough textures are used to achieve the scattering of incoming photons into the absorber layers and thus the junction characteristics of the TCO/p-layer interface must exhibit loss-free behavior. The absorption coefficient of the p–layer should be low. This is important since it is required that most of the incoming photons gets to the i-layer without being absorbed in the p-layer.

3.3.2 Microcrystalline P-layer

Figure 3.2 presents the transmittance of p–layer deposited at different deposition times for 0.2,

0.3 and 0.5 sccm of diborane as a function of wavelength. The figure shows that transmittance of photons through the p-layers in the visible range (400 nm) of the solar spectrum is highest (about

20 %) at about 12 nm layer thickness (200 s deposition time). The diborane flow content of this layer is 0.3 sccm. Transmittance decreases as the thickness increases for diborane flow of 0.5 sccm. This trend is expected because transmittance and absorption of photons are thickness dependent. As you increase the diborane flow and deposition time, the layer becomes thicker.

The layer deposited after 500 s is thicker (about 38 nm) than the one at 200 s (about 12 nm) and as such should absorb more and allow less photons to pass through. Probing further, another plot was made to have a better understanding of the thickness dependence of transmittance of photons through the p-layer. This is shown in figure 3.3. It can be seen from this figure that the same argument holds for the two figures (3.2 and 3.3).

Interestingly, if we consider the layer at 0.2 sccm of 300 s deposition time, the thickness here is about 21 nm. This layer looks better in terms of transmittance and applicable since we desire a layer that is not too thin and not very thick and highly transparent. It is expected that this layer should also be highly conductive.

39

40

20

100

80

500 s

60

80

60

40

20

300 s

0.2 sccm

0.3 sccm

0.5 sccm

0.2 sccm

0.3 sccm

0.5 sccm

80

200 s

60

40

20

0

400

0.2 sccm

0.3 sccm

0.5 sccm

600

Wavelenght (nm)

800 1000

Figure 3.2 Transmittance of different thicknesses for varying diborane flow.

40

100

80

0.2 sccm

60

40

20

80

0.3 sccm

60

40

20

500 s

300 s

200 s

500 s

300 s

200 s

40

20

80

60

0.5 sccm

500 s

300 s

200 s

0

400 600 800

Wavelenght (nm)

1000

Figure 3.3 Transmittance in p–layers at different diborane flow.

41

3.3.3 Effect of diborane flow on p-layer crystallinity and conductivity

Generally, as doping amount increases, the conductivity of a layer also increases correspondingly and the crystallinity decreases [54]. The increase in electrical conductivity is because of the addition of more free charge carriers (dopant). In figure 3.4, we see a drop in activation energy with increasing diborane flow for all the thickness series. Another interesting trend is the dependence of both the conductivity and crystalline volume fraction on deposition time for all flow series. These two parameters increase with increasing deposition time and the activation energy drops with increasing deposition time (figure 3.5).

However, comparing with figure 3.2, the transmittance drops with increasing thickness hence a trade-off is necessary in this aspect in order to choose an optimum deposition time.

Moreover, increasing deposition time leads to a drop in activation energy for all the diborane flow. On dilution with diborane, the Fermi level of the material moves towards the valence band, hence a drop in the activation energy. Activation energy is defined as the distance between the

Fermi level and the edge of the valence band.

42

0.5

0.4

0.3

0.2

(a)

0.1

200 s

300 s

500 s

1e-1

1e-2

1e-3

1e-4

1e-5

1e-6

(b)

200 s

300 s

500 s

60

55

50

45

40

(c)

200 s

300 s

500 s

35

30

25

0.20

0.25

0.30

0.35

0.40

Diborane flow (sccm)

0.45

0.50

Figure 3.4 Crystallinity volume fraction, conductivity and activation energy against diborane flow.

43

0.3

0.2

0.1

0.6

0.5

0.4

(a)

1e-1

1e-2

1e-3

1e-4

1e-5

1e-6

(b)

0.2 sccm

0.3 sccm

0.5 sccm

0.2 sccm

0.3 sccm

0.5 sccm

60

50

40

30

20

(c)

0.2 sccm

0.3 sccm

0.5 sccm

10

200 250 300 350

Deposition time (s)

400 450 500

Figure 3.5 Crystalline volume fraction, conductivity and activation energy against deposition time.

At 0.2 diborane flow, 200 s deposition times, the layer thickness is about 14 nm. This layer is very thin because of the doping content and deposition time compared to the layer deposited at

44

500 s with 0.5 diborane flow. The trend as seen in figure 3.5 indicate that the conductivity increases with crystallinity.

The deposition rate of the p-layer is indicated in figure 3.6. As expected, the deposition rate increases with increasing diborane flow. More diborane flow means more dopants to the material, hence higher deposition rate. This is observed in all the three series.

50

40

0.2 sccm

0.3 sccm

0.5 sccm

30

20

10

0

200

Deposition rate

0.2 sccm = 0.0532 nm/s

0.3 sccm = 0.0725 nm/s

0.5 sccm = 0.0867 nm/s

250 300 350 400 450 500

Deposition time (s)

Figure 3.6 Thickness and deposition rate of µc-Si p-layer at different deposition time and doping.

At 0.2 sccm, 300 s deposition time, the activation energy of the layer is about 0.5 eV with corresponding conductivity of 1 × 10

-6

-1

cm

-1

. The conductivity of this layer is high enough to allow the generated holes to be collected at the electrode. In addition, the crystalline fraction (47

%) is also sufficient to promote the nucleation of microcrystalline i-layer.

45

3.4 Application to solar cells

In order to verify the effect of the p-layers properties on solar cells, the optimized layer was applied to a single junction p-i-n type µc-Si:H solar cells with i-layer thickness of about 1000 nm and amorphous silicon n-layer with thickness of 20 nm. The structure of this solar cell is glass/ZnO/p-type µc-Si:H/intrinsic µc-Si:H /n-type a-Si:H/Ag/Al. The solar cells were deposited for the p-layer diborane flow and power series. Shown in figure 3.7 is the quantum efficiencies of the solar cell plotted for the different diborane flow series.

1.0

0.8

0.6

0.4

0.2

0.3sccm

0.2 sccm

0.5 sccm

0.0

400 600 800 1000

Wavelenght (nm)

Figure 3.7 Spectra response of solar cells as a function of p-layer diborane flow.

From the figure, the layer with diborane flow of 0.2 sccm has a quantum efficiency at 400 nm of approximately 38 % while at 0.3 and 0.5 sccm gave values of 35 and 12 % respectively. At lower diborane flow as in the case of 0.2 sccm, the layer is more crystalline and conductive hence results in higher short-wavelenght response when compared with the 0.3 and 0.5 sccm. In the long-wavelenght range, the quantum efficiency of these solar cells is almost the same except for a slight increase in the layer with 0.2 sccm diborane flow. This results shows that the shortwavelenght spectral response is highly sensitive to the opto-electronic properties of the p-layer.

At 0.2 sccm the layer is transparent enough to allow more photons to pass through hence less

46

absorption and high photogeneration of charge carriers. In addition, the crystallinity of the player is high leading to an increase in quantum efficiency.

5.4

5.3

5.2

5.1

5.0

4.9

4.8

4.7

(a)

0.64

0.63

0.62

0.61

(b)

(c)

20

19

18

17

16

(c)

0.48

0.47

0.46

(d)

0.45

0.44

0.43

0.20

0.25

0.30

0.35

0.40

Diborane flow (sccm)

0.45

0.50

Figure 3.8 µc-Si:H solar cells external parameters versus diborane concentrations of p-layers at different deposition time.

Figure 3.8 shows the external parameters of a single junction p-i-n type µc-Si:H solar cells of p-layers of different thicknesses deposited at varying diborane flow. The layer layer with 0.2

47

sccm diborane concentration will be preferred as our optimized p-layer in line with the other parameters of interest. The V oc

increases as the diborane flow increased from 0.2 to 0.5 sccm.

This is because the layer becomes more amorphous with increase in diborane flow. The layer deposited at 0.2 sccm diborane and 300 s deposition time has crystallinity volume fraction of about 40 % while 500 s at the same diborane has crystallinity of about 58 %.

We see an interesting trend in the short-circuit current density (J sc

) results. At 0.2 sccm we observed a value of 20 mA/cm

2

, which dropped to 18.5 mA/cm

2

for 0.3 sccm and finally to 16 mA/cm

2

at 0.5 sccm. This shows a decrease in J sc as the doping increases. Increasing diborane flow from 0.2 sccm to a higher value implies a corresponding increase in hole concentration (i.e. the p-layer becomes heavily doped). However, the conductivity of the layer increases with diborane flow, but the reason the short-circuit current density drops is that the absorption of the material increases with diborane concentration). Another factor contributing to low J sc is the fact that at 0.5 sccm diborane flow, the p-layer is more amorphous; as a result, the p/i interface absorbs more light hence reducing the effective photons for current generation in the i-layer.

Therefore, the lower the diborane flow, the higher the short wavelength spectral response of the solar cell. This implies a higher J sc.

Furthermore, from the same figure the results of the fill factor and the efficiency of the solar cells are shown with respect to diborane flow of the microcrystalline p-layers. For diborane flow of 0.2 sccm, the FF is 62.4% while at 0.3 and 0.5 sccm, the FF gives 59 and 62.5% respectively.

The values are very close irrespective of the doping. The maximum conversion efficiency of 5.4

% is obtained for 0.2 sccm flow of diborane. This dropped to 5.25 % and 4.65 % for 0.3 sccm and 0.5 sccm flow of diborane respectively.

3.5 Effect of µc-Si:H p-layer deposition power on the performance of µc-Si:H solar cell.

The optimization process of the µc-Si:H p-layer continued with the variation of the deposition power at fixed diborane flow (0.2 sccm) and deposition time (300 seconds). With these values, the optimized thin film µc-Si:H p-layer was obtained with respect to our study. Other deposition parameters remained unchanged. The deposition power was varied between 30 to 80 W.

48

Transmittance of the deposited layers for the power series was investigated and the results obtained are shown in figure 3.9.

60

40

100

80

30 w

70 w

80 w

20

0

400 600 800

Wavelenght (nm)

1000

Figure 3.9 Transmittance in µc-Si:H p-layer deposited at varying deposition power .

It can be seen from figure 3.9 that there is not much variation in transmittance with increasing power. However, the 30 W sample has the highest transmittance and the 70 W the lowest in the long wavelenght region, so no real trend. At 400 nm, so much of the photons are absorbed (about

85%) in all the power series.

From figure 3.10, we see a trend in all the three plots. Activation energy and thickness increases with increasing power up to a certain value and dropped sharply at 80 W. The dark conductivity of the p-layer reduces with increasing power.

49

1e-5

1e-6

1e-7

1e-8

1e+1

1e+0

1e-1

1e-2

1e-3

1e-4

(a)

0.45

0.40

0.35

0.30

0.25

0.20

(b)

24

23

22

26

25

(c)

21

30 40 50

Power (W)

60 70 80

Figure 3.10 Thickness series of µc-Si:H p-layer deposited at 0.2 sccm diborane flow and 300 seconds deposition time

As in figure 3.10, the layer deposited at 30 W presents the best dark conductivity and high activation energy. Its transmittance is not better than the other power series. Looking further into the figure 3.10, the layer deposited at 70 W presents a much better results in the three plots with

50

reasonable activation energy and dark conductivity than the layer deposited at 80 W. The structural property of the p-layer with relation to deposition power can be expressed from figure

3.11. The trend shows an increase in crystallinity fraction with increased power. The layer deposited at 80 W has crystallinity volume fraction of about 63 %.

65

60

55

50

45

40

35

30

30 40 50 60 70 80

Power (W)

Figure 3.11 µc-Si:H p-layer structural properties dependence on deposition power

Generally, from the obtained results, we could conclude that an increase in deposition power results in increase in the crystalline volume fraction of the layer. The optical transmittance however does not change much with the deposition power. It is observed that the dark conductivity of the p-layer drops with increasing power. The unexpected behaviour can arise from the non-uniformity of the deposited layer at such high power as 80 Watt. To ascertain the material quality and the effect of the p-layers properties on solar cells, this layer was then applied to a single junction p-i-n type µc-Si:H solar cells with the same configurations and composition as in stated in section 3.4. The spectral response of the solar cell is shown in figure 3.12.

51

0.6

0.4

0.2

1.0

0.8

30 W

60 W

70 W

80 W

0.0

400 600 800

Wavelenght (nm)

1000

Figure 3.12 Spectra response of solar cells deposited at a fixed diborane flow (0.2 sccm) and deposition time (300 seconds).

The result shows an increase in quantum efficiency of the solar cell at 80 W deposition power over the solar cell in section 3.4. At 400 nm, the quantum efficiency value increased from about

38 % to about 62 % but there is no significant change in the long wavelength region.

Figure 3.13 shows the external parameters of a single junction p-i-n type µc-Si:H solar cells of players of different at 300 s deposition time and 0.2 sccm diborane flow. The solar cell has the highest efficiency at 80 W deposition power. The trend in this figure shows an increase in the

V oc

, J sc

, FF and efficiency with increasing deposition power except for the layer deposited at 70

W. The reason for this disparity could be as a result of low transmittance of light in this layer at long wavelenght region as already observed in figure 3.9.

52

0.56

0.54

0.52

0.50

0.48

0.46

0.44

5.2

5.0

4.8

4.6

4.4

4.2

4.0

3.8

3.6

3.4

22.0

21.5

21.0

20.5

20.0

19.5

0.43

0.42

0.41

30 40 70 80 50

Power (W)

60

Figure 3.13 µc-Si:H solar cells external parameters versus power at 0.2 sccm diborane flow and

300 s deposition time.

53

3.6 Conclusions

In line with the obtained results and the purpose of this experiment, we have succeeded in preparing an optimized µc-Si p-layer suitable for µc-Si:H solar cells. This was achieved by investigating the influence of deposition parameters of p-layer on TCO/P-layer configuration of microcrystalline silicon solar cells using a PECVD multi-chamber system at a frequency of 13.56

MHz .We varied the doping gas (diborane) and also made series of the p-layers at different deposition time and thicknesses. The deposition power was also optimized.

The results show that the layer deposited at 300 seconds with 0.2 sccm diborane flow has the optimum value. This layer is thin ~25 nm, sufficiently transparent and conductive. It gave a high

FF and reasonable V oc when applied to a single junction p-i-n type µc-Si:H solar cells. The efficiency of the solar cell was 5.4%.

Significant gain in quantum efficiency of the solar cell was recorded at the short-wavelength region. With the optimized p-layer and at 80 W deposition power, the quantum efficiency increased to about 65 % at 400 nm when compared to the obtained value of about 35% with the same optimized p-layer deposited at 60 W. Moreover, there is no significant change in the infrared region in the two cases. The overall results shows that spectral response is highly sensitive to diborane flow at short wavelength.

54

Chapter Four

Sensitivity study of µc-Si:H I-layer

4.0 Abstract

Systematic variation of µc-Si:H deposition parameters was done to study the sensitivities and the effects of these parameters on the material properties of µc-Si:H intrinsic layer. Series of i-layer depositions were made in this experiment in order to obtain the transition from the amorphous to crystalline phase materials with crystallinity volume fraction between 50 and 55%. The µc-Si:H intrinsic layers were deposited using PECVD multi-chamber system at a frequency of 13.56

MHz. The structural properties of the samples were investigated by Raman scattering spectroscopy while FTPS and FTIR measurements and analysis were done to ascertain the optoelectrical properties. The results obtained indicates µc-Si:H i-layer deposited at a low power but a higher pressure reveals layer with high photoresponse.

4.1 Introduction

The absorber layer is the heart of thin-film silicon solar cell. Here in this layer, absorption of photons required for effective generation of charge carriers takes place. However, µc-Si:H is known to be material with an indirect band gap coupled with a lower absorption coefficient

(lower than the a-Si:H) in the visible part of the solar spectrum. This implies that the layer has to be thick in order to obtain sufficient absorption of photon and photogeneration of charge carriers

[29, 34]. Therefore, thickness of the layer is one of the key factors contributing to the enhancement of the photogeneration process. Relatively, the intrinsic layer of the microcrystalline solar cell is required to have a film thickness at least five times greater than the

55

amorphous silicon solar cell. Typical thickness value of this layer is usually more than 1 µm [55] thereby making deposition rates an important factor. However, it was sugested [56] that cell performance deterioration is attributed to increase in grain size and microcrystalline volume fraction with thickness when cell performance and the microstructure were correlated. But cell performance increased as the intrinsic layer thickness increases by varying the hydrogen dilution in the gas mixtures during deposition, thereby controlling microstructure evolution [56].

In this research, we studied the effect of varying the rf power, silane concentration and deposition pressure in obtaining device grade µc-Si:H film and increasing the performance of

µc-Si:H solar cell.

4.2 Experimental Details

The layer deposition began with substrate preparations. Corning glass and silicon wafers were used as substrates for the layers and were deposited simultaneously. The glass was etched in a solution containing HF (40 %), HNO

3

(65 %) CH

3

COOH (100 %) and water in order to enhance adhesion of the µc-Si:H layers to the glass. FTIR analysis of the layers deposited on the silicon wafers were carried out. The FTIR analysis of most of the layers could not be carried out because the deposited layers peeled off the silicon wafer. This we attribute to stress in the material arising from the high pressure, high power deposition and the effect of post deposition oxidation. The crystalline volume fraction of the layers was extracted from the Raman measurements. The microstructure factor was estimated using the from FTIR spectra. All the deposited layers had a crystalline volume fraction of 50-55%, a range that we consider for films deposited at the amorphous-to- microcrystalline transition referred to hence forth as transition materials. The following deposition parameters were varied during the deposition in order to maintain the crystallinty as specified above: silane concentration (SiH

4

]/ [SiH

4

+H

2

), rf power (60-80 W), deposition pressure (7-11 mbar). The depositions were done at a fixed substrate temperature of

180

0

C and 8 mm electrode distance in chamber 4 of the AMOR deposition system.

The layers were characterized by measuring the following: RT using the Perkin-Elmer

Photospectrometer, Lambda 900 setup to obtain the nk data. This data is required in the OPTA program for fitting with the data from the FTPS measurement to obtain the thickness, Urbach energy, Tauc gap, E

04

gap and optical gap of the material. Activation energy measurement was

56

done to determine the position of Fermi-level while light and dark conductivity measurement was done to determine the photoresponse (σ ph

/σ d

) of the layers. To ascertain the structural composition of the deposited material, we carried out the FTIR measurement. This reveals the hydrogen contents and the microstructures of the layers.

4.3 Results and Discussion

4.3.1 Electrical measurements

Activation energy determination is one of the indicators for presence or absence of impurities in an intrinsic semiconductor material. The Fermi level is essentially halfway between the valence and the conduction bands. This applies to µc-Si:H intrinsic layer with band gap E g

, of 1.1 eV and the activation energy is half of this value ( 0.55 eV). This shows that an µc-Si:H intrinsic layer with higher value is unintentionally doped with impurities such as oxygen causing an upward shift of the Fermi level.

0.66

0.64

0.62

0.60

0.58

0.56

0.54

0.52

(a)

7 mbar

9 mbar

11 mbar

0.50

60 70 80 90 100

Power (Watt)

Figure 4.1 Activation energy of µc-Si:H i-layer at different deposition power

In figures 4.1, At 100 Watt, the activation energy between 0.52 and 0.60 eV is obtained for all the deposition pressures investigated. However, at 60 W power, the activation energy increases.

The materials for all the series show intrinsic property expecially at 100 W deposition power.

This can be linked to the increased crystallinity of the films as power increases. At higher power

57

the crystalline volume fraction increase implying a more microcrystalline film. On the other hand, at 60 W deposition power, the amorphous component of the film is increased hence an increase in the activation energy.

Photoresponse is defined as the ratio of the photoconductivity to dark conductivity, σ ph

/σ d

. It reflects the optoelectrical quality of a material to a large extent [49] and it is an important factor contributing to the overall efficiency of the solar cells. The photoresponse of the deposited µc-

Si:H i-layers are shown in the figures 4.2.

From figures 4.2, we see a correlation between deposition power, pressure and crystallinity with respect to photoresponse of the layer deposited at 60 Watt power and 11 mbar deposition pressure. From these graphs, it cannot be concluded how the deposition power and pressure affect the transition material.

2000

1800

1600

1400

1200

1000

800

600

400

200

0

60 70 80

(a)

90

7 mbar

9 mbar

11 mbar

100

Power (Watt)

Figure 4.2 Photoresponse of µc-Si:H i-layer at different deposition power

58

4.3.2 Material properties

The crystallinity volume fraction values for the µc-Si:H i-layers as in figure 4.2 is between 50 % and 55%. We see a steady increase in crystalline volume fraction with increasing power especially for the layer deposited at 11mbar deposition pressure.

56

(a)

55

54

53

52

51

7 mbar

9 mbar

11 mbar

50

60 70 80

Power (Watt)

90 100

Figure 4.3 Crystallinity of µc-Si:H i-layer at different deposition power

The trend shown in figures 4.4 indicates that the silane concentration increases with increasing deposition power and pressure in order to sustain the crystallinity of the transition materials between 50-55% . Increase in the silane concentration increase the amorphous fraction and increase in the deposition power increase the crystalline fraction. These two opposing effects ensure that the desired crystallinty is maintained. However, it is said in literature that for a fixed plasma power, a lower silane concentration is required at high pressure to achieve µc-Si:H growth [56].

59

2.4

2.2

2.0

1.8

1.6

7 mbar

9 mbar

11 mbar

1.4

1.2

60 70 80

Power (Watt)

90 100

Figure 4.4 Silane concentrations at different deposition power

Figure 4.5 b show the variation of the deposition rate with the deposition power at different pressure respectively.

0.75

0.70

0.65

0.60

0.55

0.50

0.45

0.40

7 mbar

9 mbar

11 mbar

0.35

60 70 80 90 100

Power (Watt)

Figure 4.5 Deposition rate of µc-Si:H i-layer versus deposition power and pressure.

60

Figure 4.5 shows an increase in deposition rate and layer thickness with increasing power, which may suggest that SiH

4

depletion occurs due to decomposition of SiH

4

molecules into SiH x radicals efficiently with increasing deposition power. The graphs also show an increase in deposition rate and layer thickness for the pressure series.

Deposition of high-rate µc-Si:H layer, using a SiH

4

–H

2

system, requires high flux of film precursors (such as SiH

3

) and sufficient flux of atomic H during monolayer growth [57]. The atomic hydrogen is generated from a dissociation of silane, whereas it recombines with silane as

H + SiH

4

H

2

+ SiH

3

Increasing the atomic hydrogen flux density on the growing surface then requires that the partial pressure of silane should therefore be lowered. This means that high hydrogen dilution or silane depletion is necessary for microcrystalline growth [58, 59]. This can be supported with the earlier observation made above in section 4.3.

However, (Menno van den Donker, 2007 et. al. showed that good microcrystalline silicon can be obtained without hydrogen dilution. The required atomic hydrogen is obtained by decomposition of silane. Obviously other issues play a role, in particular the back diffusion of SiH

4

, which is a specific problem of research reactors.)

Thicker material means increase in light absorption and photogeneration of charge carriers but charge carrier collection may be difficult because of the distance they have to travel before collection at the electrodes. They may be trapped in the defect states and recombine thereby reducing the conversion efficiency of the µc-Si:H solar cell. The deposition rate of about 0.38 nm/s and about 920 nm thickness was recorded for the layer deposited at 60 Watt and 11 mbar deposition pressure while the layer deposited at 9 mbar has about 0.45 nm/s with about 1050 nm thickness.

4.3.3 Absorption Coefficient and Defect Density.

The optical absorption of the µc-Si:H layers deposited at different power, pressure and silane concentrations are shown in figure 4.6. Generally, looking at the figure 4.6, at photon energy between 1.4 to 2.0 eV, the optical absorption coefficient α increases with increasing silane concentration due to gradual increase in the amorphous fraction of the materials. This is more

61

pronounced in the layers deposited at 100 W deposition power and all pressure series as the silane concentration increases. High defect density in the material is attributed to grain boundaries. At 11 mbar deposition pressure, 60 Watt power and silane concentration of about 1.3

%, α (0.8 eV) is the lowest, the value is approximately 1 cm

-1

. However, we see a strange result with the sample deposited at 100 W, 11 mbar deposition conditions. This could be attributed to measurement error.

Figure 4.6 can further be described in terms of deposition pressure. The sub gap absorption decreases with increasing pressure. This can attributed to the fact that at high deposition pressure, the material is more compact hence low defect density [60].

The defect density (N

d

) for µc-Si:H film is determined from the FTPS data fitted in the OPTA program with photon absorption coefficient (α) at 0.8 eV photon energy using equation 14 [61] .

T he proportionality constant is 1.7 × 10

17

cm

-2

.

ܰ

=

1.7 × 10

17

cm

-2

α (1.12 eV) (14)

T

he value of the optical absorption coefficient at a photon energy of 0.8 eV scales roughly with the density of dangling bonds and represents a figure of merit for the quality of a solar cell material [61].

62

1e+6

1e+5

7 mbar

1e+4

1e+3

1e+2

1e+1

1e+0

60 W

80 W

100 W

1e+5

9 mbar

1e+4

1e+3

1e+2

1e+1

1e+0

60 W

80 W

100 W

1e+5

11 mbar

1e+4

1e+3

1e+2

1e+1

1e+0

1e-1

0.8

1.0

1.2

1.4

Energy (eV)

1.6

60 W

80 W

100 W

1.8

2.0

Figure 4.6 Optical absorption of µc-Si:H transition layers.

4.3.4 Hydrogen Content and Microstructure

The hydrogen content, c

H

, and microstructures R

*

, were determined from the results obtained from FTIR analysis using equations 4, 5 and 6. The c

H

was estimated using the method of evaluation developed for hydrogenated amorphous silicon. At deposition power higher than 60

Watt, we did not notice any peak at 2100 cm

-1

(HSM) and 2000 cm

-1

(LSM) FTIR spectra. The reason for this is high oxygen contamination after deposition of the material. Also, the material

63

may have been stressed at this high deposition power giving room for cracks and voids thereby enhancing speedy post oxidation immediately it was brought out of the deposition system for

FTIR analysis.

However, the results obtained at 60 watt deposition power, shows that at 7 mbar deposition pressure, c

H

is the highest (6.9 %). The silane concentration, SC, in this layer is the highest in this series (about 1.7 %) while at 9 mbar, c

H

= 4.41 %, SC = 1.47 %. The last layer with deposition pressure of 11 mbar has c

H

= 2.28 % and SC = 1.3 %. We could report from the results that c

H

decreases with increasing deposition pressure and SC at fixed deposition power.

The layer deposited at 7 mbar deposition pressure is more amorphous than the 9 and 11 mbar series.

The microstructure factor, R

*

, increases with increase in deposition pressure. At 7 mbar deposition pressure, the R

*

= 0.06, while at 9 and 11 mbar deposition pressure, the values are 0.1 and 0.6 respectively. From the R

*

, values obtained, the layer deposited at 7 mbar has the lowest value. This could indicate that the material deposited at 7 mbar pressure is less porous. A low value of R

*

shows that small fraction of Si-H

2 bonds is in the material and this corresponds to a compact material. From the trend obtained, it can also be concluded that the decrease in R

* is in line with the decreasing hydrogen content in the high stretching mode. According to [63], R

*

of fully amorphous material is always lowest but it increases with deposition pressure. For a device quality, the criteria is that the R

*

value should be less than 0.1 [49].

4.4 Application to solar cells

In order to verify the effect of i-layer properties on solar cells, some solar cells were deposited after the maintenance of the PECVD deposition machine (AMOR machine) and the newly installed aluminum doped ZnO target for TCO deposition. The layers deposited at 60 Watt, 9 and

11 mbar pressures were applied to a single junction p-i-n type µc-Si:H solar cells. For purpose of comparison, we also included the layer deposited at 80 Watt, 7 mbar.

The p-layer used for the solar cell is based on the the optimized deposition parameters reported in chapter three of this work. Amorphous silicon n- layer with thickness of 20 nm was used in the solar cell. The structure of this solar cell is glass/ZnO/ µc-Si-p/ µc-Si-i / a-Si-n/Ag/Al.

Shown in figure 4.7 is the quantum efficiencies of the solar cells.

64

The external parameters of the layers are shown in table 4.1 and 4.2. Table 4.1 shows the best cells out of the thirty measured dots while 4.2 shows the external parameters obtained for average of ten best cells. We see from table 4.2 that at a high deposition power, silane concentration but low pressure, the lowest fill factor (FF) and efficiency (ŋ) values were obtained. The V oc

and J sc values in all the series are almost the same.

Table 4.1 Solar cells parameters obtained from the best cells.

Sample /

Dot

A3996 a_15

A3994 a_20

A3995 a_14

Power

(W)

60

60

80

Pressure

(mbar) SC (%) V oc

[V] J sc

(mA/cm

9

11

7

Table 4.2 A verage of best ten cells.

1.45

1.27

2.03

0.439

0.432

0.415

18

16.7

18.2

2

) FF[%]

0.601

0.585

0.516 ŋ [%]

4.75

4.22

3.89

Sample

A3995

A3996

A3994

Power

(W)

80

60

60

Pressure

(mbar)

7

9

11

SC (%)

2.03

1.45

1.27

V oc

[V]

J

0.4281

0.4458

0.4254 sc

(mA/cm

2

)

17.37

17.5

16.9

FF[%]

0.521

0.6025

0.5716 ŋ [%]

3.87

4.695

4.109

The layer deposited at 60 Watt, 9 mbar pressure has the highest efficiency value of 4.7 %. This is higher than the layer deposited at 11 mbar with efficiency of 4.2 %. However, the layer photoresponse is lower than the layer deposited at 11 mbar. Photoresponse is an important parameter in i-layer consideration of a solar cell and high pressure depletion method can reduce defects in the grain boundary region by an amount magnitude [16] due to the lower ion bombardment. The overall low efficiency values could be attributed to the low values of fill factor and V oc.

Moreover, as already indicated at the beginning of this section (4.4), recalibration and optimization of the deposition parameters after the AMOR's maintenance and the newly installed target will help to ascertain further increase the solar cell performance.

65

0.8

0.6

0.4

0.2

60 W, 11 mbar

80 W, 7 mbar

60 W, 9 mbar

0.0

400 600 800 1000

Wavelenght (nm)

Figure 4.7 Spectra response of solar cells as a function of i-layer deposition power (W), pressure

(mba) and silane concentration (SC) in percentage.

From figure 4.7, we see a slight difference in the spectral response of the solar cells. At wavelength region between 500 – 600 nm, the cell deposited at 9 mbar show higher quantum efficiency than the 11 mbar cell. The reason could be attributed to the crystalline volume fraction of the cells. The cell with 9 mbar deposition pressure has a higher crystalline volume fraction than that of 11 mbar.

4.5 Effect of µc-Si:H i-layer thickness on hydrogen content and microstructure

The sensitivity study progressed with further investigation of the layer thickness at 60 Watt deposition power and other deposition conditions as stated in section 4.2. Fresh depositions were made at 7, 9 and 11 mbar deposition pressures but lower thicknesses. The reason for the reduction in layer thickness is to have layers with reduced stress, cracks and voids, post oxidation effect and defect in the grain boundaries. These defects leads to high recombination of

66

charge carriers in the µc-Si:H i-layers thereby reducing the overall conversion efficiency of the solar cells.

0.20

0.18

0.16

0.14

0.12

0.10

0.08

0.06

0.04

0.02

7 8 9

Pressure (mbar)

10 11

Figure 4.8 Photoresponse of µc-Si:H i-layer at 60 Watt deposition power

From figure 4.8, we see a drastic drop in electrical properties compared to the earlier investigated layers. The drop could be associated to poor generation and collection of charge carriers due to low absorption of photons. This is also evident in the microstructure and hydrogen content values.

Table 4.3 R

*

and c

H

values at 60 Watt deposition power

Deposition

Pressure

(mbar)

7

9

11

C

H

6.9

(%)

First series

4.41

2.28

C

H

0.7

0.7

0.6

(%)

Second series

R

*

(%)

First series

0.06

0.1

0.6

R

*

(%)

Second series

0.46

0.9

0.9

Thickness

(nm)

First series

1042

1117

925

Thickness

(nm)

Second series

561

340

370

There is no clear dependence of the microstructure on the hydrogen content of the layers. The

FTIR analysis of these series shows a much higher R

*

values but lower hydrogen content. At 7 mbar deposition power, the R

*

= 0.46, c

H

= 0.7 % while at 9 and 11 mbar, the values are 0.7, 0.9

% and 0.6, 0.9 % respectively.

67

From table 4.3, we see that the thicker layer has more c

H

values than the less thick layers at the same deposition conditions. The R

*

also show a similar trend. The second series show a more amorphous properties than the first in terms of the R

*

and c

H

.

The crystalline volume fractions of the 60 Watt deposition power layers with the earlier deposited series is 51 % in all the deposition pressure series. We compared the crystalline volume fraction of the layer deposited on silicon wafers and on glass and it was discovered that there is no much disparity in the values. This is important to know since the growth of microcrystalline material is sensitive to the substrate material.

4.6 Conclusions

With respect to the purpose of this study and the results obtained, an optimized µc-Si:H i-layer was prepared using PECVD multi-chamber system at a frequency of 13.56 MHz. We discorvered that the deposited µc-Si:H i-layer transition material at 60 Watt, 11 mbar and silane concentration of 1.27 %, reveals layer with relatively high photoresponse . An efficiency of 4.1

% was recorded when this layer was incorporated into a microcrystalline silicon solar cells.

68

Chapter Five

CONCLUSIONS AND RECOMMENDATIONS

Hydrogenated microcrystalline silicon, p and i-layers were grown in this study using the plasma enhanced chemical vapour deposition method. The layers were fully characterrized for thier structural, optical and electrical properties in the Laboratory of Photovoltaic Materials and

Devices, TuDelft. We optimized the p-layer of µc-Si:H by optimizing the rf PECVD deposition parameters. The parameters investigated includes the doping, layer thickness and deposition power. The results show that the layer deposited at 300 seconds with 0.2 sccm diborane flow has the optimum performance parameters in terms of the crystallinty, transmittance and conductivity.

This layer is thin ~25 nm, it gave a high FF and V oc when applied to single junction p-i-n type

µc-Si:H solar cells. The efficiency of the solar cell was 5.4%. Significant gain in quantum efficiency of the solar cell was recorded at the short-wavelength region. With the optimized player and at 80 W deposition power, the quantum efficiency increased to about 65 % at 400 nm when compared to the obtained value of about 35% with the same optimized p-layer deposited at

60 W. Moreover, there is no significant change in the infrared region in the two cases. The

69

overall results shows that the short wavelength spectral response of µc-Si:H is highly sensitive to p-layer doping and deposition power.

The sensitivity of the µc-Si:H intrinsic layer deposition parameters to the amorphous to crystalline phase materials properties was investigated. The transition materials had crystalline volume fraction between 50 and 55%. The structural properties of the samples were investigated by Raman scattering spectroscopy while FTPS and FTIR measurements and analysis were done to ascertain the optoelectrical properties. The results obtained indicates µc-Si:H i-layer deposited at a low power but a higher pressure reveals layer with high photoresponse.

For further improvement of this study, we will like to propose that an optimized and good quality TCO should be carried out. This is very important for the electrical properties of the solar cell. We recommend that further investigation be done on the effect of substrate temperature.

We also like to recommend a study on the correlation between electronic transport properties and the microstructure of hydrogenated microcrystalline silicon layers material properties of microcrystalline silicon can be studied.

70

References

1. Zoe Noonan, The depletion of fossil fuels, www.annesley.sa.edu.au/amep.

2. M. Zeman, Lecture notes on Solar Cells, Delft University of Technology, (2008)

3. Shriniwas Surendra Nayak, B.E. Thermal Imagery and Spectral Reflectance Based

System to Monitor Crop Condition December 2005, Unpublished master thesis

Submitted to the Graduate Faculty of Texas Tech University.

4. Roger Messenger, D. Yogi Goswami, Photovoltaic Fundamentals, Technology and

Application. Boca Raton: Taylor and Francis group, 2007.

5. D. Yogi Goswami, Frank Kreith, Energy Conversion. Boca Raton: Taylor and Francis group, 2008.

6. Roger Messenger, Jerry Venture, Photovoltaic System Engineering. Boca Raton: Taylor and Francis group, 2005.

7. J. Hüpkes, J. Müller, B. Rech, Transparent Conductive Zinc Oxide — Basics and

Applications in Thin Film Solar Cells, in: K. Ellmer, A. Klein, B. Rech (Eds.), Springer

Series in Materials Science, vol. 104, Springer, Berlin, 2008, p. 359.

8. David E. Carlson, Amorphous- Silicon Solar Cells, IEEE Transactions on Electron

Devices, Vol. 36, No 12, December 1989.

9. A. V. Shah, H. Schade,M. Vanecek, J. Meier, E. Vallat-Sauvain, N.Wyrsch, U. Kroll, C.

Droz, J. Bailat, Thin-film Silicon Solar Cell Technology Progress in Photovoltaics:

Research and Applications, Vol 12, pp. 113-142, 2004.

10.

Kasturi L. Chopra, Suhit Ranjan Das,

Thin film solar cells. New York, Springer, 1983.

11. B. Rech, T. Repmann, M.N. van den Donker, M. Berginski, T. Kilper, J. Hupkes, S.

Calnan, H. Stiebig, S. Wieder, “Challenges in microcrystalline silicon based solar cell technology”, Thin Solid Films, 511-512, p.548-555, (2006).

12. M. A. Green, Solar Cells - Prentice Hall, 1982.

13. Arvind V. Shah, MilanVaněček, Johannes Meier, Fanny Meillaud, Joelle Guillet , Diego

Fischer, Corinne Droz, Xavier Niqille, Sylvie Faÿ, Evelyne Vallat-Sauvain, Vanessa

Terrazzoni-Daudrix and Julien Bailat - Basic efficiency limits, recent experimental results and novel light-trapping schemes in a Si:H and ‘micromorph tandem’ solar cells -

Journal of Non-Crystalline Solids, 2004.

14. N. Pellaton Vaucher, J.-L. Nagel, R. Platz, D. Fischer and A. Shah, Proc. of the 2nd

WCPEC, Vienna, Austria, July 6–10

(1998), p. 728.

15.

Richard H. Bube, Photovltaic materials, Imperial college, 1998.

16. M. Kondo, T. Matsui, Y. Nasuno, H. Sonobe and S. Shimizu, Thin Solid Films, Volume

501, Issues 1-2, 20 April 2006, Pages 243-246.

17. B. Rech, T. Roschek, T. Repmann, J. Müller, R. Schmitz and W. Appenzeller, Thin Solid

Films

427 (2003), p. 157

18. K. Yamamoto, M. Toshimi, T. Suzuki, Y. Tawada, T. Okamoto and A. Nakajima, Proc.

Mater. Res. Soc. Symp. 507 (1998) p.131.

19. J. Meier, P. Torres, R. Platz, S. Dubail, U. Kroll, J.A. Anna Selvan, N. Pellaton Vaucher,

Ch. Hof, D. Fischer, H. Keppner, A. Shah, K.-D. Ufert, P. Giannoulès and J. Koehler. In:

(ed. 8),Mater. Res. Soc. Symp. Proc. 420 (1996), p. 3

20. H. Sakata, T. Nakai, T. Baba, M. Taguchi, S. Tsuge, K. Uchihashi & S. Kiyama, Proc.

28th IEEE PVSC, Anchorage 15-22 Sept. (2000).

21. J. Melskens, Msc Thesis, Delft University of Technology (2007).

71

22.

E. Katsia, E. Amanatides, D. Mataras and D.E. Rapakoulias Effect of plasma parameters on the amorphous to microcrystalline silicon transition “Thin Solid Films, 511-512 285

(2006)

23.

T. Moustakas, in: J. Pancove (Ed.), Semiconductors and Semimetals 21A, Academic

Press, New York, 1984, p. 55.

24. T. Saito, S. Muramatsu, T. Shimada, M. Migitaka, Appl. Phys. Lett. 42 (1983) 678.

25. T. Fuyuki, K.-Y. Du, S. Okamoto, S. Yasuda, T. Kimoto, M. Yoshimoto, H. Matsunami,

J. Appl. Phys. 64 (1988) 2380.

26. Y. Mishima, M. Hirose, Y. Osaka, K. Nagamine, Y. Ashida, K. Isogaya, Jpn. J. Appl.

Phys. 22 (1983) L46.

27. Michio Kondo, Microcrystalline materials and cells deposited by RF glow discharge.

28. Fischer, A. Shah, in: Tech. Digest of Int. PVSEC-11, Sapporo, Hokkaido, Japan, 1999, pp. 221.

29. A.V. Shah, J. Meier, E. Vallat-Sauvain, N. Wyrsch, U. Kroll, C. Droz and U. Graf,

Material and solar cell research in microcrystalline silicon, Solar Energy Mater. Solar

Cells

78 (2003), p. 469.

30. Swati Ray, Chandan Das, Sumita Mukhopadhyay, S.C. Saha. Substrate temperature and hydrogen dilution: parameters for amorphous to microcrystalline phase transition in silicon thin films, Solar energy materials and solar cells, 2002, vol. (4 ref.), pp. 393-400,

0927-0248

31. U. Kroll, J. Meier, A. Shah, S. Kikhailov and J. Weber, Hydrogen in amorphous and microcrystalline silicon films prepared by hydrogen dilution, J. Appl. Phys. 80 (1996), p.

4971

32. L. Houben, M. Luysberg, P. Hapke, R. Carius, F. Finger and H. Wagner, Structural investigations of microcrystalline silicon in the transition from highly crystalline to amorphous growth, Phil. Mag. A 77 (1998) (6), p. 1447.

33. B. Rech, T. Roschek, J. Müller, S. Wieder and H. Wagner, Amorphous and microcrystalline silicon solar cells prepared at high deposition rates using RF, sol

(13.56 MHz) plasma excitation frequency, Energ. Mater. Solar Cells 66 (2001), p. 267

34. T Roschek, T. Repmann, J. Müller, B. Rech, and H. Wagner, Comprehensive study of microcrystalline sillicon solar cells deposited at high rate using 13.56MHz plasmaenhanced chemical vapour deposition, J. Vac. Sci. Technol. A Volume 20, Issue 2, pp.

492-498 (March 2002)

35. J. Krc, M. Zeman, A. Campa, F. Smole and M. Topic. Novel approaches of light management in thin-film silicon solar cells - Material Research Society, 2006.

36. Light Scattering at Rough Interfaces of Thin Film Solar Cells to Improve the Effciency and Stability - Roelof Schuitema, Wim Metselaar and Miro Zeman - IEEE/ProRisc, 1999.

37. J. Meier, J. Meer J.Spitznagel, U.Kroll, C. Bucher, S. Fay, T. Moriarty, A. Shah,

Potential of amorphous and microcrystalline silicon solar cells, Thin Solid Films Volumes

451-452, 22 March 2004, Pages 518-524

72

38. M.A. Green, K. Emery, D.L. King, Y. Hishikawa and W. Warta, Prog. Photovoltaics:

Res. Appl.

15 (2007), p. 35

39. A.Pawlikiewicz and S.Guha. Proc. Mat Res. Soc. Symp. 118 (19880 599.

40. A.H. Mahan, B.P. Nelson, S. Salamon, R.S. Crandall, Mater. Res. Soc. Proc. 219 (1991)

673.

41. J. Meier, E. Vallat-Sallvain, S. Dubail, U. Kroll, J. Dubail, S. Golay, L. Feitknecht, P.

Torres, D. Appl. Phys. Lett. 65, p. 860 (1994).

42. L. Raniero, N. Martins, P. Canhola, S. Zang, S. Pereira, I. Ferreira, E. Fortunato, R.

Martins. Influence of the layer thickness and hydrogen dilution on electrical properties of large area amorphous silicon p-i-n solar cell, Solar energy materials and solar cells (2005), vol. 87, pp. 349-355, 0927-0248

43. Thin Film Solar Cells. J.Poortmansad V.Arkhipov, 2006.

44. J. Mullerova, P. Sutta, G. Van Elzakker, M. Zeman, M. Mikula, Microstructure of hydrogenated silicon thin films prepared from silane diluted with hydrogen, Applied

Surface, Science Volume 254, Issue 12, 15 April 2008, Pages 3690-3695

45. K.W. BOER. Survey of Semiconductor Physics, Vol. II: Barriers, Junctions, Surfaces and

Devices. Van Nostrand Reinhold, New York (1995).

46. C. Droz, E. Vallat-Sauvain, J. Bailat, L. Feitknecht, J. Meier and A. Shah, Relationship between Raman crystallinity and open-circuit voltage in microcrystalline silicon solar cells, Solar Energy Materials and Solar Cells, Volume 81, Issue 1, 25 January 2004,

Pages 61-71

47. A. H. M. Smets, W. M. M. Kessels, and M. C. M. van de Sanden, J. Appl. Phys,

102(2007) 073523

48. U. Kroll, J. Meier, A. Shah, S. Mikhailov, J. Weber, J. Appl. Phys. 80 (1996) 4971.

49. R. E. I. Schropp and M. Zeman, ‘New development in Amorphous Thin- Film Silicon

Solar cells’, IEE Trans. On Electron Devices, 46 (10). (1999).

50. US Patent 5661556 - System for measuring the total integrated scatter of a surface

S Patent Issued on August 26, 1997.

51. S. Jimenez-Sandoval, Micro-Raman spectroscopy: a powerful technique for materials research Microelectronics Journal ,Volume 31, Issue 6, 30 June 2000, Pages 419-427

52. J.Mullerova, P. Sutta, G. Van Elzakker, M. Zeman, M. Mikula, Microstructure of hydrogenated silicon thin films prepared from silane diluted with hydrogen, Applied

Surface Science

Volume 254, Issue 12, 15 April 2008, Pages 3690-3695

53. C. Smit, R.A.C.M.M. van Swaaji, H.Donker, A.M.H.N. Petit, W.M.M.Kessels and

M.C.M van de Sanden, Determining the material structure of microcrystalline sillicon from Raman Spectra, Journal of Applied Physics, Vol. 94, No. 5. (2003), pp. 3582-3588

54. Paula C.P. Bronsveld, Arjan Verkerk, Tomas Mates, Antonin Fejfar, Jatindra K. Rath,

Ruud E.I. Schropp, Relation between electronic properties and density of crystalline agglomerates in microcrystalline silicon, (Mater. Res. Soc. Symp. Proc. Vol. 989,

Warrendale, PA, 2007), 0989-A07-01

55. T. Matsui, M. Kondo and A. Matsuda, Proceedings 3rd World Conference Photovoltaic

on Solar Energy Conversion

(2003), p. 1570.

56. J.K. Rath, R.H.J. Franken, A.Gordjin, R.E.I. Schropp and W.J. Goedheer, Growth mechnism of microcrystalline silicon at high pressure conditions, Journal of Non-

Crystalline Solids

Volumes 338-340, 15 June 2004, Pages 56-60

73

57. Guo Lihui, Lin Rongming, Studies on the formation of microcrystalline silicon with

PECVD under low and high working pressure, Thin Solid Films, vol. 376, issue 1-2, pp.

249-254

58. Kondo M, Fukawa M, Guo L, et al. High rate growth of microcrystalline silicon at low temperatures. J Non-Crystal Solids, 2000, 266–269: 84–89

59. L. Guo, M. Kondo, M. Fukawa, K. Saito and A. Matsuda, High rate deposition of microcrystalline silicon using conventional plasma-enhanced chemical vapor deposition,

Jpn. J. Appl. Phys.

37 (1998), pp. L1116–L1118

60. Guha and J. Yang, Microcrystalline Silicon Solar Cells, Final Technical Progress Report

1 July 2001 – 31 August 2004.

61. G. Elzakker, V. Nadazdy, F.D Tichelaar, J.W. Metselaar, M. Zeman, Analysis of structure and defects in thin silicon films deposited from hydrogen diluted silane, Thin

Solid Films

, vol. 511-512 (2006) pp. 252-257.

62. S.K. Ram, Ph.D. thesis, I.I.T. Kanpur, India, (2006).

63. Stefan Klein, Friedhelm Finger, Reinhard Carius, Martin Stutzmann, Deposition of microcrystalline silicon prepared by hot-wire chemical-vapour deposition: The influence of the deposition parameters on the material properties and solar cell performance, Thin

Solid Films

395 (2001), 305.

64. R. Jimenez Zambrano, R.A.C.M.M. van Swaaij, M.C.M. van de Sanden, Optimisation of

Microcrystalline Silicon Deposited by Expanding Thermal Plasma Chemical Vapor

Deposition for Solar-Cell Application, (Mater. Res. Soc. Symp. Proc. Vol. 989,

Warrendale, PA, (2007), 0989-A07-02.

65. B. Strahm, A.A. Howling, L. Sansonnens and Ch. Hollenstein, Plasma silane concentration as a determining factor for the transition from amorphous to microcrystalline silicon in SiH4/H2 discharges, Plasma Sources Science and technology

16 (1) (2007), pp. 80–89.

74

Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement