<%BANNER%>

Zinc Oxide Sputter Deposition and Modeling of Copper-Indium-Gallium-Diselenide-Based Thin Film Solar Cells

University of Florida Institutional Repository
Permanent Link: http://ufdc.ufl.edu/UFE0019981/00001

Material Information

Title: Zinc Oxide Sputter Deposition and Modeling of Copper-Indium-Gallium-Diselenide-Based Thin Film Solar Cells
Physical Description: 1 online resource (138 p.)
Language: english
Creator: Liu, Wei
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Highly transparent and conductive aluminum doped zinc oxide (AZO) thin films were successfully obtained through RF magnetron sputtering using argon as the sputtering gas. Thin film AZO grown under different conditions such as base pressure, deposition power and working pressure was characterized. A correlation between various operational parameters and the optical and electrical properties of AZO was developed and used to suggest optimum of operational conditions. The sputtering process was improved by adding a small amount of hydrogen into the sputtering gas. A significant improvement of the conductivity of AZO thin films was observed. This observation together with results obtained under different base pressures supports the literature prediction that hydrogen plays a favorable role in n-type zinc oxide thin films. Hydrogen participates in the doping process and causes increased carrier concentration. The AZO thin films sputtered with the Ar and H2 mixture gas were incorporated in copper-indium-gallium diselenide (CIGS) solar cells and devices with a thin film layer structure of the form ZnO/CdS/CIGS/Mo were fabricated. It is found that the sheet resistance and thickness of the AZO layer can be an important factor determining the performance of CIGS solar cells. The devices were measured under AM 1.5 radiation and a conversion efficiency of approximately 9% was achieved. To better understand solar cell devices fabricated using chalcopyrite semiconductors such as CuInSe2 and CuGaSe2, an inverse modeling process was developed. The inverse modeling process was realized by integrating two software tools, namely Matlab and Medici, under a Linux environment. Matlab provides the values of candidate input parameters to Medici, which in turn outputs device simulation results that are fed back to Matlab. This process continues until the outputs reach the target values. The inverse modeling process successfully identified the defect distribution in a CIGS layer that matched a known cell performance. In an ensuing optimization step, optimal defect concentrations were found, revealing that the shallow acceptor defect concentrations are favorable for increasing the efficiency.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Wei Liu.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Crisalle, Oscar D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0019981:00001

Permanent Link: http://ufdc.ufl.edu/UFE0019981/00001

Material Information

Title: Zinc Oxide Sputter Deposition and Modeling of Copper-Indium-Gallium-Diselenide-Based Thin Film Solar Cells
Physical Description: 1 online resource (138 p.)
Language: english
Creator: Liu, Wei
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Highly transparent and conductive aluminum doped zinc oxide (AZO) thin films were successfully obtained through RF magnetron sputtering using argon as the sputtering gas. Thin film AZO grown under different conditions such as base pressure, deposition power and working pressure was characterized. A correlation between various operational parameters and the optical and electrical properties of AZO was developed and used to suggest optimum of operational conditions. The sputtering process was improved by adding a small amount of hydrogen into the sputtering gas. A significant improvement of the conductivity of AZO thin films was observed. This observation together with results obtained under different base pressures supports the literature prediction that hydrogen plays a favorable role in n-type zinc oxide thin films. Hydrogen participates in the doping process and causes increased carrier concentration. The AZO thin films sputtered with the Ar and H2 mixture gas were incorporated in copper-indium-gallium diselenide (CIGS) solar cells and devices with a thin film layer structure of the form ZnO/CdS/CIGS/Mo were fabricated. It is found that the sheet resistance and thickness of the AZO layer can be an important factor determining the performance of CIGS solar cells. The devices were measured under AM 1.5 radiation and a conversion efficiency of approximately 9% was achieved. To better understand solar cell devices fabricated using chalcopyrite semiconductors such as CuInSe2 and CuGaSe2, an inverse modeling process was developed. The inverse modeling process was realized by integrating two software tools, namely Matlab and Medici, under a Linux environment. Matlab provides the values of candidate input parameters to Medici, which in turn outputs device simulation results that are fed back to Matlab. This process continues until the outputs reach the target values. The inverse modeling process successfully identified the defect distribution in a CIGS layer that matched a known cell performance. In an ensuing optimization step, optimal defect concentrations were found, revealing that the shallow acceptor defect concentrations are favorable for increasing the efficiency.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Wei Liu.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Crisalle, Oscar D.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0019981:00001


This item has the following downloads:


Full Text





ZINC OXIDE SPUTTER DEPOSITION AND MODELING OF
COPPER-INDIUM-GALLIUM-DISELENIDE-BASED THIN FILM SOLAR CELLS




















By

WEl LIU













A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA

2007


































O 2007 Wei Liu



































To my beloved parents and family









ACKNOWLEDGMENTS

Dr. Crisalle is the chair of my supervisory committee and I express my sincere

appreciation for his guidance and constant encouragement through the last Hyve years. I am also

obliged to thank Dr. Hoflund, Dr. Svoronos and Dr. Eisenstadt for their valuable discussions and

serving as my research committee members. Of course my gratitude should be given to Dr.

Holloway for his supports and suggestions based on his profound knowledge on vacuum science.

Dr. Anderson, Dr. Li also provided a lot of support during the last Hyve years and I would like to

express my appreciation. I also thank Dr. Craciun for his valuable discussions on thin film

characterization.

This work would not have been achieved without the collaborations from my colleagues

working on the CIGS solar cell projects at University of Florida. In particular, I hereby extend

my thanks to Xuege Wang, Ryan Kaczynski, Lei Li, Ryan Acher and Andre Baran for their

valuable discussions and collaborations on device fabrication. Dr. Mark Davidson, Evan Law

and Dr. Lei Qian from Dr. Holloway's group also provided their kind helps at different points of

this work. I would like to show my sincere thanks to them.

Of course the acknowledgement would never be complete without expressing my gratitude

to my parents. I own everything to them and the word is never enough to express my

appreciation. Without their support and encouragement, I would not have made it to this point of

my life.












TABLE OF CONTENTS


page

ACKNOWLEDGMENTS .............. ...............4.....


LIST OF TABLES ......_. ................ ...............8.......


LIST OF FIGURES .............. ...............9.....


AB S TRAC T ............._. .......... ..............._ 12...


CHAPTER


1 INTRODUCTION ................. ...............14.......... ......


2 LITERATURE REVIEW ................. ...............18................


Solar Light ................... .......... ...............18.......
The Photovoltaic Effect ................. ...............19........... ....
History of Photovoltaic Technology ................. ...............19........... ...
Device Physics............... ....... ..............2
Solar Cells using Different Materials .............. ...............22....
Crystalline Silicon Solar Cells............... ...............23.
Single crystalline silicon ................. ...............23........... ....
Polycrystalline silicon............... ...............24
Thin Film Solar Cells .............. ...............25....

Amorphous silicon............... ...............25
CulnSe2 and CuGaSe2 .............. ...............27....
Transparent Conductive Oxide ................. ...............28....... ......
Sputter Deposition .............. ...............30....
Introduction .............. ..... ... ..............3
Advantages of Sputter Deposition ................. ...._ ....___ ............3
Sputter Deposition of ZnO ............ ..... __ ...............32..
Summary ............ .... __ ...............33..


3 EXPERIMENTAL METHODS .............. ...............40....


ZnO Sputtering System............... ...............40.
Sub state Cl meaning ............ ...... __ ...............42..
Characterization Techniques .............. ...............43....
Electrical Property Characterization............... ............4
Four-point probe............... ...............43.
Hall measurement............... .... ..............4

Optical Spectrophotometer Characterization .............. ...............45....
Other Characterizations ........._.._ ..... ..__ ...............46....
Profilom etry .............. ...............46....
X-Ray Diffraction .............. ...............47....












Atomic Force Microscopy ........._. ....... .__ ...............49....
Auger Electron Spectroscopy ................. ...............51.........._ ....

4 GROWTH AND CHARACTERIZATION OF ZINTC OXIDE ................. ......................56


Optimization of Sputter Deposition of Aluminum Doped ZnO ............... ...................5
Experiments at Different Base Pressures .............. ...............56....
Optical property............... ...............57
Electrical property ................. ...............58.......... .....
Growth rate............... ...............58..
Power Effect ................... ...............59.......... .....
Working Pressure Effect............... ...............61.
Characterization ................. ............ ................62......

Auger Electron Spectroscopy Analysi s ................ ...............62........... ...
Surface scan............... ...............63..
Al sensitivity .............. ...............64....

Depth profile .............. ...............65....
Atomic Force Microscopy ................. ...............65................


5 EFFECT OF HYDROGEN ON SPUTTERED ZINC OXIDE ................ ......................78


Motivation from the Base Pressure Effect ................. ...............78...............

Experiments and Results............... ...............78
Hall Characterization............... ............7
AFM and SEM Characterization ................. ...............80......__._....
Film Uniformity Study .............. ...............81....
Sum m ary ................. ...............83........ ......


6 DEVICE FABRICATION AND CHARACTERIZATION .........__. ........_. ..............92


Device Fabrication Procedure............... ...............9
Substrate ................ ...............92...

Back Contact Sputtering ............_ ..... ..__ ...............92...
Absorber Growth ............_...... ._ ...............93....

Buffer Layer Deposition ............. ...... ._ ...............94....
Window Layer Deposition .............. ...............95....
M etal liz ati on ................... ...............96......... ...
Device Characterization System .................. ......._. ........_.._.........9
Devices with Ab sorbers from UF PMEE Reactor ................. ...............97..............
CIGS Devices .............. ...............97....
CGS Devices .............. .......... .. ................9
Effect of Zinc Oxide Window Layer on Device Performance ................. ............ .........99
Effect of Window Layer Transmission ............__......___....._ ............0
Effect of Sheet Resistance ............_...... ._ ...............101..

Impact of the Intrinsic ZnO Layer. ............ .....___ ...............102
Sum mary ............. ...... ._ ...............102....












7 INVERSE MODELING OF CIGS SOLAR CELLS ................. ............... ......... ...110


Definition of Inverse Modeling ................. ...............110......... .....
Inverse Modeling for Photovoltaic Cells ............................... ................. 111...
Application of Inverse Modeling Performance Optimization ................. ................ ...112
Approach................ ...............11
For ward M odeling ................. ................. 113........ .....
Cell structure generation ................. ...............113...............
Input param eters ................... ...............114......... ......
Example of a forward modeling ................. ...............114........... ...
Modeling of NREL champion device .........___....... ......__...........1
Inverse Modeling............___......... ... .. .....___..........11
Inverse Modeling of NREL champion device ......____ ...... .. .................1 17


8 CONCLUSION............... ...............13


Parameter Effects on Sputtered AZO .............. ...............130....
Device Fabrication ..........._.... ...............13 1......_.......
Defects in CIGS Absorbers .............. ...............132....
Future Work............... ...............132.


LIST OF REFERENCES ............._... ...............134...._.........


BIOGRAPHICAL SKETCH ............._.. ......._..._ ............ 138..










LIST OF TABLES


Table page
4-1 Estimated grain size using XRD data obtained for AZO deposited at different power.....77

4-2 Parameters used for collecting the Auger depth profie of AZO thin fi1ms.............._....__...77

4-3 AFM surface roughness evolution with working pressure. ............. .....................7

5-1 Growth parameters comparison for AZO and H-AZO thin films............... ..................9

5-2 Deposition condition comparison between two AZO thin films shown in Figure 5-5......91

5-3 Thickness, Rq and estimated grain size comparison between two AZO thin films
shown in Figure 5-5. ............. ...............91.....

5-4 Roughness data evolution with increasing film thickness. ................ ..................9

5-5 Hall measurements results on two films with different positions under the target..........9 1

5-6 Comparison of Hall and four-point-probe measurement results. ............. ...............91

6-2 Recipe for CdS buffer layer deposition. ............. ...............104....

6-3 Composition and thickness data for UF CIGS absorbers. ............. ......................0

6-4 Calibration of the reference cell. ................ ................ ..........................104

6-5 I-V character of the UF CIGS devices ................. ...............105.............

6-6 Growth recipe and composition for UF CGS absorbers. ................ .................10

6-7 I-V character of the UF CGS devices. ................ ...............105.............

6-8 Performance of CIGS solar cells with different window layer transmission. ..................1 05

6-9 Performance of CIGS solar cells with and without i-ZnO layer............... ..................105

7-1 Typical input parameters for Medici ................. ...............129..............










LIST OF FIGURES


Figure page

2-1 Energy current density per wavelength of AMO spectrum(heavy line) compared with
blackbody radiation at 5800K(thin line). .............. ...............34....

2-2 Schematic of the Air Mass definition. ................ ...........__........ ......__.......34

2-3 Energy current density per photon energy of AM1.5 spectrum (heavy line)
compared with blackbody radiation at 5800K (thin line). ................ ..................3

2-4 Formation of built-in potential between p-n junction. ................. ....__. ................3 5

2-5 An equivalent circuit of a typical solar cell. ............. ...............36.....

2-6 Effect of series resistance and shunt resistance on solar cell performance..............._._. ....36

2-7 Effect of series resistance on solar cell performance. ....._.__._ .... ... ._ .................3 7

2-8 World PV market share by technology in 2002............... ...............38..

2-9 Development hi story of thin film solar cells. (Courtesy of NREL) .............. .........._....3 8

2-10 Thin film CIGS solar cell structure................ ...............3

3-1 Perkin-Elmer 4400 sputtering system. .............. ...............53....

3-2 Schematic of the ZnO Sputtering System (Perkin-Elmer 4400) ................. ................. .53

3-3 Schematics of four-point probe measurement. .........___ ..... .___ ......_.............54

3-4 Square shaped van der Pauw sample geometry. ............. ...............54.....

3-5 Generation of Cu Koc radiation. ............. ...............55.....

3-6 Bragg' s condition. ........._.._.. ...._... ...............55...

4-1 The effect of base pressure on the optical transmission of sputtered AZO films. .............68

4-2 Effect of base pressure on the resistivity of sputtered AZO films .................. ...............68

4-3 The effect of base pressure on the growth rate of ZnO:Al thin films (Fixed
conditions: Dts = 5 cm, P = 4 mTorr, Power = 400W) ................. .......... ...............69

4-4 The effect of power on the resistivity of as grown ZnO:Al thin films (Fixed
conditions: Dts = 5 cm, Pb = 8.0 x107 Torr, and P = 4 mTorr)............... .................6










4-5 The effect of power on the growth rate of ZnO:Al thin films (Fixed conditions: Dts =
5 cm, Pb = 8.0 x10-7 Torr, and P = 4 mTorr). ............. ...............70.....

4-6 X-ray diffraction pattern of thin films deposited from a reduced ZnO target .................. .70

4-7 X-ray diffraction pattern of thin films deposited from an oxidized ZnO target ................71

4-8 High resolution XRD pattern for AZO films deposited using different power (Fixed
conditions: Dts = 5 cm, Pb= 8.0 x 10-7 Torr, Pressure 4mTorr). ................ ................. .71

4-9 The effect of working pressure on the resistivity of ZnO:Al thin films ( Fixed
conditions: Dts = 5 cm, Pb 8 .0 x10-7 Torr, and Power = 500 W ) ................. ................72

4-10 The effect of working power on the growth rate of ZnO:Al thin films ( Fixed
conditions: Dts = 5 cm, PbO 8.0 x10-7 Torr, and Power = 500 W ) ................. ................. 72

4-11 High resolution XRD pattern of AZO thin films deposited at different working
pressures (Fixed conditions: Dts = 5 cm, Pb=4.0x 10-7 Torr, Power 500 W). ................. .73

4-12 AES survey scan pattern for an AZO thin film with organic contamination on the
surface. .............. ...............73....

4-13 AES scan results after sputtering for 3 minutes. .....__._.. .... .._.... ......_... .......7

4-14 Al Auger peak shown by using increased primary beam energy ........._...... ........._.......74

4-15 Depth profile for each element of AZO obtained with AES. ............. .....................7

4-16 AFM graph of AZO thin films deposited with different working pressure. a) 2mTorr,
Rq=2.2nm; b) 3mTorr, Rq=2.8nm; c) 4mTorr, R,=3 .6nm. ........._.._... .......__. ........._.76

5-1 Effect of inj ecting hydrogen gas on the resistivity of sputtered AZO films....................85

5-2 Hall measurement results for aluminum-zinc-oxide thin films deposited using only
argon as the working gas (sample labeled as AZO) and using a mixture of argon
with 0. 1 wt%/ of hydrogen (sampled labeled H-AZO). ....... ................ ............... ..85

5-3 AFM image of two AZO films deposited using different sputtering gas. a) AZO b)
H -AZO .............. ...............86....

5-4 AFM image of H-AZO thin films with increasing thickness. a)200nm b)400nm
c)600nm d)800nm ........... __..... ._ ...............87....

5-6 Variation of AZO thickness with different positions along the axial direction. ...............89

5-7 Variation of sheet resistance with different positions along the axial direction. ..............89

5-8 Variation of resistivity with different positions along the axial direction. ......................90










6-1 Schematic of the PMEE reactor ................. ...............106..............

6-2 Photo of CIGS solar cells fabricated on a 2"xl" soda-lime glass substrate. ................... 106

6-3 Cross section SEM image of absorber #582. .......................... ........107

6-4 Cross-section SEM image of absorber #588 ................. ...............107........._ ..

6-5 Cross-section SEM image of CGS device #640. ............. .....................108

6-6 AZO window layer optical transmission of three CIGS solar cells. .............. .............109

6-7 CIGS cell efficiency versus the AZO layer sheet resistance. ............. .....................10

7-1 Schematic diagram illustrating the forward modeling and inverse modeling
operations for a given physical system S............... ...............120...

7-2 Parameter classification with examples for solar devices............... ...............121

7-3 Procedure chart for solar cell inverse modeling. ............. ...............122....

7-4 A typical CIGS cell structure. ........._... ...... .___ ...............123..

7-5 Cell structure described in Medici. ............. ...............123....

7-7 A typical I-V curve generated by Medici. ............. ...............124....

7-8 Medici simulation result for an NREL champion device. ............. ......................2

7-9 Iteration of inverse modeling process for there solar cell performance variables: (a)
Ac,, (b) Voc, and (c) FF. (Case 1)............... ...............126..

7-10 Iteration of inverse modeling process for there solar cell performance variables: (a)
Ac,, (b) Voc, and (c) FF. (Case 2)............... ...............127..

7-11 Iteration of inverse modeling process for there solar cell performance variables: (a)
Ac,, (b) Voc, and (c) FF. (Case 3)............... ...............128..









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

ZINTC OXIDE SPUTTER DEPOSITION AND MODELING OF
COPPER-INDIUM-GALLIUM-DISELENIDE-BASED THIN FILM SOLAR CELLS

By

Wei Liu

December 2007

Chair: Oscar D. Crisalle
Major: Chemical Engineering

Highly transparent and conductive aluminum doped zinc oxide (AZO) thin films were

successfully obtained through RF magnetron sputtering using argon as the sputtering gas. Thin

fi1m AZO grown under different conditions such as base pressure, deposition power and working

pressure was characterized. A correlation between various operational parameters and the

optical and electrical properties of AZO was developed and used to suggest optimum of

operational conditions.

The sputtering process was improved by adding a small amount of hydrogen into the

sputtering gas. A significant improvement of the conductivity of AZO thin films was observed.

This observation together with results obtained under different base pressures supports the

literature prediction that hydrogen plays a favorable role in n-type zinc oxide thin films.

Hydrogen participates in the doping process and causes increased carrier concentration.

The AZO thin films sputtered with the Ar and H2 mixture gas were incorporated in copper-

indium-gallium diselenide (CIGS) solar cells and devices with a thin film layer structure of the

form ZnO/CdS/CIGS/Mo were fabricated. It is found that the sheet resistance and thickness of

the AZO layer can be an important factor determining the performance of CIGS solar cells. The










devices were measured under AM 1.5 radiation and a conversion efficiency of approximately 9%

was achieved.

To better understand solar cell devices fabricated using chalcopyrite semiconductors such

as CulnSe2 and CuGaSe2, an inVeTSe modeling process was developed. The inverse modeling

process was realized by integrating two software tools, namely Matlab and Medici, under a

Linux environment. Matlab provides the values of candidate input parameters to Medici, which

in turn outputs device simulation results that are fed back to Matlab. This process continues until

the outputs reach the target values. The inverse modeling process successfully identified the

defect distribution in a CIGS layer that matched a known cell performance. In an ensuing

optimization step, optimal defect concentrations were found, revealing that the shallow acceptor

defect concentrations are favorable for increasing the efficiency.









CHAPTER 1
INTTRODUCTION

With the decline of the traditional fossil fuel resources in the world, renewable energy

sources are emerging as the future solution to maintain a sustainable development of human

society. Among various kinds of renewable energy solutions including hydroelectric,

geothermal, wind, and solar (photovoltaic and thermal), photovoltaic technology has attracted

more and more attentions for its advantages over other options. Photovoltaic technology has the

following advantages. The first advantage is the abundant energy source. The sun is a reliable

energy source and it constantly radiates a tremendous amount of energy toward the earth. The

second advantage is clean and environmentally friendly. Photovoltaic devices, commonly

known as solar cells, convert the energy from sunlight directly into electricity. There is no

emission, noise or pollution of any kind during the operation. With more and more concerns

about the "greenhouse gases" and their detrimental effect on our planet, with the awareness of

the possibility of irreversible environmental pollution that could be brought by nuclear power,

the value of PV technology as a clean, renewable energy source has been increasingly

appreciated. The third advantage is reliability. PV technology was originally developed for

space applications where solar cells were used to power satellites. With no moving parts, solar

cells can operate reliably and require little maintenance. With all these advantages, solar energy

is emerging as a promising alternative to traditional energy sources.

The amount of light energy available annually in a particular landmass on earth depends on

the climate and the latitude. In areas closer to the equator, available energy per unit area is

larger. On average between 2 and 3 megawatt-hours (MWh) of solar energy falls on a square

meter of land in the United States annually. Accordingly, computed annual solar energy

available over the total land area of the United States is about 2.4x 1016kWh, which is 10,000









times larger than the annual electricity consumption of the United States [1]. The sun is a huge

energy reservoir that could provide the increasing need for energy consumption.

Solar cells have been successfully applied in various filed to provide electricity. Examples

of these applications are powering devices on satellites for space applications, providing

electricity for remote areas where the power infrastructure is not available. Although cost is not

a big concern in these situations, it becomes a major problem for terrestrial applications where

power grid is available. When compared with fossil fuels, the relatively high costs of

semiconductor materials and processing are a maj or barrier for the large-scale implementation of

solar devices. The high material and fabrication cost keeps the price of solar panels at a high

level. As a result, the huge investment on these solar panels may keep consumers from choosing

solar electricity considering the fact that it may require tens of years to return the initial

investment and start saving consumer's money. Until low cost, highly efficient photovoltaic

devices are ready to be fabricated, the application of solar electricity is unlikely to replace the

traditional energy source and become the mainstream. Therefore, it is imperative to develop low

cost processes and improve performance of PV materials to bring down the cost per watt for

solar electricity.

In order to reduce the cost of PV systems, which is usually evaluated in dollars per

kilowatt hour ($/kWh), the current R&D is concentrated on both improving the conversion

efficiency and reducing the semiconductor material cost as well as the processing cost. The

average residential price for utility-generated electric energy in the U. S. in 1994 was

US$0.079/kWh. In order to reach a cost target as US$0.06/kWh for electricity from a PV plant

operating for 30 years, the module efficiencies are required to be in the range of 15% to 20% for

a flat plate panel system and 25% to 30% for a system operating under concentrated sunlight.









These correspond to module area costs of $45 to $80/m2 and $60 to $100/m2, TOSpectively [2].

Although the current fabrication cost is still beyond these values, with the favorable green energy

policy from governments, such as government solar rebates, more and more solar panels are

installed to provide clean utility energy.

Single crystal silicon solar cells have achieved good performance with efficiencies beyond

20%. However, due to the high cost nature of the Czochralski process used to produce single

crystal silicon, the price for silicon solar panels remains high. Direct bandgap thin film materials

have much higher light absorption coefficient than silicon and require a much less material

thickness to absorb the light. They are possible alternatives that could significantly lower the

material consumption and consequently the cost. Three typical thin film PV materials are

amorphous silicon, cadmium telluride, and copper chalcogenides. Among these thin film PV

materials, copper indium gallium di-selenide (CulnxGal-xSe2 Of Simply CIGS) is very attractive

for its high performance, long-term stability and relatively little toxicity. The work in the scope

of this dissertation includes both theoretical analysis of CIGS solar cell device physics through

simulation and experimental fabrication and characterization of this kind of solar cells.

A process to grow device quality zinc oxide window layer using an industry sputtering

system was developed and optimized. Aluminum doped zinc oxide (AZO) thin films with good

optical transmission and resistivity between lx10-30-cm and 2x10-30-Cm Were repeatedly

obtained. It is also observed that residue water vapor in the sputtering chamber is beneficial to

the electrical conductivity of aluminum doped zinc oxide. Motivated by this observation, a

modified process using hydrogen and argon mixture as the sputtering gas was implemented.

Resulted zinc oxide thin films show a lower resistivity. Hall measurement shows that this

lowered resistivity was achieved through the increase of carrier concentration in the film.









This process was then successfully applied in CIGS photovoltaic device fabrication.

Efficiency of approximately 8% and 9% was achieved on CIGS from Energy Photovoltaic Inc.

(EPV) and those from the PMEE reactor at the University of Florida, respectively. This is the

first time that CIGS photovoltaic devices are completely fabricated in house, achieving the goal

of two previous attempts [3], [4].

Although laboratory scale CIGS devices with efficiency of 19.5% have been achieved

experimentally, this material system is not fully understood and many issues remain unknown.

For example, the defect distribution in this layer has big impact on the performance of the Einal

device and it is of great interest to investigate the relation between defects and cell performance.

In order to do this, software Medicio" from Synopsys Inc was utilized in simulation to predict

solar cell performance under different conditions. An inverse modeling was implemented to

identify the defect distribution and their impact on CIGS devices. Solar cell performance was

also optimized using the inverse modeling program.









CHAPTER 2
LITERATURE REVIEW

Solar Light

The sun is a hot sphere of gas whose internal temperatures reach over 20 million degrees

Kelvin due to nuclear fusion reactions at the sun's core converting hydrogen to helium. The

radiation from the inner core is strongly absorbed by a layer of hydrogen closer to the sun's

surface. Heat transfers through this layer by convection. The surface of the sun, called the

photosphere, is at a temperature of about 6000K and can be represented as a blackbody radiation

source.

An immense amount of solar energy radiates toward the earth at any time. The radiation

power density at the sun's surface is about 5.96x106 mW/cm2 and it decreases with increasing

the distance from the sun's surface as the radiation sphere gets larger. When reaching the

atmosphere of the earth, the sunlight carries a power density of about 135.3mW/cm2 [5]. This

spectral distribution is referred to as Air Mass Zero (AMO) since it has not passed through the

air. Figure 2-1 shows a comparison between AMO radiation and a blackbody radiation at 5800K

[6]. During the process of going through the earth atmosphere, sunlight will be absorbed and

scattered and its energy is reduced. Different incident angles will cause different amount of

energy loss and this is reflected by Air Mass. In the situation shown in Figure 2-2, Air Mass is

defined by

Pl 1
AM= P co (2-1)

where Pl is the actual path length that light takes through the atmosphere and P2 is the

shortest possible path length, i.e., the normal incident distance. For example, AM1.5 refers to

sunlight passes through 1.5 atmospheres with a solar zenith angle (6 in Figure 2-2) of 48.190









AM1.5 is usually used as incident light for testing terrestrial solar devices. Figure 2-3 shows the

comparison between AM1.5 radiation and a blackbody radiation at 5800K [6].

The Photovoltaic Effect

Photovoltaic effect is the creation of an electromotive force by the absorption of photons in

a semiconductor device. When photons with energy higher than the energy bandgap of a specific

material are radiated on the material, they can be absorbed and the photon energy will excite

electrons from the valance band to the conduction band and consequently generate electron hole

pairs. To make a solar cell, the electron hole pairs need to be separated before they recombine

with each other. Semiconductor p-n junctions were successfully applied to realize this process.

When a p-type semiconductor contacts with an n-type semiconductor, a built in electric filed will

be generated. This electric field drives any electrons generated in the p-type semiconductor

across the junction to the n-type semiconductor. Thus the electron-hole pairs are separated.

History of Photovoltaic Technology

The photovoltaic effect was first discovered by Frenchman Edmund Becquerel in 1839.

He observed a special effect while experimenting with an electrolytic cell made up of two metal

electrodes. When the metal plates immersed in a suitable electrolyte are exposed under sunlight,

a small voltage and current were produced. The first solid-state materials that showed a

significant light-dependent voltage were selenium in 1876 and later cuprous oxide, which

indicated that semiconductors would eventually become the most promising materials for

photovoltaic energy conversion.

PV technological development began with the build up of a silicon p-n junction by a group

of researchers at Bell laboratory in 1954 [7]. But the powerful driving force for the PV

technological development did not come until the advent of space exploration. The urgent

demand for a reliable, long lasting energy source to power the spacecraft became the motivation









of solar cell development. By 1958, the first silicon solar cells were applied on spacecraft.

Interest in solar cells as a clean, renewable terrestrial energy source arose after the oil embargoes

in 1973 and 1979.

Shay et al. [8] at Bell Labs first discovered that the I-III-VI2 semiconductor (CulnSe2) can

be effective used as a photovoltaic material in the 1970s. A polycrystalline CulnSe2 Cell Of

nearly 6% efficiency was developed at the University of Maine in 1976 [9]. Over a period of

more than one decade starting from early 1980s, a research group at Boeing was able to

consistently push up the efficiency of polycrystalline chalcopyrite solar cells to a higher value

when it reached 13.7% in 1993 [10]. The Boeing CIS project was ceased not long after that.

The research of CIGS technology are later led by researchers at the National Renewable Energy

Lab (NREL), who reported an conversion efficiency beyond 19% in 2003 [l l]. CIGS

technology has also been transferred to industry pilot production both in Europe and the United

States. It is becoming a promising thin film technology that can dramatically lower the

production cost.

Device Physics

Most solar cells rely on p-n junction to separate the photon generated electron and hole

pairs. The p-n junctions are formed by j oining a p-type and n-type semiconductor together.

Once connected, diffusion happens due to the concentration difference of electrons and holes

between these two types of semiconductors. Since the n-type semiconductor has a high electron

concentration and the p-type a high hole concentration, electrons diffuse from the n-type region

to the p-type region. Similarly, holes flow by diffusion from the p-type region to the n-type

region. Because electrons and holes are charged, their diffusion causes charge redistribution in

the junction area and an internal electric Hiled is formed across the junction. The region where

the electric Hield is present is called "space charge region". All free carriers in the space charge










region are quickly swept away by the electric field and therefore this layer is also called as

"depletion region" due to the unavailability of free carriers. As a result of the electric field, a

built-in potential is present across the junction. This process can also be explained by the Fermi

level alignment between the p-type and n-type region due to the diffusion of electrons and holes,

which is shown in Figure 2-4. Under thermal equilibrium, both diffusion and drift happens and

their current balance and there is no net current flow through the junction.

When the p-n junction is illuminated by photons whose energy is larger than the bandgap,

electron and hole pairs are generated. These free carriers generally have very short lifetime and

they will recombine with each other if they are not separated in a timely fashion. However, the

built-in potential across the junction will swiftly sweep the electrons and holes across the

junction so that photon generated electrons reaches the n-type region and holes the p-type region,

where they become the so called maj ority carriers and no longer suffer a fast recombination. If

the p-n junction is connected with external circuit, the light generated current flows out and the

semiconductor p-n junction becomes a solar cell. If the p-n junction is not externally connected,

i.e., under open circuit condition, the photon generated current will produce a forward bias across

the diode, which reduces the barrier for diffusion and promotes the diffusion current to balance

out the photon generated current. At equilibrium, the net current still remains zero. This

forward bias created is referred to as open circuit voltage (Voc).

Figure 2-5 shows an equivalent circuit of a solar cell [12]. Rs represents series resistance.

Series resistance comes from the resistance of the material that constitutes the solar cell and the

contact resistance between the body material and external circuit. Rsh represents shunt

resistance, which is the resistance in parallel with a solar cell that results in current leakage

losses within the cell. An ideal solar cell would have zero series resistance and infinite large









shunt resistance. The effects of series resistance and shunt resistance on solar cell performance

can be found in Figure 2-6 [13]. From the calculated results shown in the graph, the solar cell

current-voltage (I-V) curve (i. e., performance) is strongly affected by the series resistance while

shunt resistance plays a less important role. The results show that the I-V curve changed

significantly when the series resistance increased from 0 to 502. Instead when the shunt

resistance reduced from infinite large to 10002, there is no maj or change for the I-V curve. Thus

series resistance is a dominating factor in determining the power available from a solar cell. The

performance deterioration mainly comes from the rapid drop of fill factor (FF), a response of a

slight increase of series resistance. Figure 2-7 shows how quickly an increasing series resistance

affects the performance of a solar cell.

This result shows that it is critical to obtain a low series resistance, i. e., to reduce the

resistance of the internal material that constitutes the device. This is the motivation for the

efforts to deposit transparent zinc oxide with extremely low resistivity, which is part of the work

in the scope of this dissertation.

Solar Cells using Different Materials

Among the various materials that have been tested as absorber layer for PV devices,

several are quite successful in their applications. Examples of them are crystalline silicon,

amorphous silicon and a wide variety of compound semiconductors such as GaAs, InP, CulnSe2

and CdTe. These materials can be generally divided into two categories based on their energy

band structure. One is called direct band gap material and the other is referred to as indirect

band gap material. For direct band gap materials, the maximum of the valence band energy and

the minimum of the conduction band energy show at the same crystal momentum values. The

indirect band gap material does not have this property. This band gap structure difference leads









to the fact that direct band gap materials have a much higher absorption coefficient than indirect

materials. Consequently, for indirect band gap material such as crystalline silicon, it requires

more thickness of the material and more critical preparation technique to produce a qualified

absorber layer with reasonable efficiency. For example, a silicon film of thickness about 100

microns is required to absorb 95% of the photons in the solar spectrum [14] while for a direct

band gap material, a film of less than 1 micron thickness is capable of absorbing more than 90%

of the incident sunlight. Because of the large thickness of material required, the silicon film has

to be ultra-high pure for the generated minority carriers to travel the large distance towards the

junctions without being captured by defects. Single crystal silicon used in PV applications is

usually fabricated by the Czochralski (CZ) process [15], which significantly elevates the

fabrication costs. Nevertheless, silicon technology still dominates the current PV market as a

mature technology. Figure 2-8 shows the technology market share in 2002 [16]. Some of the

important PV materials are briefly introduced in the following sections.

Crystalline Silicon Solar Cells

Silicon is the most extensively studied semiconductor and was the first commercially used

PV cell material. It is a cheap and abundant element that can be found in sand and other

minerals. Silicon processing is very mature through the continuous development of integrated

circuit processing technique since the 1950s. This provides great background theory and

technical support to silicon solar cell research.

Single crystalline silicon

High purity single crystal silicon solar cells are the first commercially developed solar

cells. Efficiency of single crystal silicon has been beyond 20% as early as in the 1980s [17].

Commercial single crystalline silicon cells usually have lower efficiency.










Maj or drawback of single crystalline silicon solar cells are the high cost and silicon supply

problem. Silicon wafers need to be cut off from ingots produced by Czochralski (CZ) process,

which is a costly method. As an indirect bandgap material, it requires thicker silicon

consequently more amount of materials than thin films to build a high efficiency cell. About half

of the cost of a silicon module comes from the cost of single crystal silicon wafers. With the

high demanding form the semiconductor industry, price of silicon wafers remains high. The PV

industry has in the past used rej ected silicon from the semiconductor industry that was available

at a lower cost. This creates a dependence on the source materials and restricts the development

scale of this technology. The high material cost and processing cost together with the

competition of wafer consumption from the integration circuit industry led to the development

of alternative solar cell materials.

Polycrystalline silicon

Polycrystalline silicon is typically produced by block casting and solidification of

silicon[18], which avoids the costly pulling process used by the CZ method, resulting a process

much more economic compared with the strict CZ process. Silicon is melted and poured into a

square graphite crucible to create silicon blocks with a square shape, which yields better

utilization of the wafer when later assembled into modules. With controlled cooling process,

polycrystalline silicon block with a large crystal grain structure can be produced. The grain size

can be obtained is about millimeters or centimeters. The silicon blocks are then sawn into wafers.

A waste of about 50% of the silicon can happen during the process of sawing.

Polycrystalline silicon is not used in integration circuit industry and is only produced for

fabrication solar cells. Although it typically has a lower solar cell efficiency, it is cheaper than

single crystal silicon. Another advantage is that the square shaped wafers can easily be

manufactured into square solar cells compared to round shaped pulled silicon crystals. Silicon









based PV technology is the most advanced and they share the largest percentage of the current

market.

Thin Film Solar Cells

Materials used for thin film solar cells are direct band gap semiconductors with high

absorption coefficients. With a thickness less than a few microns, they are able to absorb most of

the incident sunlight. Because the film thickness is small, generated electron-hole pairs only need

to travel a small distance to be gathered in the collector. The loss due to recombination is greatly

reduced compared with crystalline silicon. Thus, the purity requirement is not as strict as that for

silicon, which may lead to a significant drop for the manufacturing costs. Amorphous silicon,

CdTe and CulnSe2 (Or its alloys) are three types of most promising material systems for thin film

solar cells because of a good combination of reasonable efficiency, reliability and low production

costs. The development history of these three thin film materials is shown in Figure 2-9.

Amorphous Silicon (a-Si:H) is the first thin-film material that has been commercially

fabricated into photovoltaic cells and it holds the largest market share in thin film photovoltaic

applications. It is also the most commercially developed thin film technology. CdTe and CIS

technology have also been commercially developed recently. A brief introduction of amorphous

Si and CIS technology is introduced next.

Amorphous silicon

Amorphous silicon is a disordered cousin of crystalline silicon. Although each silicon atom

in amorphous Si has the same number of nearest neighbors as crystalline silicon, the bond angles

connecting it to the neighbors are unequal. In stead of the 1090 angles in crystalline silicon, the

bond angles of a-Silicon are not constant. This disordered structure leads to a significant

difference between amorphous Si and crystalline silicon. For example, instead of having an

indirect band gap of 1.1eV (as for crystalline silicon), it has a direct band gap of about 1.75eV.









The average ab sorption coefficient of a-Silicon is much larger than that of single crystalline

silicon over the visible light spectrum, which makes it possible to absorb the incident sunlight

with a much smaller film thickness.

As stated before, a-Si has the most commercially advanced technology among all thin film

materials. It is actually the only thin film material system that has an impact on the current PV

market. But there is a maj or problem with a-Si solar cell system which holds back the

development of reliable high efficiency a-Si modules. It is the performance degradation under

illumination caused by light-induced photoconductivity degradation [19], a phenomenon known

as Staebler-Wronski effect [20].

To improve the performance of a-Si solar cell, a critical technique called light trapping

[21] is utilized to increase the light path length in a-Si, thus enhance the amount of photons to be

absorbed. Transparent Conductive Oxide (TCO) plays a critical role in realizing light trapping.

TCO can be deposited or post etched to form a rough surface with features serving as reflection

mirrors. Light reaching these surface features can be scattered back into the device. By applying

this technique appropriately, short circuit current can be significantly increased. Traditional

TCOs like Indium Tin Oxide (ITO) and SnO2 WeTO Originally applied to serve light trapping

functions. But for superstrate structured a-Si solar cells, a-Si is typically deposited on top of

TCO using PECVD (Plasma Enhanced CVD) process, during which the TCO is exposed to

hydrogen plasma. It is found that the electrical conductivity and optical transmission of these

traditional TCOs deteriorate when exposed to hydrogen plasma while ZnO as a new TCO

material shows more resistance to hydrogen plasma [22].

Radio frequency magnetron sputtering can be used to deposit highly conductive and

transparent ZnO:Al films to serve the light trapping purpose. But films grown by this technique









generally are smooth and do not have a light trapping structure. A wet etching process using

hydrochloric acid is typically used to create surface roughness [23].

Amorphous Si based solar cells can be fabricated in two configurations: substrate and

superstrate structure. Substrate type modules are built on stainless steel foil which has been first

covered with a textured back reflector and superstrate structured devices are built on glass coated

with TCOs. Tandem structured Si solar cells with a-Si as top cell are also fabricated to further

increase the total conversion efficiency.

CulnSe2 and CuGaSe2

In the mid-1970s, a small group at Bell Laboratories in New Jersey began to work on

several compound semiconductors for advanced photovoltaic applications. Among the materials

studied, there was a category called chalcopyrite such as CulnSe2 (CIS), CuGaSe2 (CGS) and

CulnS2. They were successful to produce a singe crystal CulnSe2 based cell which reached 12%

efficiency in 1975. This spurred the interest with the exploration of ternary and quaternary

compound semiconductors on the application in photovoltaic filed.

CIS is a direct band gap material and is the most light absorbing semiconductor known.

The absorbance can be as high as 105 cm l. Half a micron of CIS is sufficient to absorb 90% of

the incoming photons, making it possible to achieve high efficiency with much less material

consumption than the silicon PV technology. Besides, this material system also possesses

excellent stability and it is especially radiation hardness [24], which leads to a long life-time with

consistent performance.

Figure 2-10 shows a typical CIGS solar cell structure. The typical substrate is glass.

Flexible and light weight substrate such as metal foil can also be used. A thin layer of

molybdenum is followed as the back contact. Then p-type CIGS compound is grown on top of

Mo. After CIGS is a thin layer of cadmium sulfide (CdS) with n-type conductivity. Next to CdS









are the double transparent layers of intrinsic and doped ZnO. Optional MgF2 antireflection layer

can be evaporated on top of doped ZnO. Finally, it is nickel and aluminum dual metal layer

serving as the top contact of the device, where nickel serves as a diffusion barrier between

aluminum and zinc oxide.

CIS has a bandgap of 1.04 eV. This is smaller than the bandgap of silicon. Devices made

out of CIS typically have a small open circuit voltage and a large short circuit current. This leads

to a relatively large series resistance loss and compromise cell performance. Gallium can be

added to increase the band gap of the compound and form an alloy typically referred to as CIGS.

Sulfur can also be added to increase the band gap. High performance CIGS based devices

generally require very high temperature to produce the absorber layer and the fundamental

reason for that is not clear.

With a relatively large bandgap of 1.68 eV, ternary compound CGS is also a very

interested material under investigation. Large bandgap CGS can be used as top cell for multi-

junction solar cells (tandem cells) and it responds to the short wavelength part of the solar

spectrum and allows long wavelength photos to pass through itself and reach the bottom cell. It

is estimated that CGS/CIS tandem structure cells can reach efficiency beyond 30%. For

CGS/CIGS tandem structure, modeling work predicted 25% conversion efficiency can be

achieved [25].

Transparent Conductive Oxide

Because transparent electrode is an essential ingredient in most optical-electronic devices,

extensive research has been done on both material selection and fabrication techniques. The

most popular materials used to develop highly transparent and conductive thin films are n-type

semiconductors consisting of metal oxides.









Transparent and electrically conductive fi1ms have been prepared from a wide variety of

materials, including semi-conducting oxides of tin, indium, zinc and cadmium nitride [26], and

they are commonly referred to as Transparent Conductive Oxide (TCO). Among these, the TCO

thin films composed of binary compounds such as SnO2 and In203 have been developed and

doped SnO2 (Sb or F doped SnO2, e.g. SnO2:Sb or SnO2:F) and In203:Sn (indium tin oxide, ITO)

are in practical use. In addition, ternary compounds such as Cd2SnO4, CdSnO3 and CdIn204 have

also been developed but have not yet been used widely [27]. One advantage of binary compound

TCOs is that their chemical composition during the fi1m deposition is relatively easier to control

than that of ternary compounds and multi-component oxides. Typical undoped binary compound

TCO films such as SnO2, In203, ZnO, and CdO are n-type degenerate semiconductors, with

electron concentrations of the order of 1020 CA-13. To further lower the resistivity, doping

materials are added into the fi1ms to enhance the free electron concentration. The lowest

resistivity obtained is on the order of 10-5 G-cm, based on CdO:In films prepared for flat-panel

displays. However, these were not practically used because of the toxicity of Cd. In practical

films, the lowest resistivity comes from ITO films with a value of approximately 1 x 10-4 2-Cm.

Indium is an expensive material because the amount of Indium in nature is very small. There are

no mines intentionally built to obtain indium simply because of the trivial amount available. In

fact, indium is only obtained as a byproduct from mines for Zn and other metals. To obtain even

lower resistivity than that oflITO and find an alternative to ITO, doped ZnO films were

developed in the 1980s. Unlike In, Zn is an inexpensive, abundant and harmless material.

Recently, ZnO:Al and ZnO:Ga doped thin films, with a resistivity of the order of l x 10-4 "2Cm,

have attracted much attention for deployment as the transparent electrode for thin film solar cells.

It is believed that doped ZnO is a promising alternative to ITO.









Thin film ZnO has an energy bandgap greater than 3eV and can be deposited with high

transparency. If heavily doped with Al or Ga, it can be very conductive as well. The high

transparency allows the light to go through, reaching the absorber layer, where electron-hole

pairs are generated. Highly doped conductive ZnO fi1ms easily form an ohmic contact with Al,

which is essential for collecting the current generated from the cell.

Intensive research work on ZnO has been done for solar cell applications [28], organic

light-emitting diodes [29-3 1] and transistors [32-34]. Thin films of zinc oxide can be prepared

using various techniques such as spray pyrolysis [35-37], evaporation [38-40], chemical vapor

deposition (CVD) [41-43], and sputtering [44-46]. The sputtering technique can produce as-

deposited, highly conductive ZnO fi1ms without any post deposition treatment. Key advantages

of sputtering are discussed in the next section.

Sputter Deposition

Introduction

Sputter deposition of films, which is often referred to as sputtering, was first reported by

Wright in 1877. Edison patented a sputtering deposition processes to deposit silver on wax

photograph cylinders in 1904[47].

Sputter deposition is a physical vaporization of atoms from a surface by momentum

transfer from bombarding energetic particles, which generally are ions generated from a glow

discharge or a plasma process. The ions developed in the plasma are accelerated in an electric

Hiled and bombard at a target, which is the source of the deposited material. After ion impact, the

eroded material then condenses from the gas phase onto the substrate where the film is grown.

To attract the positive ions, the target is applied a negative electric filed. Therefore the sputter

deposition source is also referred to as the cathode.










Sputter deposition can be used deposit films of either element or compound material.

When sputtering a compound thin film, the target can be either compound or elemental. In the

later case, a reactive gas is incorporated in the sputtering gas and the compound is formed during

the sputtering process. In some cases, sputtering a compound target can result a loss of some of

the more volatile material (such as oxygen from ZnO) and the stoichiometry of the fi1m may be

different from that of the target. By adjusting the deposition parameters or add the volatile

component into the sputtering gas to compensate the loss, resulted films can reach a desired

stoichiometry, which is important for transparent conductive zinc oxide.

Advantages of Sputter Deposition

Sputter deposition is a flexible thin film growth process. It can be done under different

vacuum conditions varying from below 10 STorr up to approximately 50mTorr. High vacuum

sputter deposition using ion beams can reduce the damage of substrate from the plasma

environment. Sputter deposition can deposit films of element, alloy and compound from either

conductive or insulating target. There are also some of other advantages can be achieved

through sputter deposition.

One of the major differences between sputtering as a plasma process and thermally excited

thin-film preparation methods such as evaporation, chemical vapor deposition is the much higher

energy input into the growing film. The relatively high kinetic energy of the adatoms with

values of a few electron voltage can be achieved in contrast to around 100 meV for evaporation

process. The high kinetic energy increases the thermal surface mobility of the adatoms, allowing

more rearrangement of the atoms on the surface, consequently yielding improved film properties.

The target is generally cooled and the energy of ej ected particles comes from momentum

transfer form the bombarding ions. The cold surface presents little radiant heating to the vacuum

system, which is an advantage over evaporation, where radiant heating can be significant,










especially in the case of long period intensive evaporation. The small amount of heating makes

sputter deposition an appropriate process to deposit films on heat sensitive substrates.

The sputtering target can provide a stable, large area, solid vaporization source. Because

the vaporization comes from a solid surface, the sputter geometry can be up, down or sideways,

in contrast to evaporation from a melted source inside a crucible.

Sputter Deposition of ZnO

Thin film ZnO can be produced with different sputtering techniques such as radio

frequency sputtering, direct current reactive sputtering and middle frequency reactive sputtering.

The properties of sputtered ZnO fi1ms are dependent on many factors. Comprehensive studies

on parameters such as deposition power, working pressure, substrate temperature, oxygen

concentration, and target doping concentration have been carried on [48,49]. Individual

parameters such as RF power [50], target to substrate distance [51] are also investigated.

Although the research on sputtered ZnO is extensive, the reports from different authors are quite

different. It seems that the optimum operation condition is equipment specific and there is not a

universal recipe that guarantees ZnO fi1ms with good electrical and optical properties. Therefore,

it is essential to determine the optimum parameter set for the UF oxide sputtering system.

Sputtered ZnO thin films have been successfully used for various optoelectronic devices

such as organic light emitting diodes (OLED) and solar cells. Particularly for thin film solar cells,

ZnO works well for different absorber materials such as CdTe [52], a-Si [53], CIGS [54]. There

is also effort made to reduce the visible and particularly near-infrared optical absorption of ZnO

fi1ms to further improve the performance of solar cells [55].

Recently, producing p-type ZnO has been a popular research area due to the interest in

fabricating homojuction wide bandgap ZnO devices for optoelectronic applications. Sputtering









has been successfully demonstrated to produce p-type ZnO [56]. The thermal stability of p-type

ZnO needs to be improved for reliable applications.

Summary

Thin film ZnO has broad applications in optoelectronic devices including light emitting

diodes, lasers and solar cells. Sputter deposition is an effective method in producing n-type ZnO.

The optical and electrical properties of sputtered ZnO can be affected by various deposition

parameters. An optimization of the sputtering condition is necessary to produce highly

conductive yet transparent ZnO films that are required for photovoltaic applications.










2.5


E
a
E
ce
5
r


0.5



0 0


"1 2 3


Figure 2-1. Energy current density per wavelength of AMO spectrum(heavy line) compared with
blackbody radiation at 5800K(thin line).


Figure 2-2. Schematic of the Air Mass definition.


~ ~eP











>i
culO~'


2
w
r;r'


2 3 t~wleV4


Figure 2-3. Energy current density per photon energy of AM1.5 spectrum (heavy line) compared
with blackbody radiation at 5800K (thin line).


Selectrons
O holes


Drift


O~ n-type E
Diffusion E

O E,
Drift X



Figure 2-4. Formation of built-in potential between p-n junction.



























Figure 2-5. An equivalent circuit of a typical solar cell.


1 ~J I ~1~


,,


I I... I I d el._ I lI


I I I 1 90


_ _ ___


are ,a0 p a Ig




VOk J I l


20 -


Rlh~c#


He


YY


Figure 2-6. Effect of series resistance and shunt resistance on solar cell performance.


te = 10 emp.


* 4QvltT' ,~l















R .0 *


R MM
n


I


voLT


P 4


0 10 IR 14 13 it 2


Figure 2-7. Effect of series resistance on solar cell performance.













81 film l

Ribbon Si a-Si onCz slice
3.30% 5.890% /


Amorphoslsw SI
15,50%6
_Cadmiumn t~ellurdde
/0;40%$
Other
0 80%~








Single crystal Si
2936%a


PolyLcystal Si
54,44%


Figure 2-8. World PV market share by technology in 2002.


* CulnSe2
NREL
CdTe e*
Univ. of So. FoidRL
o Amorphous silicon B oa uoI
(stabihized) Boeing Boeing
Lodak ,ARCO ~ Univ. of So. FL~g
Boeing, --Y A M E e
...****United Solar
Monosolar ,oa Boeing .**
Matsushita .CT
Boeing ~....**** ECD
univormaThe Best One-of-a-Kind Laboratory
..0'"Cell Efficiencies for Thin Films
.0**(Standard Conditions)
CP".'I- --~-~-~ RCAI


12

8


1975 1980 1985 1990 1995 2000
Figure 2-9. Development history of thin film solar cells. (Courtesy of NREL)


2005


World Market 2002 bsy Technology












Ni/Al
MgF2
ZnO:Al (3.3eV)


i-ZnO
CdS (2.4eV1


0.05/0.5Cim
0.1 Clm

0.5-1.5 Cim

0.05 Clm

0.05 Clm


1.5-2.0 Cim




Alo
0.5-1.5 Cim

Glass


Figure 2-10. Thin film CIGS solar cell structure









CHAPTER 3
EXPERIMENTAL IVETHODS

ZnO Sputtering System

Sputter deposition was adopted as a room temperature process which could keep the

pre-deposited CIS and CdS layers from being destroyed by high temperatures. The equipment

used for ZnO sputtering is a Perkin-Elmer 4400 industry sputtering system. A photo and

schematic of the equipment are shown in Figure 3-1 and Figure 3-2, respectively. It is equipped

with a stainless steel, load-locked vacuum system to generate the vacuum required for sputter

deposition. Typical base pressure of the system is approximately 6x107 Torr. However, it

requires long period of pumping to reach this pressure due to the large size and surface area of

the substrate transfer pallet.

The system originally has three RF planar sources, each handling an 8 inch target in a

sputter down geometry. Targets currently being used are Mo, ZnO:Al203 (98wt%:2wt%) and

Zn. An 8 inch diameter magnetron was later added on top of the ZnO:Al203 target to confine

electrons close to the target and increase the sputter rate. Uniformity shields are integrated 1.6

cm below each target to compensate the short exposure time of substrates placed at large radial

distance due to the higher angular velocity when rotating. These shields modify the flux rate

from the target (blocking flux in small radial distance areas) to account for the different exposure

time along the radial direction of the platen, thus generate a uniform film when the platen is

rotated. Because the plasma distribution for magnetron sputtering and diode sputtering is

different, the shape of the uniformity shields for the two cases are different. These uniformity

shields are removed for experiments without substrate motion.

The sources and the substrate table are cooled by chilled water. The substrate platen has

rotary motion capabilities, allowing deposition of uniform films on up to fourteen 4 inch wafers









in a single batch process. The motion system for the substrate platen is controlled with a

stepping motor, which gives accurate control of the substrate motion. It can generate full

rotation in either clockwise or counter clockwise direction at precise speed and acceleration,

designated number of oscillations about a region below the target. Although these capabilities

can be of utility for complicated procedures, accurate positioning of the substrates below the

sources are the primary uses for the motion system. Generally motion on the substrate platen is

not adapted during deposition to avoid generating particles, which could contaminate the vacuum

environment. The throw or substrate to target distance can be set between 5 and 9 cm.

The RF power is supplied by a Randex 2kW RF generator which provides a radio

frequency of 13.56 MHz. This power is delivered to the target through an impedance matching

network. There is a switch available to route the power generated by the single generator to the

appropriate source, or to the substrate table when sputter etching is used to clean or modify the

sub state.

A CTI CryoTorr 8 cryopump and a Leybold Trivac 60cfm rotary vane pump provide the

high vacuum and rough pumping, respectively. A throttle valve between the cryopump and the

deposition chamber reduces the pump's effective pumping speed by up to a factor of 50 when

engaged. Two mass flow controllers (MFCs) are used to regulate introduction of Ar and 02 gaS,

and are used in conjunction with the throttling valve to generate a controlled gas ambient over

the pressure range of 2-150 mTorr, a range typical for sputter deposition. Both MFCs have a

range of 0-100 standard cubic centimeters per minute (SCCM), and control gases from ultrahigh

purity gas bottles. The system operates at a typical base pressure of 6x107 Torr. Vacuum

gauging is provided by an array of two Bayard-Alpert ionization gauges, two Convectron

gauges, three thermocouple gauges, and one MKS Baratron capacitance manometer. The









Bayard-Alpert gauges are used for measurements at high vacuum conditions (<10-5 Torr), and

are therefore used to determine the base pressure of the system. The base pressure is a critical

factor in controlling contamination in the growing film. The other critical gauge is the MKS

Baratron gauge, which gives accurate, precise, fast, and direct measurement of the pressure for

the range of 10-5 to 1 Torr, and therefore spans the pressure range used for sputter deposition. A

direct gauge refers to the fact that the gauge measures the gas pressure directly, instead of using

heat transfer rate to determine the ambient pressure. A direct gauge's calibration is therefore

independent of the gas being measured. This is critical for generating ambient with controlled

partial pressures of a multiple gas species.

Substrate Cleaning

ZnO thin films were deposited on 1mm thick soda-lime glass substrates for

characterization purpose. The substrates are first rinsed in sprays of deionized (DI) water to

remove any major contaminations attached on the surface. They are then vertically placed in a

Teflon substrate holder. The substrate holder is then put in a hot DI water bath at approximately

800C for 10 minutes with all glass substrates completely immersed in the bath. The substrates

are then taken out and dipped in DI water dissolved with Alconox detergent. They are then

individually taken out of the holder and physically scrubbed with a clean brush and then put back

in the holder. Both the sides and edges are cleaned to remove all possible contaminations. The

substrates are rinsed again in sprays of DI water and put in a second hot DI water bath in an

ultrasonic cleaner. All substrates are completely immersed in the bath and the ultrasonic

cleaning takes about 10 minutes. The holder are then removed from the ultrasonic cleaner and

placed in a three stage cascade rinser agitated with nitrogen bubbles. Finally the substrates are

individually dried with nitrogen gun and placed in storage boxes.









For CIS device fabrication, the soda-lime glass substrates coated with Mo, CIS, CdS are

stored in vacuum sealed bags to prevent possible oxidation from air. The bags are only opened

before the ZnO deposition and the substrates are quickly put into the vacuum chamber after they

are exposed in air.

Characterization Techniques

To fully understand the various aspects of semiconductor property, a number of material

characterization techniques are utilized. The electrical, optical and other properties are

comprehensively investigated using the following techniques.

Electrical Property Characterization

Electrical property characterization includes four-point probe measurement to acquire the

sheet resistance of the thin film and Hall measurement to obtain carrier type, concentration and

mobility information.

Four-point probe

The four-point probe contains four thin collinearly placed tungsten wires probes which are

made to contact the sample under test. In a configuration shown in Figure 3-3, a constant current

I flows between the outer probes, and the voltage V is measured between the two inner probes,

ideally without drawing any current.

The equipment used was an Alessi four point probe. It has a probe head with tungsten

carbide tips with a point radius of 0.002", a probe space of 0.05". Current supplier was a

Crytronics model 120 current source which can provide currents from 1CIA to 100mA. Voltage

was measured with a Keithley model 181 nanovolt electrometer. Sheet resistance with units of

O2/square, can be obtained from the four-point probe using the equation

4.532V
Rs -I (3.1)









where V and I are voltage and current respectively and 4.532 is the geometric factor when

the size of the thin film is much larger than the probe spacings. Resistivity p with units of G-cm

can be obtained by multiplying the thickness of the film by Rs, shown in the following equation

4.532Vt
p = Rt =
I (3.2)

Hall measurement

Hall measurement is used to obtain the resistivity, carrier mobility and concentration

information of zinc oxide thin film. The physical theory behind this characterization technique is

the Hall effect. When electric current flows in the presence of a perpendicular applied magnetic

field, the Lorentz force deflects the moving charge carriers to one side of the sample and

generates an electric field perpendicular to both the current and the applied magnetic field. This

is called the Hall effect. The Hall coefficient is the ratio of the perpendicular electric field to the

product of current density and magnetic field, while the resistivity is the ratio of the parallel

electric field to the current density.

A Lakeshore 7507 Hall system was used to conduct the Hall measurement. The samples

were cleaved into squares with dimension 1cmxlcm. They are mounted on Lakeshore Hall

sample mount (Part number 671-201) using double stick tape. Care was taken to make sure the

sample surface is in parallel to the sample surface. Lakeshore Hall sample wires (Part number

671-260) are used for connection between the contact spots on the sample mount and the corners

of the sample. The wires are soldered with small amount of indium which forms Ohmic contact

with the zinc oxide thin film.

Square shaped van der Pauw geometry is used, which is shown in Figure 3-4. In this

geometry, the contacts are on the circumference of the sample and the dimension of the contacts









is made as small as possible. For the square van der Pauw geometry, the estimated correction

factor for resistivity can be shown in the following equation[57]

Ap c .2 (3.3)
p a

where c is the dimension of the contact spot and a is the length of the Hall sample. Efforts

were made to ensure the dimension of the contact spot as small as 1 mm. Therefore, a good

accuracy of the resistivity can be achieved.

Before conducting the variable filed Hall measurement, a forward and reverse I-V curve

measurement is always conducted to verify the Ohmic property of all of the four contacts spots.

Optical Spectrophotometer Characterization

The optical transmission data was measured from the UV region to near IR region using a

Perkin-Elmer Lambda 900 spectrophotometer. The system is equipped with a monochromator

and a chopper, which is used to generate a sample and a reference beam. For all the

measurements, the reference beam path was left open. Background correction leaving both the

reference and sample beam path open is always performed before taking the transmission data.

The samples are cleaned with acetone to remove possible contaminations such as

fingerprints and dust particles, which may cause scattering of the incident light. All transmission

data was taken between 300nm and 900nm using the normal incidence setup.

When an incident light with intensity lo enters the sample, the absorption process can be

expressed as

I = Ioe-A 0 -a~d (3.4)

where I is the remaining intensity and A is the absorbance, which is the product of the

absorption coefficient oc and the sample thickness d. The transmission T is defined as










T fe = ed (3.5)


The absorption coefficient oc can be estimated by measuring the transmission of two

samples with different thickness dr and d2. Then the absorption coefficient is derived using the

equation


a =In( ) (d, -d,) (3.6)


where T1 and T2 are the transmission of sample 1 and sample 2 respectively.

Other Characterizations

Profilometry, X-ray diffraction, atomic force microscopy and Auger electron spectroscopy

are used to characterize various material properties for sputtered zinc oxide thin films.

Profilometry

Because sputtering is a reproducible process, the growth rate can generally be maintained

at a repeatable rate by using identical sputtering conditions. Therefore, there is no in-situ film

thickness monitor set up in the UF zinc oxide sputtering system. The film thickness is

determined using profilometer after the deposition and the growth rate is calculated. With

growth rate obtained from previous runs, the film thickness can be controlled by choosing

appropriate deposition duration.

A Tencor P2-long scan profilometer was used to measure the thickness. Profilometer is a

measurement device that probes the sample surface by a sweeping stylus. The vertical

movement of the stylus is recorded when it runs over the sample surface. This gives the height

of the surface as a function of location.

To measure the thickness of the deposited films, a trench step needs to be created on the

surface of the sample. Generally, two methods were utilized to create the step. The first method










is to cover part of the substrate before deposition and remove the cover later. A portion of the

substrate was either masked with permanent marker or covered with a scotch tape. After the

deposition, the marker was removed with acetone and a swab or the tape was simply removed by

hand, which effectively removed the deposited film on top, yielding a well defined trench step.

The second method to create the step is etching. The film area to be preserved are covered with

tape and the unwanted area are exposed on top of a beaker containing aqua regia. The solution is

heated to 500C and the vapor etches away the oxide creating the step. The profilometer was then

used to measure the height of the step, and therefore the film thickness.

X-Ray Diffraction

X-ray is the generic term used to describe electromagnetic radiation with wavelengths

between roughly 0.01 nm and 10 nm. This is shorter than ultraviolet but longer than gamma

radiation. The spacing of adj acent atoms in a crystal structure is typically about 0.3 nm. The

wavelengths of X-rays are therefore of the same order of magnitude as the atomic spacing in a

cry stal .

X-rays are generated by accelerating electrons toward an anode target under vacuum

conditions. The high-energy incident electrons interact with the target atoms and knock out

some of the core electrons. Then the outer shell electrons fill in the holes left through relaxation

process and the energy difference is released generating X-ray to conserve the energy. The

energy or wavelength of the radiated X-ray is determined by the energy difference between the

two electrons and is not related with the acceleration energy of the incident electrons. The Cu

Ku radiations are widely used in X-ray diffractometers. It is generated via transition between

the Cu L shell electrons and K shell electrons. The intensity ratio between Kal and Ku2

radiation is 2:1. Figure 3-5 shows the schematic how Cu Ku radiations are generated. Other









series of X-rays such as KP and L series are generated at the same time. But they are filtered

away and only the Kal and Ku2 are used to generate the diffraction pattern. The wavelengths

of these two radiations are very close and the peaks are overlapped especially at the low 26 angle.

There are X-ray pattern processing software having the capabilities of stripping off the Cu

Ku2 diffraction pattern allowing more accurate evaluation of the data.

X-Ray Diffraction (XRD) can generally be categorized as single crystal XRD and powder

XRD. Single crystal XRD examines a single crystal whose unit cells are perfectly aligned. It

requires careful orientation of the sample in order to obtain the diffraction pattern. X-ray powder

diffraction is more widely used and the general term "X-ray diffraction" refers to the powder

diffraction unless specified otherwise. The samples used for powder diffraction can be either

powders or thin film samples. The diffraction geometry tells that only the crystallites with

planes in parallel to the specimen surface will contribute to the diffraction intensity. An ideal

powder sample consists of a collection of small crystallites randomly distributed with all

available orientations contributing to the diffraction pattern. The diffraction pattern is a

fingerprint of a specific crystalline substance (phase) and the International Center of Diffraction

Data (ICDD, formerly known as JCPDS) maintains a diffraction pattern database of various

minerals, inorganic and organic standards. By comparing the diffraction pattern of an unknown

sample with the data of standards, the possible phases available in the unknown may be

identified. It is noteworthy that XRD is a technique for phase identification. In most cases, it is

still necessary to know the chemical compositions of the sample to accurately identify the phases

in exist. For thin film samples, the peak intensity can vary significantly from the standards due

to preferred orientation. Besides phase identification, careful examination of the peak

broadening can be used to estimate grain size and stress associated with the sample. Finally,










quantitive analysis of the diffraction pattern an be used to estimate the ratio between different

phases in a specimen.

The Bragg's law is the cornerstone of X-ray diffraction analysis. The relation was

discovered in 1912 by W. L. Bragg. Bragg law describes the condition that the scattered rays

from two parallel planes interact with each other in such a way as to create constructive

interference.

As shown in Figure 3-6, the extra distance that ray B must travel is the distance X-Y-Z.

Thus X-Y-Z = n h to create constructive interference. Notice that X-Y = Y-Z = d sine. Hence,

the Bragg's law can be expressed with the equation

h = 2dhkl sine ,t (3.7)

where 26 is the diffraction angle, which is the angle between the incident and diffracted x

rays. In a typical experiment, the diffracted intensity is measured as a function of 26 and the

orientation of the specimen, which yields the diffraction pattern.

Atomic Force Microscopy

Atomic Force Microscopy (AFM) and Scanning Tunneling Microscopy (STM) are two

forms of Scanning Probe Microscopy (SPM), a technique that can reveal information down to

atomic scale by using a sharp probe to scan across the surface of a sample. It is relatively new

techniques developed in 1980s.

Depending on the different operational conditions of the probe during scanning, there are

three operation modes for AFM, namely contact mode, tapping mode and non-contact mode.

During contact mode measurement, a tip mounted at the end of a cantilever scans across the

sample while contacting the surface. The deflection of the cantilever is monitored using a laser

shining on the tip and redirect the laser to a split photodiode, which measures the signal variation









coming from the vertical movement of the cantilever. A feedback control loop maintains a

constant deflection of the cantilever by adjusting the vertical position of either the piezoelectric

scanner or the sample. This position change is monitored and converted to the surface

morphology of the sample. In tapping mode, the cantilever oscillates at or slightly below its

resonance frequency at a position very close to the sample surface. Instead of direct contacting

the sample surface, the tip only taps on the sample surface. The root mean square (RMS) of the

oscillation signal is maintained by the feedback control loop by adjusting the positions of the

scanner of the sample and the position change is converted to morphology. For non-contact

mode, the tip oscillates close to the surface but never touches the sample surface. The oscillation

amplitude and frequency will be changed by the van der Waals force, which extends a few

nanometers above the sample surface. The feedback control loop maintains a constant

oscillation amplitude and frequency by vertically moving the scanners and this movement is

converted to surface morphology.

Contact mode AFM has high scan speed and good resolution, but the lateral forces exert on

the sample surface may damage soft samples. Non-contact mode AFM exerts little force on the

sample surface and is not likely to scrap the sample, but the scan speed is low and the sample

should be extremely hydrophobic to prevent the tip from getting stuck by the adsorbed fluid

layer on the sample surface. Tapping mode AFM has less surface damage than contract mode

AFM and can be used on most samples comparing to non-contact mode AFM. The AFM images

shown in this dissertation were obtained using a Veeco Dimension 3100 AFM operated under

tapping mode AFM. The tips used are tapping mode etched single crystal silicon probes

produced by Veeco.










Auger Electron Spectroscopy

Auger electron spectroscopy (AES) is a technique that can identify chemical composition

information by analyzing the characteristic energy of Auger electrons from an element. Auger

electron spectroscopy has been primarily used to determine the chemical compositions of

materials, but it is being increasingly used to characterize chemical state properties. [58] It is

applied to sputtered ZnO thin film characterization to evaluate the composition of the fi1m and

detect any possible contaminations.

When a core level electron is removed leaving a vacancy, electrons from higher energy

level can relax and fall into the left vacancy. The energy difference between these two energy

levels is released to balance the total energy. This released energy can be transferred to a third

electron, which is excited and escape from the atom. The third electron is called Auger electron.

Auger electrons have characteristic energy relating to the energy difference between involved

energy levels. Three electrons are involved to generate an Auger electron. Therefore, AES can

only analyze materials whose atoms have at least three electrons. Hydrogen and Helium are not

detectable using AES and Lithium is material having the least atomic number required for AES

analysis.

AES is a surface sensitive analytical technique because Auger electrons have low energy

and they can only escape from areas a few atomic layers under the surface. Nevertheless,

combined with ion-beam sputtering, which removes material from the surface, this technique can

provide depth information. Depth profie data was taken by alternating between sputter etching

the sample and collecting small windows of Auger spectra focused on the peaks for elements of

interest.

A PHI model 660 Scanning Auger Microprobe (SAM) was used in this work. It has a

LaB6 Eilament as the thermionic emission electron source. A cylindrical mirror analyzer (CMA)









is used to analyze the kinetic energy of the secondary and Auger electrons. The system is

equipped with an ion pump and normally operates at a base pressure of about 3 x 10-10 Torr. An

Ar ion gun is equipped to sputter off material from the samples for the purpose of either cleaning

the sample surface or depth profiling. During the sputtering process, the system typically works

under a pressure in the range of middle 10-s Torr.





























Figure 3-1. Perkin-Elmer 4400 sputtering system.


Stepper Drive


Figure 3-2. Schematic of the ZnO Sputtering System (Perkin-Elmer 4400).





















L

Figure 3-3. Schematics of four-point probe measurement.


SB loriented out of the pag-e)


Figure 3-4. Square shaped van der Pauw sample geometry.













=_' I =


1 -


I


I


Kal


'
V


I


Figure 3-5. Generation of Cu Koc radiation.


Figure 3-6. Bragg's condition.


L shell


K shell









CHAPTER 4
GROWTH AND CHARACTERIZATION OF ZINC OXIDE

Optimization of Sputter Deposition of Aluminum Doped ZnO

The optical and electrical properties of Al doped ZnO (AZO) can be affected by various

deposition parameters such as base pressure, working pressure and deposition power. Films

obtained under different deposition conditions are routinely characterized with four-point probe

and profilometry to obtain the resistivity value. Various other techniques such as X-ray

diffraction, atomic force microscope, Hall measurement and Auger electron spectroscopy are

employed to further characterize these thin films. The results will be presented in this chapter.

Experiments at Different Base Pressures

Due to the large surface area of the substrate platen, it requires a long period of time for

outgassing. The starting base pressure can vary over a large range depending on the pumping

time. An investigation which extends the base pressure between 6.0x10-7 Torr and 2x10-6 Torr is

carried out. The base pressures are categorized into three regions to better illustrate the obtained

results, namely the low, medium and high base pressure region. The low base pressures

represent approximately 6 x10-7 Torr, obtained by adopting a very long period (typically greater

than 24 hours) of pumping of the sputtering chamber before starting the deposition. This ensures

that a minimum amount of residual water vapor is present during the sputtering process. The

medium base pressure, approximately 9x10-7 Torr, is associated with a shorter pumping period,

under conditions that the chamber' s pressure still continues to drop when the deposition process

starts. In this case, the pumping and outgassing have not reached equilibrium at the beginning of

the deposition, and as a consequence the amount of water vapor present in the chamber is larger

than in the low base-pressure region. Finally, the high base-pressure condition, approximately

2x10-6 Torr, is obtained after a short pumping period, and as a consequence the amount of









residual water vapor is relatively large. All of these three base pressure regions ensure a

reasonable clean environment for the sputter-deposition process.

Optical property

The optical transmission of AZO films deposited under the three base-pressure regions is

measured using Perkin-Elmer Lambda900 spectrophotometer over the wavelength between

300nm and 900nm, the spectrum region where solar light has most of its intensity. Figure 4-1

shows plots of the optical transmission of the resulting films as a function of incident-light

wavelength for the three pressure-regimes investigated. The results clearly demonstrate that

increasing the base pressure improves the optical transmission. For the film deposited under low

base pressure, the average optical transmission between wavelength 300 nm and 900 nm is

below 70%. The two films deposited under medium and high base pressure show better optical

property and meet this requirement. Generally, for solar cell applications, an average optical

transmission of 80% is desired for the visible and near infrared wavelength region.

It is known that the optical transmission and conductivity of zinc oxide is directly related

to the stoichiometry, i.e., the ratio between zinc and oxygen content. Higher concentrations of

oxygen vacancies or zinc interstitials in turn provide higher concentrations of intrinsic carriers

and result in a more conductive film. However, under such circumstances the optical

transmission would be expected to drop. This is exactly what reflected in Figure 4-1, where the

optical transmission indeed increases when the base-pressure favors the presence of water vapor

in the chamber. Under the plasma environment the residual water vapor dissociates and the

active oxygen species from the water vapor contributes the incorporation of oxygen in the

sputtered AZO films. This explains why the optical transmission increases with an increasing of

the base pressure.









Electrical property

Figure 4-2 plots the effect of each of the three base-pressure regimes on the resistivity of

the sputtered AZO thin films. An arbitrary continuous line is drawn between the experimental

points simply to suggest a trend and to aid in the discussion of the results. First consider the

results obtained at the medium and high base-pressure regions. Note that the resistivity

decreases as the base pressure is reduced from the high to the medium region. This result is

expected because under medium base-pressure conditions the AZO fi1ms are more zinc-rich (i. e.,

oxygen defieient) because less water vapor is present, leading the more conductive fi1m.

Next consider the results obtained at the low and medium base-pressure regions. In

contrast to the previous case, the resistivity now increases as the base pressure is reduced from

the medium to the low region. Obviously, in this case a different mechanism must be in playing

a significant role to a point where it causes a reversal in the resistivity trend. The explanation for

this observation is based on the fact that less residual water vapor is present at the low base

pressure, promoting the presence of insufficient amounts of hydrogen. The observation of this

resistivity trend reversal lends evidence indicating that hydrogen plays a critical role in the

resistivity of sputtered AZO thin films. Details of hydrogen effect on sputtered AZO will be

addressed in the next chapter.

Growth rate

The effect of base pressure on growth rate was investigated. A series of films were

deposited under all same deposition conditions except starting from different base pressures,

which are 8 x10-7 Torr, lx 10-6 Torr, 2x 10-6 Torr, and 3 x10-6 Torr respectively. From the results

shown in Figure 4-3, an increase of the base pressure from 8.0 x 10-7 Torr to 3 x 10-6 Torr caused a

growth rate decrease from 3.4 A/lsecond to 2 A/second. This effect is also due to the existence of

water vapor. The oxygen generated in the highly energetic plasma process not only reacts with









the deposited film, it also reacts with the target and increases the oxidation extent of the target. It

is known that the sputtering rate for an oxide target is generally much lower than that of a metal

target under the same sputtering conditions. Thus, the growth rate from a more oxidized target is

expected to be smaller than that from a more reduced target, when all other sputtering conditions

are held constantly.

Power Effect

The effect of deposition power on the resistivity of sputtered ZnO:Al films was

investigated by holding all other deposition parameters as constant except the power.

All other deposition conditions are fixed with target to substrate distance of 5 cm, base pressure

of 8.0x107 Torr, and working gas pressure 4 mTorr.

The results show that the power can directly affect the resistivity. As shown in Figure 4-4,

it is clear that with all other deposition parameters held as constant, the film deposited at 400W

shows lower resistivity than those obtained using other values of power.

The relation between growth rate and deposition was also evaluated and the result is shown

in Figure 4-5. It can be seen that the trend is increasing the deposition power causes the increase

of the film growth rate. This is caused by an increase of the sputter yield upon increasing the

deposition power.

Discharge power is a critical factor that determines the properties of sputtered ZnO. When

sputtering from a compound target, the more volatile component can be more rapidly lost into

the gas phase without being condensed on the substrate. This may result a different

stoichiometry of the film from that of the target. For the case of ZnO sputtering, oxygen is more

volatile than zinc. With a careful adjustment of the deposition power, a desired Zn rich

stoichiometry can be achieved.









Through experiments, it is observed that the deposition parameters, especially the power,

can affect the surface stoichiometry of the ZnO target. For magnetron sputtering of ZnO, when

large power density (greater than 2 Watt/cm2) is exerted on the target for a long period of time,

the target surface could be significantly reduced. Thin films sputtered from a heavily reduced

target can contain strong separated Zn phase. This conclusion is obtained by employing X-ray

diffraction characterization on sputtered thin film from a reduced target. The equipment used

was a Philips APD 3720 powder diffractometer using Cu Ku radiation with h(Kal1) equals

0.154056 nm and h(Ku2) equals 0.154439 nm. The X-ray generation conditions on the

diffractometer were set at 40KV and 20mA.

For survey scanning, continuous scans with a 26 range between 100 to 800, a step size of

0.030, and a step time of 0.5 second was used. Figure 4-6 shows the XRD survey pattern of a

thin film sputtered from a heavily reduced target. The vertical lines indicate the 2 theta angle

and relative intensity of Zn taken from JCPDS database. It is evident that there is Zn phase

available in the resulted thin fi1m. In contrast, Figure 4-7 shows the XRD survey pattern of a

typical ZnO thin film sputtered from an oxidized target. The pattern shows a strong ZnO (002)

peak indicating the sputtered film has preferred orientation. These results show evidently that

the target condition is important determining the property of sputtered ZnO thin fi1m. A control

of deposition power with in certain limit is required to maintain the oxidization status of the ZnO

target.

High resolution continuous scans were also implemented using the following parameters:

26 ranging from 33.50 to 35.50; step size of 0.010; and 1.0 second per step. Figure 4-8 shows the

high resolution scan pattern of three AZO thin films deposited at 300 W, 400 W and 500 W,










respectively. The Full Width at Half Maximum (FWHM) values were obtained from the high

resolution scan and their values are list in table 4-1.

Table 4-1 also contains grain size information. The mean grain size z was evaluated using

Scherrer equation [59] :

KAl
S=~ (4-1)
/7 cos B

where K is a constant called shape factor, whose typical value is 0.9[60]. The term P

represents the line broadening due to the effect of small crystallites. Note it is the additional

broadening in diffraction peaks beyond the inherent peak widths due to the instrument effect.

The instrument broadening contribution can be obtained by measuring a standard material that

has no significant stress and grain size effect. Here for approximation, only the measured

FWHM in radians was used. The other two terms h and 6 are the wavelength of the X-ray and

the diffraction angle. With this simplification, the estimated grain sizes are 36.3, 32.3, and 30.7

nm for films deposited using 300W, 400W and 500W, respectively.

Working Pressure Effect

Similarly, the relationship between working pressure and film resistivity is investigated by

changing the working pressure while fixing all other deposition parameters. Experiments are

done under four working pressure evenly distributed between 2 mTorr and 5 mTorr. The results

are shown in Figure 4-9. It is clear that 4 mTorr working pressure gives the lowest film

resistivity for this specific experiment set.

The film growth rate is also dependent on working pressure. As shown in Figure 4-10, the

general trend is increasing the working pressure would lead to an increase of the film growth

rate. This effect is especially strong at the relatively low working pressure region with pressure










changing from 2 mTorr to 3 mTorr. The curve start to flat out at high pressure region and only a

slight increment was observed when increasing the pressure from 3 mTorr to 5 mTorr.

High resolution XRD scans of films deposited with different working pressures are shown

in Figure 4-11. The three working pressures used are 2 mTorr, 3 mTorr and 4 mTorr. All other

deposition conditions are fixed with target to substrate distance of 5 cm, base pressure of 4.0x10-

STorr, and deposition power of 500W. It is found that the measured FWHM of the three XRD

rocking curves decreases with the increased working pressure, indicating larger grain size can be

achieved by using a higher working pressure.

Characterization

Besides four-point-probe, profilometry and X-ray diffraction, other characterization

techniques were used in analyzing sputtered ZnO thin films. Auger electron spectroscopy was

applied for chemical composition and depth profile analysis. Atomic force microscopy was used

to evaluate surface morphology properties.

Auger Electron Spectroscopy Analysis

Auger electron spectroscopy was utilized to verify the chemical composition of the

deposited films and to detect any unexpected impurities that may exist in the deposited films.

Results reveal that the films contain only Zn, O and Al in the bulk, with C contamination only in

the surface. The source of the carbon is organic contaminations residing on the surface of the

sample. Depth profile analysis shows a uniform distribution ofZn and O through the film,

indicating good stoichiometry control during the deposition.

Auger electron spectroscopy is a surface sensitive analytical technique because the Auger

electron, generated with a small energy, can only escape from a few atomic layers on the top.

Like SEM, Auger electron spectroscopy uses a electron beam hitting the sample to generate

secondary and Auger electrons. The resulted secondary electrons can be collected and









transferred into secondary electron images, which is just like what a scanning electron

microscopy does. Beyond that, the Auger electrons are analyzed by a cylindrical mirror analyzer

(CMA) and transferred to information regarding the chemical composition.

The distance between the CMA and the sample surface was always calibrated using the

elastic scattering peak from a 3KeV electron beam before taking the survey spectrum. Typical

energy of the electron beam used for survey spectrum was varied between 5KeV to 10KeV. The

spectrum of counted electrons at an energy (n(E)) versus kinetic energy (KE) were taken with an

energy step size of 1.0 eV, and a dwell time of 30ms at each energy step. Ten sweeps of this

spectrum were taken over the energy range of 50 to 2050 eV and the average value was used for

numerical processing. This spectrum is then numerically differentiated using algorithm from the

computer acquisition software (AugerScan) to yield the spectrum in terms of d(n(E))/d(E) versus

kinetic energy. The differentiated spectrum is then used to compare standard results and

consequently determine the elements present on the sample's surface.

Surface scan

Figure 4-12 shows a survey spectrum of an AZO thin film. The differentiated energy

distribution d(n(E))/d(E) was plotted against the kinetic energy. The key instrument parameters

used for this survey were as follows: e-beam acceleration voltage 5kV, probe current 50nA, and

sample tilting 30 degrees. The Auger peaks of C, O, and Zn are identified and labeled in the

figure. The result verifies the film composition as Zn, O and it also suggests that there is carbon

contamination at the surface of the ZnO film. The source of carbon is from organic

contaminations on the sample surface. These contaminations can be reduced by careful handle

and storage of the sample after it is taken out of vacuum.

Al peak was not distinguishable from Figure 4-12. Various factors are responsible for it.

First of all, the noise signal level is quite significant due to the surface contamination. Secondly,









Al only constitutes less than 2% of the target, therefore, it is estimated the film would only

contain a relatively small amount of Al. The third reason is Al has a small relative sensitivity

factor Sx (~0.09, 10keV) compared with Zn (~0. 16, 10keV) and O (~0.4, 10keV).

Ar ion beam was used to sputter the sample surface for 3 minutes to remove the surface

contamination. Another survey spectrum was taken using identical survey conditions and the

result is shown in Figure 4-13. Compared with result shown in Figrue 4-12, it is clear that the

carbon peak is no longer present. This verifies the previous assumption that carbon is only a

surface contamination and not part of the film constitution. Ar peak is detected indicating

implanted Ar from the ion beam exist on the surface of the sample. Comparing with previous

results, the noise signal in the second survey (shown in Figure 4-13) is much smaller than that in

the previous survey (shown in Figure 4-12). Considering the fact that two scans used identical

survey parameters, the noise reduction was achieved through the removal of surface

contaminations. Semi-quantitive data of the concentration of each species detected are also

labeled in the diagram.

Al sensitivity

Because of the reasons presented earlier, the aluminum peak was not clear in either of the

previous cases. To identify the aluminum peak and verify its existence, another scan was

conducted using higher e-beam energy. Increasing the acceleration voltage to 10kV, the new

survey result is shown in Figure 4-14. In this spectrum, Aluminum Auger peak at approximately

1375 eV kinetic energy was clearly distinguished. The increased beam energy caused more

excitation of Al Auger electrons, which increased the size of the peak, making it distinguishable

from the noise signal.









Depth profile

Depth profile was measured to examine the distribution of those main elements along the

depth of the sample. Depth profile analysis can be performed using the window method by

alternatively sputter etching the sample and collecting small energy windows of Auger spectrum

focused on the peaks for elements of interest. The energy window is defined using the 3-point

method, which uses two background energy position and one peak energy position to define the

area of the peak. The system uses the difference between the peak value and the background

values to evaluate the intensity of the peak. This intensity is then plotted as a function of the

sputter etching time, giving a distribution of the element distribution along the depth direction.

Parameters used for collecting depth profiles are presented in Table 4-2. The sputtering rate can

be controlled by adjusting the ion beam parameters. With the Ar ion beam energy of 3KeV,

beam current of 90 nA and raster pattern of 3x3 mm, the sputter etch rate is estimated to be

approximately 100 A/min.

Figure 4-15 shows the Auger depth profile of a ZnO thin film. 30 sputtering cycles was

used in the particular measurement and during each sputtering cycle, the sample is sputtered for

6 seconds. Therefore, the estimated sputter etch rate is approximately 10 A/cycle and a total

depth of approximately 300 A+ into the bulk of the film was characterized. The results show that

carbon contamination is highest on the surface, and that it stablizes at nosie level at the position

approximately 100 A+ below the surface. The depth profile analysis also shows that the

concentration of Zn and O inside the bulk of the film is uniform.

Atomic Force Microscopy

Atomic force microscopy was applied to examine the surface morphology of sputtered

AZO thin films. A Veeco DI Dimension 3100 system was utilized for this purpose. The surface










of the sample was carefully cleaned using acetone to remove attached contaminations before the

examination.

The AFM image was digitally processed before any analysis is performed to eliminate the

effects coming from artifacts such as tilt and bow. Tilt happens when the sample is not perfectly

leveled and bow is generally caused by large area scans using a large scanner. These unwanted

effects need to be eliminated before surface roughness analysis is performed.

A "Flatten" command from the diNanoScope Software 7.0 was first executed to eliminate

unwanted features from scan lines. A third order flatten is used. The software calculates a third

order least-squares fit then subtracting it from the scan line. This effectively removes the Z

offset between scan lines and also eliminates sample tilt and bow from each scan line. After

flattening, "Plane Fit" command which computes a polynomial of a selectable order for an image

and subtracts it from the image was used. A third order "Plane Fit:" command was applied on

both X and Y direction. Arithmetic roughness Ra and root mean square roughness R, values are

then obtained from the roughness analysis. Ra is the arithmetic average of the absolute value of

the surface height deviation and R, is the standard deviation of the surface height value. They

can be expressed using the following equations:


R (4-1)



R (4-2)
4 N

where zi is the height value of a specific position and N is the total number of positions

evaluated.

Figure 4-16 shows the AFM micrograph ofAZO thin films deposited using identical

deposition parameter except the working pressure. Sample a,b, and c are grown are Corning









1417 glass substrates under working pressure 2mTorr, 3mTorr and 4mTorr, respectively. Care

was taken to ensure the samples have similar thickness to eliminate the thickness effect on

surface morphology. It is found that with the increase of the working pressure, larger grain size

is found on the sample surface. Generally, larger grain size is favorable for polycrystalline AZO

because it can be beneficial to reduce grain boundary effect and therefore increase the electron

mobility, which then improves the conductivity of the film. The surface roughness also increases

along with increasing the working pressure. With a working pressure change from 2mTorr to

4mTorr, the root mean square roughness increased from 2.2nm to 3.6nm. Details of the

roughness data change with the working pressure is shown in Table 4-3.

The AFM analysis shows that 4mTorr is a favorable working pressure for larger grain sizes.

Since larger grain size tends to yield a better conductivity, this is consistent with the resistivity

characterization presented earlier, which indeed showed that AZO sputtered using 4mTorr

working pressure has lower resistivity.





20O

300 400 500 600 700 800 900
Wavelength (nm)
Figure 4-1. The effect of base pressure on the optical transmission of sputtered AZO
films.


Low Medium High
Base pressure


Figure 4-2. Effect of base pressure on the resistivity of sputtered AZO films.





3.5




S2.5

2-

1.5 -
5.00


10.00 15.00 20.00 25.00 30.00


35.00


Base pressure (x 10 Torr)


Figure 4-3. The effect of base pressure on the growth rate of ZnO:Al thin films
(Fixed conditions: Dts = 5 cm, P = 4 mTorr, Power = 400W).


300


400


500


Power (W)


Figure 4-4. The effect of power on the resistivity of as grown ZnO:Al thin films
(Fixed conditions: Dts = 5 cm, Pb = 8.0 x10-7 Torr, and P = 4 mTorr).






















r;


,4


400


Power (W)




Figure 4-5. The effect of power on the growth rate of ZnO:Al thin films

(Fixed conditions: Dts = 5 cm, Pb = 8.0 x10-7 Torr, and P 4 mTorr).


20 30 40


400-


350-


300-

c
3
o ?50-

L
cr, 200-
c
m
E
150-


100-


50-

0


50 60 70


2-Theta


Figure 4-6. X-ray diffraction pattern of thin films deposited from a reduced ZnO target



































uluw~l~~"~LIl~~~lI.~-~-~Y1- YI_LILILIII~~Y~Y [~~~YY~~IIYLYI~/


1000


500


'1'~'


'~~' '''1-'117'


2-Theta

Figure 4-7. X-ray diffraction pattern of thin films deposited from an oxidized ZnO target


10000 -

9000

8000
7000

6000

5000

4000

3000
2000

1000


33.5


34.0 34.5 35.0 35.5

26


Figure 4-8. High resolution XRD pattern for AZO films deposited using different power
(Fixed conditions: Dts = 5 cm, Pb=8.0 x 10-7 Torr, Pressure 4mTorr).















7r1T


Working pressure (mTorr)


Figure 4-9. The effect of working pressure on the resistivity of ZnO:Al thin films
( Fixed conditions: Dts 5 cm, Pb = 8.0 x107 Torr, and Power = 500 W )


-~7r1r


Working pressure (mTorr)


Figure 4-10. The effect of working power on the growth rate of ZnO:Al thin films
( Fixed conditions: Dts 5 cm, Pb 0 8.0 x 10-7 Torr, and Power = 500 W )















16000


14000

12000

10000

8000

6000

4000

2000

0-
33.5


34.0 34.5 35.0

26


35.5


Figure 4-11. High resolution XRD pattern of AZO thin films deposited at different working
pressures (Fixed conditions: Dts = 5 cm, Pb=4.0 x 107 Torr, Power 500 W).


50 250 450 650 850 1050 1250 1450 1650 1850 2050
Kinetic Energy (eV)
Figure 4-12. AES survey scan pattern for an AZO thin film with organic contamination on the
surface .






















0.7%~


50 250 450 650 X50 1050 1250 1450 1650 1850 2050
Kinetic Energy (eV)
Figure 4-13. AES scan results after sputtering for 3 minutes.


A12
5.3%~


14.0% o


23.1 %


50 250 450 650 850 1050 1250 1450 1650 1850 2050
Kinetic Energy (eV)
Figure 4-14. Al Auger peak shown by using increased primary beam energy




























0 5 10 15 20 25
Cycles
Figure 4-15. Depth profile for each element of AZO obtained with AES.





















20 Dnni 20.0 nm
10nm 1 10 Inm
D4 nm 00 Onm

400 400
500 nhi \ 500mi
300 \ 300
400 -~400
200 200
300 300
\100 100II lo
2001 20C

100 500 nm 100? 500 nm
300 300







a) b)





20.0 nm
0nm
0.0 nm

400
500 af

400

300
100



1001 0nm
0400







c)

Figure 4-16. AFM graph of AZO thin films deposited with different working pressure. a)

2mTorr, Rq=2.2nm; b) 3mTorr, Rq=2.8nm; c) 4mTorr, Rq=3 .6nm.












Table 4-1. Estimated grain size using XRD data obtained for AZO deposited at different power.
Deposition power FWHM Grain size
300 W 0.2290 36.3 nm
400 W 0.2570 32.3 nm
500 W 0.2700 30.7 nm


Table 4-2. Parameters used for collecting the Auger depth profile ofAZO thin films.
Peaks of Peak energy Background El Background E2 Dwell time
interest (eV) (eV) (eV) (mS)
01 506.0 523.0 479.0 250
Zn1 985.0 1017.0 975 250
Cl 264.0 298.0 200.0 250

Table 4-3. AFM surface roughness evolution with working pressure.
Working pressure (mTorr) R, (nm) Ra(nm)
2 2.2 1.7
3 2.8 2.3
4 3.6 2.9


Number of
sweeps
2
2
2









CHAPTER 5
EFFECT OF HYDROGEN ON SPUTTERED ZINC OXIDE

Motivation from the Base Pressure Effect

The observation of base pressure's effect on AZO resistivity indicates that residue water

vapor plays a critical role in determining the resistivity of sputtered zinc oxide thin film. Under

the plasma environment during sputter deposition, the water vapor decomposes into hydrogen

and oxygen. With the base pressure variation in the order of 107 Torr, the change of oxygen

coming from H20 is quite small and would not affect the amount of oxygen in the resulted film

in dramatic fashion. Therefore, it is highly suspected that hydrogen, the other ingredient from

H20 plays certain roles in sputtered ZnO.

In most semiconductors interstitial hydrogen presents itself as an amphoteric impurity [61],

and acts in a fashion that counteracts the prevailing conductivity. This behavior makes hydrogen

an unusual candidate for semiconductor doping. However, in the case of zinc oxide, Van de

Walle [62] predicted that interstitial hydrogen can provide a donor level just below the

conduction band and that therefore it could contribute to enhance n-type conductivity in this

material. Experimentally, hydrogen implantation experiments showed the change of carrier

concentration and mobility in ZnO films [63]. However, hydrogen doping ofZnO needs to be

experimentally implemented in our Perkin-Elmer sputtering system to see if a more conductive

film can be obtained.

Experiments and Results

To directly investigate the effect of hydrogen on the electrical properties of sputtered AZO

thin films, two groups ofAZO films were deposited on soda-lime glass using identical

conditions, except the composition of the sputtering gas. One group was deposited using UHP

argon only, and the second group using a gas mixture of argon and 0. 1% hydrogen gas. Special










care was taken trying to maintain an identical deposition conditions. This includes identical

sputter parameter, deposition duration, substrate position, etc. The relevant experimental

conditions are given in Table 5-1. The soda-lime glass substrates were thoroughly cleaned using

the method described in Chapter 3 before depositing with the AZO films.

The resistivity obtained through four-point-probe and profilometry measurement of thin

films from the two groups is plotted in the graph presented in Figure 5-1. AZO stands for the

thin films sputtered using pure argon and H-AZO represents those grew with argon and hydrogen

mixture.

The data reported in Figure 5-1 clearly demonstrates that the use of an argon-hydrogen gas

mixture significantly decreases the resistivity of the sputtered AZO thin film. This observation is

consistent with the claims made previously regarding the beneficial effect of the presence of

hydrogen from the viewpoint of enhancement of the film conductivity. Hence, the data provides

a strong experimental evidence to justify the claims of hydrogen' s beneficial effect on sputtered

AZO thin films in terms of lowering the resistivity.

Hall Characterization

To obtain an in-depth understanding of the mechanism on this improvement of

conductivity, Hall measurements are done to evaluate the carrier concentration and mobility

change between the AZO films deposited with pure Ar and Ar/H2 gaS mixture. Special care was

taken to make sure the samples were taken from areas with identical positions under the target.

This ensures the elimination of possible effects coming from the location difference.

Figure 5-2 shows Hall measurement results obtained for two AZO films. The sample

denoted as AZO in the figure was deposited using only argon as the working gas, and the sample

denoted as H-AZO was deposited using 0.1% hydrogen in the working gas. The results show

that the carrier concentration increased from 2.5x1020cm-3 for the AZO film to 4.7x1020cm-3 foT









the hydrogen doped H-AZO film while the hall mobility dropped from 15.1 cm2/(V s) to 12.7

cm2/(V-s). As a result, the resistivity decreased from 1.67x10-3G'Cm for the AZO sample to

1.05x10-3G'Cm for the H-AZO sample. The resistivity value is consistent with that obtained

using four-point probe measurement. It is evident that hydrogen improved the conductivity

through increasing the carrier concentration inside the AZO film, indicating that it either acts as a

dopant or increases the doping efficiency of aluminum for these sputtered ZnO thin films.

AFM and SEM Characterization

Atomic force microscopy was used to analyze the surface morphology of ZnO sputtered

using different sputtering gas. A scan area of lxl ym was used. Like mentioned in Chapter 4,

the "Flatten" and "Plane Fit" commands from the diNanoScope Software 7.0 were applied to

eliminate unwanted features such as tilt and bow. The root mean square roughness R, and the

arithmetic roughness Ra of the sample surface were then evaluated. The definition of these two

parameters can be found in Chapter 4.

Figure 5-3 shows a comparison of surface morphology between AZO sputtered with Ar

(referred to as AZO in the figure) and Ar+0. 1%H2 (referred to as H-AZO in the figure). They

are grown on soda-lime glass cleaned using the procedure described in Chapter 3. Both films

have similar thickness of 600nm and they were deposited using identical deposition conditions

except with different sputtering gas. Details of the deposition conditions are listed in Table 5-2.

The result shows that the AZO sputtered using Ar and H2 miXture Shows a much larger

grain size compared to the film sputtered using Ar only. The RMS roughness R, and Ra of these

two films are compared in Table 5-3. It can be seen that the root mean square (RMS) roughness

R, of H-AZO film almost doubled that of the AZO film. This indicates that a much larger grain

size can be achieved through using the Ar and H2 mixture gas.









The thickness effect on surface morphology of sputtered ZnO was also investigated.

H-AZO films with different thickness deposited using Ar and H2 mixture gas were characterized

and the resulted AFM micrographs are shown in Figure 5-4. It is clear that increasing the film

thickness would tend to yield larger grain size. The R, and Ra values of ZnO films with different

thickness are shown in Table 5-4. The trend is that the surface roughness increases when the

film gets thicker.

Figure 5-5 shows a typical cross-section SEM image of a H-AZO thin film grown on i-

ZnO substrate. It can be seen that the film contains large column shaped grains grown

perpendicular to the substrate.

Film Uniformity Study

To investigate the uniformity related issue, aluminum doped zinc oxide thin films were

deposited onto large size glass substrate covering areas up to 3 inches away from the center axis

of the target. The deposition duration was 30 minutes.

The thickness of the film was measured every 0.5 inch along the radius and the results are

shown in Figure 5-6. Sheet resistance was measured using four-point probe at identical positions

where the thickness was measured and the result is shown in Figure 5-7. Resistivity was then

calculated and its variation with position is shown in Figure 5-8. In these figures, a position of 0

inch indicates the position directly below the target center. The number in front of a position

indicates the distance between this position and the 0 inch center point. The positive or negative

sign of this number indicates the position locates at the right hand side of the center point or the

left hand side, respectively. For example, a position with a value of -1 inch denotes the location

one inch away from the center point sitting on the left hand side.

From Figure 5-6, it is clear that the film thickness is the lowest at the center point and it

increases with an extended distance from the center point until it reaches its maximum value of









about 600nm at the location approximately 2.5 inch away from the center point. After passing

this maximum point, the thickness starts to decrease with increasing the distance from the center

pomnt.

Figure 5-7 and Figure 5-8 indicate that the sheet resistance and film resistivity show

similar distribution along the axial direction. The resistivity at the center point is the highest and

it decreases along the direction away from the center. This trend persists until a minimum value

is reached at the position approximately 2-2.5 inch away from the center point. After this point,

the resistivity starts to increase again.

To further investigate the mechanism behind the resistivity distribution, Hall measurement

was utilized to reveal the carrier concentration and mobility variation on films with different

positions under the target. Two samples were cleaved off the large substrate from positions with

the highest and lowest resistivity indicated by four-point-probe measurement results. The

comparative results are shown in Table 5-5. It is found that the film deposited directly under the

center of the target has a much lower electron mobility and relatively higher carrier concentration

comparing to that deposited 2.5 inch away from the center. As a result, the lower electron

mobility directly leads to a higher resistivity. The lower electron mobility of the film deposited

at the center point could be caused by the heavy bombardment of the ions from the plasma

during the sputtering deposition process. It is noteworthy that for hall measurement, the

measured carrier concentration is directly related to the van der Pauw sample thickness, which is

a manual input before the measurement. Due to the nature of non-uniformly distributed film

thickness along the Hall sample, it is difficult to use an accurate description of the Hall sample

thickness. Thus, some errors could exist on the carrier concentration results.









The resistivity obtained by using Hall and four-point-probe measurements is compared in

Table 5-6. It can be found that the resistivity values obtained from different methods fall in a

reasonable close range. The difference between results obtained using these two measurement

techniques could come from the uncertainty of the position where the Hall samples were cleaved

from.

Summary

The experiments on base pressure effect on sputtered AZO thin films are extended to very

low base pressure region. The resistivity obtained at this region show different trend from those

at higher region presented in Chapter 4. Results suggest that hydrogen emanating from residual

water vapor affects in a favorable fashion the resistivity. This supports Van de Walle's theory on

hydrogen's doping effect on zinc oxide thin films.

To direct investigate hydrogen' s effect, a different sputtering gas consisted of UHP argon

and 0. 1% hydrogen was utilized to produce AZO thin films. Properties of resulted AZO thin

films were compared with control sample. Hall measurements show an increase of both carrier

concentration and mobility, resulting a much more conductive thin film. This directly provides a

second piece of evidence confirming the claimed role of hydrogen.

The results are particularly relevant from practical viewpoint, given that it is

demonstrated that improved optical and electrical properties can be obtained by the addition of

hydrogen into the working gas and/or the selection of a relatively high base-pressure that permits

the presence of residual water vapor in the chamber. This identifies a path towards the

development of a more consistent method for depositing good-quality transparent AZO films via

magnetron sputter deposition method. Further work needed to achieve this goal include

experimental studies aimed at characterizing the optimal ranges of hydrogen content in the

working gas as well as the optimal range of operating base-pressures. This is a feasible goal









given that the experimental evidence provided here conclusively demonstrates that the two

variables in question have a significant effect on the electrical and optical film properties.

The film uniformity distribution was investigated on AZO thin films sputtered using Ar

and 0.1% H2 gaS. Non-uniform distribution of the film thickness and resistivity was observed

over a range covering up to 3 inch away from the center point under the target. It is found that

the center point has the lowest film thickness and highest resistivity. The difference between the

highest and lowest resistivity is within a factor of two for freshly deposited films. This result is

helpful for determining the feasibility of large area AZO coating using this equipment for large

size solar module applications.
















r,20

15
. 1



0~t


AZO


H-AZO


Figure 5-1. Effect of inj ecting hydrogen gas on the resistivity of sputtered AZO films.


AZO H-AZO


Figure 5-2. Hall measurement results for aluminum-zinc-oxide thin films deposited using only
argon as the working gas (sample labeled as AZO) and using a mixture of argon with
0. 1 wt% of hydrogen (sampled labeled H-AZO).




















40 0 nrim 40 0nm
0 0 nrn IDn 0.0 nm 0n


750 750 75


250o 50~ 10 nn 20 5 10n
500 500


250 250





a) b)




Figure 5-3. AFM image of two AZO films deposited using different sputtering gas. a) AZO b)
H-AZO
































a) b)



10.0 nm
40 0nm
0.0 nmrr
0.0 nm

1000 n 7 00 r


T 50 0 m500
i~750

250 -- 500 250 500
20 __-- 250

c) d)
Figure 5-4. AFM image ofH-AZO thin films with increasing thickness. a)200nm b)400nm c)600nm
d)800nm








































Figure 5-5. Cross-section SEM image ofH-AZO thin films grown on i-ZnO substrate showing

large column shaped grain structure.




















-

-


-

- ------ -------- -


-


550

500


450

400


350

300


Position (inch)




Figure 5-6. Variation of AZO thickness with different positions along the axial direction.


Position (inch)


Figure 5-7. Variation of sheet resistance with different positions along the axial direction.



















e r


-2 -1


Position (inch)



Figure 5-8. Variation of resistivity with different positions along the axial direction.












Table 5-1. Growth parameters comparison for AZO and H-AZO thin films.
Film composition Base pressure Working pressure Power density Sputtering gas
AZO 1.0x10-6 Torr 4 mTorr 350W Ar only
H-AZO 1.0x10-6 Torr 4 mTorr 350W Ar + 0.1% H2

Table 5-2. Deposition condition comparison between two AZO thin films shown in Figure 5-5.
Film Power Working pressure Deposition duration
AZO 350 W 4mTorr 30Omin
H-AZO 350 W 4mTorr 30Omin

Table 5-3. Thickness, Rq and estimated grain size comparison between two AZO thin films
shown in Figure 5-5.
Film Thickness R, Ra
AZO ~600 nm 2.89 nm 2.34 nm
H-AZO ~600 nm 5.85 nm 4.63 nm

Table 5-4. Roughness data evolution with increasing film thickness.
Thickness R, Ra
~200 nm 2.38 nm 1.89 nm
~400 nm 4.15 nm 3.38 nm
~600 nm 5.85 nm 4.63 nm
~800 nm 8.18 nm 6.51 nm

Table 5-5. Hall measurements results on two films with different positions under the target.
Position Carrier concentration Hall mobility Resistivity
center 6.5 x1020 (Cm-3) 4.79cm2/(V-s) 2.00 x10-30- Cm
2.5 inch from the center 4.3x1020 (Cm-3) 14.2cm2/(V-s) 1.03 x 10-3 0Cm

Table 5-6. Comparison of Hall and four-point-probe measurement results.
Position Resistivity obtained Resistivity obtained from
from Hall measurement four-point-probe measurement
center 2.00 x10-30- cm 1.86 x10-30- Cm
2.5 inch from the center 1.03 x10-30-cm 9.74x10-40-cm









CHAPTER 6
DEVICE FABRICATION AND CHARACTERIZATION

Device Fabrication Procedure

Substrate

The fabrication of CIGS solar cells starts from choosing the right substrate. Soda-lime

glass (SLG) is generally preferred due to its chemical inactivity, thermal stability, and vacuum

compatibility. Soda-lime glass contains 15.6 wt % of sodium in the form of Na20. At the high

processing temperatures used for absorber growth, soda-lime glass starts softening. The sodium

ions can diffuse through the molybdenum back contact entering the absorber layer. It is also

found that an optimum amount of diffused sodium is beneficial to improve the performance of

the devices [64]. It is noteworthy that CIGS solar cells can also be fabricated on flexible

substrates such as metal foil and polymer substrates. Sodium doping may be required to achieve

high efficiency from devices fabricated on these sodium free substrates.

Back Contact Sputtering

Sputtering is generally used for coating the thin layer of molybdenum back contact.

Molybdenum is chosen because it forms good ohmic contact with CIGS. It is resistant to react

with the elements contained in CIGS, and stays stable in the highly corrosive Se vapor

environment during the absorber growth. The working pressure used during sputtering of Mo is

found to be related to the diffusivity of sodium. Increasing the sputtering pressure generally

increases the amount of sodium that can diffuse through. Since there is an optimum amount of

sodium diffusion for best device performance, the Mo sputtering pressure needs to be adjusted

well.

All the UF CIGS devices are fabricated on Mo coated soda-lime glass provided by Shell

Solar Industries (SSI). Small amount of Mo was sputtered using the Perkin-Elmer system for









test purpose. For diode RF sputtering (non magnetron), with the uniformity shield blocking part

of the flux, a growth rate between 10 to 15 nm/min can be achieved using 4mTorr working

pressure and 400W deposition power. The substrate is heavily heated during sputtering and the

substrate temperature is elevated beyond room temperature although the substrate platen is

constantly cooled with chilled water. The resulted Mo film generally has good adhesion on

soda-lime glass. Resistivity in the low 10 O-2cm region was achieved on 150nm thick film.

Absorber Growth

The next step of fabrication is to deposit the CIGS layer. Vacuum processes are typically

adopted for better device performance. Such vacuum processes are generally physical vapor

depositions (PVD) including co-evaporation and sputtering [65,66]. The current champion

CIGS devices are fabricated using absorber deposited through 3-stage process, which is basically

a co-evaporation process [67]. Alternative pathways are selenization of metal precursors within

a selenium environment [68-70]. Non-vacuum processes [71-73] are also reported to be feasible

in producing the CIGS layer. The non-vacuum processes generally produce devices with less

efficiency, but the production cost is lower than its vacuum counterparts. They could prove to be

a promising way of producing low cost solar cells for terrestrial applications.

Here in the University of Florida, the CIGS layer is deposited using a Plasma Migration

Enhanced Epitaxy (PMEE) system [74,75]. Figure 2-1 shows a schematic of the PMEE reactor.

The system is a variant of traditional Molecular Beam Epitaxy (MBE) system. The substrate is

put on a rotating platen and it sequentially passes under each source. The metal components, i.e.,

copper, indium, gallium are sequentially deposited during the exposure. The substrate then

rotates into the selenium zone and encounters a surplus of selenium vapor. At the high substrate

temperature (up to 4900C) used during the deposition, the components reacts and polycrystalline









CIGS is formed. The substrate is only exposed to each source once during every rotation cycle

and the system has a very low throughput compared with other PVD processes such as

co-evaporation. It is not economical for large scale production using the system but the epitaxial

ability makes the system potentially capable for exploration into the single crystal regime.

Single crystal growth of CIGS can be performed on GaAs substrates.

Buffer Layer Deposition

Following the absorber layer is the deposition of buffer layer. A wet chemical process

called chemical bath deposition (CBD) was utilized to deposit a thin layer of n-type buffer layer

on top of CIGS. CBD is a technique of controlled precipitation of a compound from its aqueous

solution. It is a relatively economic way to deposit thin films because no vacuum or high

temperature is required.

Traditional buffer layer for CIGS solar cell is cadmium sulfide (CdS). For CdS deposition,

solutions containing certain concentrations of ammonia hydroxide, CdSO4 and thiourea are

mixed in a water circulated 1 Liter j acket beaker and reaction took place in a controlled batch

under a preferred temperature of 650C. CdS with approximately 50nm thickness is deposited on

to the surface of the substrate during a 15minute deposition time. Table 6-1 shows detailed

recipe for traditional CBD deposition of CdS buffer layer.

Due to the toxic nature of Cd, there have been efforts to remove or at least reduce the toxic

cadmium component and produce more environmentally friendly CIGS solar cells. For the result

presented in this dissertation, an alternative buffer layer of Cdl-xZnxS was used. The targeted

stoichiometry is a ratio between Cd and Zn as 7:3, which gives a good device performance [76].

The recipe used is similar to that in a patent from Beoing company [77]. It is basically an

aqueous solution contains CdCl2 2(1/2)H20, NH4C1, thiourea (H2NCSNH2), ZnCl2, and NH40H









was used for the UF CBD deposition. Detailed concentration of each ingredient is documented

in Table 6-2.

The chemical bath is formed in a 1 liter j acket beaker heated by circulating hot water and

the temperature is maintained at a target of 85 oC at all times during film growth. The bath is

vigorously stirred and typical deposition duration is 45min, which gives a film thickness of

approximately 50nm.

Window Layer Deposition

After depositing the buffer layer, an intrinsic ZnO and an Al doped ZnO (AZO) layer are

sequentially deposited using a Perkin Elmer 4400 sputtering system. Details of the sputtering

system are introduced in Chapter 3. It is known that both the optical transmission and electrical

resistance of the window layer have strong impact on the performance of CIGS solar cells. The

transmission determines the amount photons entering the absorber layer while the resistance is

related with the series resistance of the device. Both of these factors are associated with window

layer thickness in a conflicting way. Decreasing ZnO thickness would help to improve

transmission but it also causes the increase of sheet resistance, which may result a deterioration

of device performance. Therefore, the thickness of the AZO layer is adjusted to obtain a

desirable sheet resistance while maintaining a reasonable optical transmission.

The intrinsic ZnO layer is deposited from a fully oxided target and the growth rate is low.

It will dramatically improve the throughput if this layer can be omitted. Ramanathan et al. [78]

reported that the conversion efficiency of CIGS solar cell does not change using a simplified

device structure without the intrinsic ZnO layer. This could be a major simplification of the

fabrication process and could significantly reduce the cost and turn around time. However, there

is contradictory report from Ishizuka et al. [79]showing different intrinsic ZnO thickness would









affect the performance of CIGS solar cell. An investigation on this issue is addressed in the

work of this dissertation.

Metallization

To finish up the device, a dual metal layer contact grid consisting of nickel and aluminum

is deposited onto ZnO:Al. A shadow mask is used to define the contact finger pattern. Nickel

acts as a diffusion barrier to keep aluminum oxide from entering the zinc oxide layer. The high

melting temperature of Nickel is out of the reach of typical thermal evaporators and makes it

necessary to use e-beam evaporation. The thickness of Ni and Al are 50nm and 1Clm

respectively. The grid coverage should be small (approximately 5%) to allow light passing. To

improve the cell performance, an MgF2 anti-reflection layer can be evaporated to reduce the light

reflection and allow maximum amount of light entering the device. For all the results shown in

the scope of this work, no anti-reflection layer was used. Figure 7-2 shows the photo of a CIGS

device completed at the University of Florida.

Device Characterization System

The device current-voltage (I-V) measurement was conducted in our optoelectronic lab.

The I-V system uses an ELH (tungsten-halogen) lamp with an integral dichroic rear reflector as

the solar simulator. It is classified as a class C solar simulator due to a lack of energy below the

wavelength of 0.5 Clm and excess energy in the wavelength between 0.6 and 0.7 Clm compared to

AM 1.5G radiation.

The radiation intensity from the solar simulator was calibrated using the reference cell

method. Basically, a calibration cell with NREL certified characterization report was used as the

reference cell to calibrate the system. This calibration cell, namely S2049-Al fabricated in

November of 2002, with a total cell area of 0.408 cm2 was measured on 3/22/04 at NREL with









the following parameters for cell #2: Voc = 0.661 V, Jse = 33.73 mA/cm2, FF =77.47 %, and rl =

17.28 %. The distance between the light source and the calibration cell was adjusted to obtain a

calibrated position where the measured Jse matches the certified Jse value.

The temperature of the device tends to increase under elongated radiation from the solar

simulator. Water cooling system and a thermal couple is embedded in the test stage to control

the temperature. Efforts were made to maintain the temperature of the test cell at 25 + loC

during the characterization.

Devices with Absorbers from UF PMEE Reactor

Devices were first time completely fabricated in house. Two kinds of absorbers, namely

CIGS and CGS, were grown in the UF PMEE reactor. These absorbers were etched in 10%

potassium cyanide solution to remove any possible Cu rich phases before putting on the buffer

layer. All the buffer layers are CdZnS approximately 50nm thick. No intrinsic ZnO was

deposited so that the structure is simplified and fewer variables are involved. AZO window layer

of approximately 600nm was deposited to ensure a sheet resistance well below 5002/0, although

later optimization on EPV absorbers indicates that thinner AZO layer would be sufficient and

should have improved the performance of these devices.

CIGS Devices

For Cunlni-GaxSe, absorbers with different x value and different Cu to group III metal ratio

were grown and devices were fabricated using the method described earlier. The substrate

temperature during absorber growth was approximately 4900C. All metal fluxes were kept

constant during the growth. Therefore there was no bandgap grading for these absorbers. Table

6-3 shows the film properties of the CIGS absorbers used for device fabrication. These are Cu










poor films with x values vary around 0.3. Film thickness is around 1Clm. Only one thick sample

with thickness of 1.9 Clm was grown due to the low throughput of the system.

The I-V properties of these CIGS devices were measured using the I-V system. The solar

simulator intensity was calibrated using the reference cell before and after the lengthy

measurement to ensure there is no maj or parameter drift during the process. The calibration cell

performance parameter is the short circuit current. Table 6-4 shows the NREL certified

parameters of cell #2 on S2049-Al and the calibrated parameters of this cell before and after the

measurement. It can be seen that the calibrated results are conservative to avoid an exaggeration

of the device performance.

Table 6-5 lists the measured I-V character of the UF CIGS devices. Device 582, the CIGS

with zero Ga content (essentially CulnSe2), showed the highest efficiency of 8.9%. It had a

relatively high open circuit voltage and the highest short circuit current among all the CIGS

devices. The highest CIGS device, #588, has good short circuit current and fill factor, but the

lower open circuit voltage compared with device 582 made its efficiency lower.

The two best performance absorbers were evaluated under high resolution scanning

electron microscope. Figure 6-3 and Figure 6-4 are the cross-section SEM images of absorber

#582 and #588 respectively. It can be seen that these two films have similar thickness. Both of

these two absorbers have relatively large grain size, which is important for high performance.

CGS Devices

High efficiency CuGaSe2 devices are required as the top cell for high performance tandem

chalcopyrite structure. Currently the efficiency of CGS is lagging far behind that of CIGS [80].

Much still needs to be done to improve the efficiency of CGS devices.









Some devices using CGS grown from the PMEE reactor were fabricated. The CGS growth

parameters are shown in Table 6-6. Generally, two growth recipes were involved in producing

these CGS absorbers. One recipe is an emulated NREL 3-stage growth process trying to create

bandgap grading within the absorber layer. The other recipe uses constant metal flux to obtain a

uniform band structure in the CGS thin films. All of these films were Ga rich with copper to

gallium ratio varied slightly between 0.91 and 0.95. Table 6-7 shows the device performances of

these CGS films. The highest efficiency of4.9% was achieved on Device 640grown via the

emulated 3-stage process.

Cross-section SEM image was taken on the UF champion CIGS device and the result is

shown in Figure 6-5. Each layer constituting the device can be clearly seen except there is no

significant boundary between the CdS and ZnO layers. This device has a fairly thick absorber

with large grain size in the order of microns. Good attachment between the absorber and the

underneath Mo layer indicating a good stress control on the relatively thin Mo back contact.

Effect of Zinc Oxide Window Layer on Device Performance

CIGS devices are also fabricated using the absorbers provided by Energy Photovotaics Inc.

These absorbers are grown on large sheet of substrates then cut into smaller 2"x2" pieces and

shipped from EPV to UF. Therefore, it is reasonable to assume that the absorber layer has

similar performance and can be used for the window layer optimization. The effect of optical

transmission and sheet resistance of the AZO layer was investigated. Another issue addressed is

the effect of the intrinsic ZnO layer.

For experiments conducted to investigate the transmission and sheet resistance, only the

AZO thin film layer was deposited and the intrinsic ZnO layer was skipped. Doing so would










simplify the solar cell structure and the fabrication process, reduce the possibility of effects

brought in by an extra step and thus focus on the issue under investigation.

The impact of intrinsic ZnO layer on solar cell performance was then assessed. The

motivation is to evaluate the necessity of this intrinsic ZnO layer. The deposition rate for this

layer is generally low because it is sputtered from a fully oxidized target. Although only a very

thin layer of intrinsic ZnO is required, it generally consumes comparable time as that for

sputtering the AZO layer due to the much smaller deposition rate. If this layer can be proved to

be unnecessary, it would significantly reduce the complication of the fabrication process and

greatly improve the process throughput. In literature there are contradictory reports regarding to

the effect of this layer and further investigation needs to be conducted. Devices fabricated with

and without the intrinsic ZnO layer are compared and the results were analyzed.

Effect of Window Layer Transmission

The AZO transmission experiments were conducted using CIGS absorbers from EPV.

Efforts were made trying to eliminate the possible variations coming from other layers. These

include cutting the absorbers off from the same 2"x2" CIGS substrate and depositing the ZnCdS

buffer layer during the same batch of CBD process. A 2"x2" substrate was cut into four 1" xl"

pieces and three of them were then coated with buffer layer during one single batch of bath

deposition. The maximum number of substrates can be processed in one batch of CBD

deposition is three due to the number of substrate holders available. The recipe used was

targeted to a stoichiometry of Cdl-xZnxS with x equals 0.3. AZO thin films with different optical

transmission were then deposited on top of the buffer layer. A dummy glass was placed beside

each sample to monitor the optical transmission and sheet resistance. Finally, the metal top

contact fingers are evaporated to finish the device.










Figure 6-6 shows the window layer optical transmission of the three aforementioned CIGS

devices. For easy comparison, the transmission is categorized into three levels: low, medium

and high. The current-voltage (I-V) properties of these solar cells were characterized and

important device parameters are listed in Table 6-8. The results showed that by increasing the

window layer optical transmission, both the open circuit voltage (Voc) and short circuit current

(Jse) were increased. The Voc improved from 0.43 6V to 0.477V and Jse increased from

24. 12mA/cm2 to 27.50mA/cm2 when the transmission improved from low to high region. As a

result, the solar cell efficiency was improved from 4.59% for a low transmission window to

6.21% for a high transmission window. This is approximately 35% improvement of the

efficiency.

The results from AZO transmission experiments showed that the window layer optical

property is critical to solar cell performance. It is a very effective way to improve CIGS solar

cell performance by enhancing the optical transmission of the AZO window layer. It is also

known that improvement of transmission by applying MgF2 antireflection layer can further

increase the device performance.

Effect of Sheet Resistance

As mentioned before, the sheet resistance affects the series resistance of a solar cell and

according the theory showed in Chapter2, it would affect the performance of a device. To

evaluate this effect, a number of devices with various sheet resistances were fabricated on CIGS

provided by EPV and the device performance was evaluated via I-V measurement.

Figure 6-7 shows the relationship between AZO layer sheet resistance and CIGS solar cell

efficiency. Many data points were taken around 500-cm region, a value adopted by literature

report[81]. But exploration of sheet resistance to regions much higher than this value also










yielded devices with good efficiency. Generally, the device performance is very tolerant to sheet

resistance over a fairly large range. Reasonable performance was obtained from devices with

window layer as thin as 100nm.

A good tolerance to sheet resistance yet being sensitive to transmission indicate device

performance can be improved by reducing the window layer thickness to achieve better

transmission while maintaining a reasonable sheet resistance. Since many factors coming from

different layers of the device can affect the efficiency, it is difficult to perform an optimization

with limited amount of EPV absorbers available. Indeed, there is no straight forward trend

observed from these performance data.

Impact of the Intrinsic ZnO Layer

Three pairs of devices were fabricated using CIGS from EPV to evaluate the effect of the

intrinsic ZnO layer. Within each pair, one device is fabricated with the intrinsic ZnO layer and

the other without. These devices were characterized using I-V measurement system and the

result is shown in Table 6-9. It is evident that devices with the intrinsic ZnO layer unanimously

showed an improvement of Voc. The Jse value is also higher for the devices with the intrinsic

ZnO layer compared to its counterpart, with the first pair having very close Jse values though. As

the result, all devices with the intrinsic ZnO layer showed improved efficiency.

These data show that intrinsic ZnO is necessary in terms of achieving high efficiency, at

least for the case of EPV absorbers. But from the standpoint of industry manufacturing, this

improvement has to be balanced with the cost of an extra step of deposition. A careful

evaluation of the benefit and cost need to be made. This is out of the scope of this dissertation.

Summary

Devices were fabricated using UF CIGS and CGS absorbers. Reasonable performance was

achieved on these devices indicating device quality ZnO has been achieved. Further










optimization of the window layer to improve device performance was conducted using EPV

absorbers. Results showed that the devices are tolerant to the sheet resistance over a large range.

Optical transmission plays a more direct role affecting the efficiency. Device efficiency can be

increased by using thinner ZnO as the window layer. The necessity of the intrinsic ZnO layer

was investigated. The results showed that this layer is beneficial to improve the device

efficiency. It is necessary to have this layer to achieve best efficiency devices.










Table 6-1. Recipe for CdS buffer layer deposition.


Constitute
DI water
Thiourea
Ammonia hydroxide
CdSO4


Concentration
N/A
1.5mol/L
28.0-30.0wt%
0.015mol/L


Volume
365mL
25mL
62.5mL
50mL


Table 6-2. Recipe for CdS buffer layer deposition.
Constitute Concentration Volume
DI water N/A 490mL
Thiourea 0.0834mol/L 100mL
Ammonia hydroxide 3.5-3.75wt%/ 0.25mL
ZnCl2 0.254mol/L 1.4mL
CdCl2 2.5H20 0.0084mol/L 121.9mL


Table 6-3.
Film ID
569
575
578
579
582
586
587
588


Composition and thickness data for UF CIGS absorbers.
Cu/III ratio Ga/(In+Ga) ratio Thickness (Cpm)
0.97 0.30 0.8
0.97 0.33 0.85
0.97 0.32 1.3
1.00 0.30 1.9
0.91 0 1.0
0.90 0.25 0.9
0.99 0.41 0.85
0.98 0.21 0.9


Table 6-4. Calibration of the reference cell.
Data Source Voc (V) Jsc (mA/cm2)
NREL certified value 0.661 33.73
Calibration result 0.642 33.62
before measurement
Calibration result 0.643 33.56
after measurement


FF (%)
77.47
76.09

76.24


Eff. (%)
17.28
16.42

16.24










Table 6-5. I-V character of the UF CIGS devices.
Device ID Voc (V) Jsc (mA/cm2) FF (%) Eff. (%)
569 0.457 24.3 39.4 4.4
575 0.476 26.3 55.5 7.0
578 0.325 26.6 34.2 3.0
582 0.457 30.5 64.0 8.9
586 0.317 21.3 26.8 1.8
587 0.308 18.1 41.7 2.3
588 0.437 29.3 67.0 8.6
Note: Device 579 gave no I-V characteristics

Table 6-6. Growth recipe and composition for UF CGS absorbers.
Film ID Process Cu/Ga ratio
640 3-stage 0.91
647 Constant flux 0.92
649 3-stage 0.95
655 Constant flux 0.95


Table 6-7. I-V
Device ID
640
647
649
655


character of the UF CGS devices.
Voc (Vi) Jsc (mA/cm2) FF (%)
0.502 16.9 57.4
0.480 13.7 55.7
0.426 16.8 46.3
0.516 14.2 57.8


Efficiency (%)
4.9
3.7
3.3
4.2


Table 6-8. Performance of CIGS solar cells with different window layer transmission.
Device ID Transmission Voc (V) Jse (mA/cm2) FF(%) Efficiency (%)
EPV5-2 Low 0.436 24.12 43.65 4.59
EPV5-1 Medium 0.467 25.79 47.89 5.77
EPV5-3 High 0.477 27.50 47.35 6.21


Table 6-9. Performance of CIGS
Pair number Device ID
~EPV3-1
Pair 1
EPV2-2


solar cells with and without i-ZnO layer
i-ZnO Voc (V) Jse (mA/cm2)
Yes 0.536 29.64
No 0.488 29.58


FF(%)
47.18
47.00
45.51
47.89
50.63
47.35


Efficiency (%)
7.49
6.78
6.02
5.77
7.25
6.21


EPV5 -4
EPV5-1
EPV 6-1
EPV 5-3


Yes
No
Yes
No


0.494
0.467
0.498
0.477


26.78
25.79
28.75
27.50


Pair 2


Pair 3





























Load-lock
Zone

Figure 6-1. Schematic of the PMEE reactor.


Figure 6-2. Photo of CIGS solar cells fabricated on a 2"xl" soda-lime glass substrate.


































rl I I II I I I II


Luunrr EH I = b.UU kV Date :19 Oct 2007
Mag 3000 X ~WD = 8 mm File Name = ClGS588_02.tif


200nm
Mag = 30.00 KX X


EHT = 5.00 kV Date :19 Oct 2007
WD = 5 mm File Name = ClGS582 02.tif


Figure 6-3.


Cross section SEM image of absorber #582.


~PE"


*~e. 1
ipuo;


r


Figure 6-4. Cross-section SEM image of absorber #588.


t 1:,Li4 5~.~B
:d
in
"r~ s
.


hj
;rl













y~n~c!
*ilFII*i?
~i~F~;i
C I *ir*r
"il~r*7 ~r~i
Yi ..~ r

*L C
*lrF.~~ J **


~IIWIF ;i*5`-;~a~-;- :a


-~i~li~ *I


"C~


u~E"


~; "


'+r
r


11-1111
Mag = 10.00 KX X


EH I = b.UU KV
WD = 4 mm


Date :19 Oct 2007
File Name = CGS640 02.tif


Figure 6-5. Cross-section SEM image of CGS device #640.









































108


I ~ ~..~ -'* ~. ~~
`~~3
~R~iWm~~Y ~
.? r.
~ L~'. u~'BlkL1~I ik rlr -

+i
Y
.j : ; Z II: k
; F
*- ~I', ~
~E* d

f ,,
6:Y
r ;iu;r'TJ;;l~j II
n





0 50 100 150 200 250 300 350

Sheet resistance (R cm)



Figure 6-7. CIGS cell efficiency versus the AZO layer sheet resistance.


300


400 500 600 700

Wavelength


Figure 6-6. AZO window layer optical transmission of three CIGS solar cells.


* **
** *









CHAPTER 7
INVERSE MODELINTG OF CIGS SOLAR CELLS

CIGS solar cell structure consists of several layers of compound semiconductors.

Although today's material characterization techniques are highly advanced, there is still much

information remains unknown. For example, the defect distribution within the CIGS layer has a

fundamental impact on the behavior of the absorber layer and it is critical to determine the cell

conversion efficiency. The work shown in this section is dedicated to develop a convenient way

of identifying some of the key unknown factors through numerical simulation. The idea is to

construct an inverse problem and solve the unknown parameters through inverse modeling.

Details of the procedure are addressed in the following sections.

Definition of Inverse Modeling

This section proposes a definition of inverse modeling, the process of solving an inverse

problem. Let a physical system under study be denoted by S. The scientific analysis of system S

can be divided into the following three aspects:

(i) Paranteterization of the system: specification of a minimal set of model parameters
whose values completely characterize the system.

(ii) Forwardnrodeling: use of the physical laws to make predictions on some observable
parameters (output values) basing on some input values.

(iii) Inverse modeling: use of the actual results of some measurements of the observable
parameters (output values) to infer the actual values of the model parameters.

Strong feedback exists among these three steps and a dramatic advance in one of them is

usually followed by the other two [20].

A schematic description of the three components discussed above is given in Figure 7-1,

which shows a system S with known parameters. The process of feeding through input values to

derive output values is called forward modeling. The process of finding the input values that

yield given output values is called inverse modeling. A special case of inverse modeling is










system identification, where both input values and output values are known, and the obj ective is

to determine the parameters of system S.

Inverse Modeling for Photovoltaic Cells

Applying the three steps defined previously to the case of solar cell, the solar cell system

can be analyzed under the following perspectives:

(i) Parameterization of the system

Examples of input and output parameters of a solar cell system are shown in Figure 7-2.

The input parameters include cell configuration parameters and material properties parameters.

The output parameters may include the open circuit voltage, short circuit current, fill factor, and

energy conversion efficiency.

(ii) Forward Modeling

The physical laws followed in the forward process are the various physical models used

in the device simulation software, such as Medici [21]. Basically, the input parameters are

provided to Medici by the user, and the software does the calculation to predict the values of the

output parameters of interest.

(iii) Inverse M~odeling

Based on specified values of cell performance parameters (output values), such as cell

efficiency, the inverse modeling method seeks to find the values of a set of input parameters that

yield the given output values. It should be noted that for the solar cell case, it is believed that for

certain efficiency, there may be more than one set of input parameters which yield the same

efficiency value. In other words, different combinations of the input parameters may result in

the same output parameter value.










Application of Inverse Modeling Performance Optimization

Numerical simulation has been adapted to optimize the performance of CIGS solar cells

[22], but a comprehensive study is far from complete. We hope to obtain better understanding

and optimization by using inverse modeling.

An extension of inverse modeling is the endeavor to optimize key performance parameters,

such as cell efficiency, to Eind a set of input parameters that generate the highest possible cell

efficiency. This is extremely attractive because improving the cell efficiency has always been the

most challenging task for the Photovoltaic community.

Approach

The inverse problem for solar cell proposed in this work consists of integrating Matlab

[23] and Medici. Matlab is responsible for the parameter analysis and for providing the

optimization algorithm. Medici is used to carry out the forward modeling part, and to predict the

values of the output parameters.

The flow chart shown in Figure 7-3 describes the procedure for solving the inverse

problem. The inverse modeling algorithm is comprised of all the steps between the Start and

Stop indicators in the flow chart. The user must first specify the target output parameters. Then

the user specifies an initial guess for the input values. At this point, an iteration procedure is

initiated by Matlab, by feeding the input values to the Medici program. In the next step, Medici

predicts the corresponding output value for the Matlab-specified parameters.

At the beginning, a set of initial attempted parameters such as defect energy distribution

and concentration were input to Matlab. Then Matlab send this set of parameters to Medici.

Medici will proceed to do the forward modeling and generate the criteria parameters such as

open circuit voltage (Voc), short circuit current (Isc) and fill factor (FF). Then values of these

criteria parameters were sent back to Matlab, where they were compared with the set value. If









the difference between the generated values and the set values is smaller than a tolerated value

(usually a very small value), the inverse process stops and the set of input parameters that will

yield the wanted criteria parameters are found. Otherwise, when there is still a significant

difference between the generated values and the set values, Matlab will generate a new set of

input parameters according the Algorithm used, and send them back to Medici and the process is

repeated until the generated values fit the set values.

The process can be easily adapted to optimize arbitrary output values, such as Voc, 1sc, FF

and cell efficiency. The process stops once the optimized parameter reaches its optimum value.

For the efficiency case, the process proceeds until the maximum cell efficiency is found.

Forward Modeling

This section describes details of utilizing Medici for the forward modeling. Cell structure

is first built in Medici by defining each layer of the device. Then relevant input values for each

layer is assigned to describe the properties of these layers. Some examples of the input values

are listed.

Cell structure generation

The cell structure used in the forward modeling is shown in Figure 7-4. At the bottom is a

layer of Molybdenum on top of a glass substrate serving as the ohmic contact for the p type

Copper Indium Gallium di-Selenide (CulnxGal-xSe2 Of Simply CIGS). Next is the absorber layer

p type CIGS. The typical thickness for the CIGS layer is about 2 microns so as to completely

absorb the incoming sun light. The OVC (ordered vacancy compound) layer is not intentionally

deposited. It is an indium rich n-type layer formed between the p-type CIGS and the n-type CdS.

The existence of this layer was initially suggested by Schimid et al. [24]. The CdS buffer layer

is deposited through Chemical Bath Deposition (CBD). On top of CdS, a thin layer of intrinsic

ZnO (i-ZnO) is deposited typically by sputtering. Finally, the top of the structure consists of a










layer of Transparent Conductive Oxide (TCO) in the form of heavily Al-doped ZnO, which

forms ohmic contact with an Aluminum grid (not shown in the Eigure) that completes the device.

The software Medici uses a mesh concept to describe the property distribution inside the

simulated device. It does calculation at each mesh node and then combines these results to form

a whole structure view of the complete device. The CIGS cell structure of Figure 7-4 should be

first described using mesh concept in order for Medici to proceed with the calculation. As

shown in Figure 7-5, the graph on the left shows the mesh generated in Medici and the graph on

the right shows each layer. Starting from the bottom to the top, the successive layers are CIGS,

OVC, CdS, i-ZnO and Al doped ZnO.

Input parameters

The simulation temperature is kept at 300K in all cases considered, and the input solar

spectrum was introduced from the "AMO.dat" file provided by Medici. This spectrum generates

an input energy density of approximately 100 mW/cm2

Some of the important parameters used in the simulation are shown in Table 7-1. The

values of these parameters are consistent with the values used in the simulation based on the

widely used solar cell simulation software AMPS1D [25].

With all the materials parameters defined in Medici, the bandgap profile can be generated.

An example is shown in Figure 7-6. The layers can be seen in the bandgap profile from left to

right are Al-ZnO, i-ZnO, CdS, OVC, and CIGS. Graded CIGS bandgap profile can also be

achieved by defining more than one layer of CIGS with different bandgap values.

Example of a forward modeling

Figure 7-7 shows a typical I-V curve generated by Medici. The input parameters were

taken from Table 7-1. Several important device performance parameters can be extracted from









the I-V curve. They are short circuit current (Isc), open circuit voltage (Voc), Eill factor (FF) and

efficiency (r).

The short circuit current Isc is defined as the current value at the spot where the I-V curve

crosses the vertical axis [7]. From the graph it can be seen that the Isc value is approximately 37

mA/cm2. The open circuit voltage Voc is the voltage value at the spot where the I-V curve

crosses the horizontal axis. Hence, the Voc value is approximately 0.61 V for the case shown in

Figure 7-7.

Efficiency ris defined as

Pm,
9 = (7.1)
P~n

where Pmax is the maximum output power of the device and P,, is the input power of the

solar light. Pmax is basically the maximum value of the product of I and V along the I-V curve.

P,, is the total input power of the solar light. Its value is approximately 100 mW/cm2 for the

input spectrum used in this simulation study. As the maximum power for the case shown in

Figure 7-7 is approximately 17.5 mW/cm2, the value of r approximately equals to 17.5% for this

case.

The fill factor (FF) is defined using the expression

P,,
FF =(7.2)


where Isc and Voc are the short circuit current and open circuit voltage respectively, which

can be identified from the I-V curve. Note Pmax can be geometrically taken as the largest

rectangular within the I-V curve and the product of Isc and Voc is basically the rectangular whose

corners are defined by the Isc and Voc spot together with the origin. FF can be geometrically









taken as the area ratio of these two rectangular. It has been discussed in early chapters that FF is

sensitive with the series resistance of the device.

Note Isc, Voc, fill factor and efficiency are not four independent parameters. The efficiency

can be calculated using Isc, Voc, and fill factor. Therefore, in the inverse process discussed later,

the three output parameters used are only Ise, Voc, and fill factor.

Modeling of NREL champion device

The champion CIGS device in 2003 from NREL was modeled using Medici. This device

has an efficiency of 19.2% with Isc = 35.71 mA/cm2, Voc = 0.698 V and FF = 78. 12%. Figure

7-8 shows the simulated I-V curve plotted together with the experimental data obtained by

NREL. A reasonable agreement between simulation and experiment was achieved. This

indicates the Medici forward modeling program may be used in the inverse process to

automatically determine the appropriate values of some input parameters.

Inverse Modeling

Using the procedure described earlier, Matlab and Medici can be integrated to fulfill the

inverse modeling process and identify important unknowns in the CIGS absorber layer. Matlab

optimization toolbox was utilized for this purpose. The criteria "y" for optimization is shown in

the following equation:

I V,- VFF FF,
y = ax( scsO )2 + b x( cocO )2 0CX 2 (7.3)
scO ocO FF

where Isco, Voco and FFo are the targeted device performance value. Parameters a, b, and c

are weighing factors to adjust the convergence preference among the three parameters. From the

equation it is clear that during the minimization process of criteria y, the device performance

parameters Ise, Voc, and fill factor approach their targeted values.









Three input variables were adjusted during the inverse process. They are three defect

concentrations in the CIGS absorber layer. The interested defects in the CIGS layer are shallow

level acceptor state, deep level electron trap and deep level hole trap. Their energy levels are

assigned with reasonable values within the bandgap which are shown in Table 7-1. The inverse

process is supposed to identify a set of input parameters that would yield certain output

parameters. The search range for the three defects is between 2x1017/cm3 and 1x1020/CM3. Note

that due to the complication of the forward model (the Medici simulation process), there could be

more than one set of input parameters that would yield certain output parameters. In other word,

the solution may not be unique.

Inverse Modeling of NREL champion device

The previous mentioned NREL champion device was used as the target for this inverse

modeling process. This means the target performance was set so that Ico = 35.71 mA/cm2, Voco =

0.689 V and FFo = 78. 12%. Inverse modeling seeks a set of input values that yields the same

performance values of the champion cell.

For the first run case, the starting values of the three defect concentrations are randomly

chosen as 1x1019, 5x1017 and 1x1019/CM3, TOSpectively. This yields a performance parameter set

of Voc = 0.500V, Ac, = 35.81 mA/cm2, and FF = 72.67%. The inverse process is then executed

following the procedures described earlier in Figure 7-3. As shown in Figure 7-9, after only

approximately 20 iterations, the three performance parameters are converged to values close to

the target performance parameters. A set of defect concentrations with values 5.3 x 1019, 6.6 x1019,

and 2.6 x10 x is identified to give a performance parameter set of Voc = 0.689 V, Ac, = 35.7

mA/cm2, and FF = 78.4%.

The first run case was successful and very efficient. This can be contributed to a good

initial guess which yields Ac, and FF close to their target value. For the second attempt, the initial









guess defect concentrations were taken as 5x1017, 3x101s and 8x1017/cm3. This reflects a

performance parameter set of Voc = 0.357V, Ac, = 40.92 mA/cm2, and FF = 53.96%. Note none

of these parameters are close to their final target value. The iteration process is shown in Figure

7-10. Compared with the first case, the second inverse process takes much more iterations to

converge and stabilize at the target values. The variation of each of the three parameters also

covers a much larger range. After approximately 80 iterations, a set of defect concentrations

with values 5.2x1019, 1.0x1020, and 2.6x101s/cm3 is identified to give a performance parameter

set of Voc = 0.688 V, Ac, = 35.7 mA/cm2, and FF = 78.4%. Note that for the second inverse

process, two defect concentrations converged to the same values that were obtained in the first

run. Yet the electron trap concentration converged to another value which had the same order of

magnitude. This suggests that the three output parameters are not very sensitive with the

electron trap concentration.

For the third demonstration case, the defect concentrations were chosen at the upper

boundary, which corresponds to values of 1.0x1020, 1.0x 1020, and 1.0x1020/CM3. This makes the

initial guess as far away as possible from the final converging point. With this initial guess, the

performance parameters are: Voc = 0.500V, Ac, = 35.38 mA/cm2, and FF = 69.58%. As seen in

the iteration process shown in Figure 7-11, after approximately 40 iterations, the performance

parameters converged and stabilized at their targeted values. This yields a defect concentration

combination of 5.3 x 1019, 1.0 x1020, and 2.6 x101s/cm3, the same point where the second case

converged .

In conclusion, the inverse process is feasible to identify some input parameters in the CIGS

layer. It has been demonstrated through different cases, no matter what the initial guess is, the

sensitive parameters all converge to the same point. Yet the insensitive parameters can be









randomly located. This process can provide an alternative way to determine or at least shine

light on some of the sensitive parameters that can not be easily measured through experiments.

But numerical simulation can never completely replace experimental work. Theoretical

predictions still need to be verified through experimental methods.














Forward modeling


Inverse modeling


Figure 7-1. Schematic diagram illustrating the forward modeling and inverse modeling
operations for a given physical system S














Voc
Ise
FF
Efficiency


Output values


Cell structure
Film thickness
Anti reflection coating, etc.


Semiconductor bandgap
Electron/hole concentration
Electron affinity
Electron/hole mobility
Density of states in both bands
etc.


Input values


Material properties r


Figure 7-2. Parameter classification with examples for solar devices
































Matlab feeds input
parameters to Medici


Matlab generates another
set of input parameters


Medici predicts
performance parameter
values (Forward
Modeling)


iteration iteration+1


Figure 7-3. Procedure chart for solar cell inverse modeling.













TC'O (ZnO0)






CdS


OVC






CIGS


vld~~dllll[l~*;r~~l:I[*: ,,


m
hi





hi
hi


ClGS Cell Structure


50nm, E,=3.3eV


30nm, E,=2.4eV


30nm, E,=1.3eV






2p~m, E,=1.0-1.7eV


Figure 7-4. A typical CIGS cell structure.


2.00


0.00


1.00
Distance (Microns)


2.00


0.00


1.00
Distance (Microns)


Figure 7-5. Cell structure described in Medici.


CTGS Cell MESH






























I-V curve of CIGS solar cell


1-llelluY


I


I


h
c'
o

n
t'
C
ol~
o



B


0.40


0.60 0.80
Distance (Microns)


1.20


Figure 7-6.


Bandgap profile of a CIGS device generated by Medici.


1.000 0.1;00


0.200 0.300 0.400 0.500 0.600 0 .700
V(Volt)


Figure 7-7. A typical I-V curve generated by Medici.



124


















30







-NREL Champion Cell
S- Medici Simulation
10






0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

V (Volt)


Figure 7-8. Medici simulation result for an NREL champion device.






















35.0



S30.0



:25.0



20.0



15.0


0.80






0.60



-0.50


0.40


0.30



0.20


0.8!

0.8(

0.7!

0.7(

0.6!


0.6(

0.5!

o E(


1 21 41

Iteration number



Figure 7-9. Iteration of inverse modeling process for there solar cell performance variables: (a)

Isc, (b) Voc, and (c) FF. (Case 1)


-



-



-


(a)


5




5

0


5

0

5

O













15.0


40.0





S30.0


25.0


20.0


15.0



0.80


0.70 .. ..


0.60


0.50


0.40


0.30


0.20


0.85





0.70

0.65

0.60

0.55

0.50

0.45

0.10
1 21 41 61 81 101

Iteration number



Figure 7-10. Iteration of inverse modeling process for there solar cell performance variables: (a)
1sc, (b) Voc, and (c) FF. (Case 2)



















^30.0



25.0





20.0



15.0


0.80


0.70


0.60


0.50


0.40


0.30


0.20


0.85


0.80

0.75


0.70


0.65


0.60

0.55


0.50
1 21 41 61 81

Iteration number



Figure 7-11. Iteration of inverse modeling process for there solar cell performance variables: (a)
1sc, (b) Voc, and (c) FF. (Case 3)









Table 7-1. Typical input parameters for Medici.
Parameters i-ZnO CdS


OVC


CIGS


Thickness (nm)
Electron affinity X
(e V)
Eg (eV)
Electron mobility
(cm /V-s)
Hole mobility
(cm2/V-s)
Density of States
in conduction
band N, (cm-3
Density of States
in valence band
Nv (cm-3
Doping
Concentration
(cm-3>
Acceptor State
Concentration
NAS CA-13)
Electron Trap
Concentration NET
(cm-3>
Hole Trap
Concentration
NHT CA-13)
Acceptor State
Energy level (eV)
Electron trap
Energy level (eV)
Hole Trap
Energy level (eV)


30

3.95

1.3

10


2000

3.8

1.12-1.4

300

30


3x101s


1.5x1019


N/A


5 x10s


1 x1019


2.8 x10 x


Ev+0. 5E,-0.44

Ev+0. 5E,+0.2

Ev+0. 5E,-0.2


3.75


1x1019


1x1019


5 x10


N/A


N/A


N/A


N/A


1x1019


1x1019


6 x1016


N/A


N/A


N/A


N/A

N/A


6.6x101


1.5x1019


1 x1012


N/A


N/A


N/A


N/A

N/A

N/A


N/A

N/A


N/A









CHAPTER 8
CONCLUSION

Parameter Effects on Sputtered AZO

The sputtering system setup and deposition parameters greatly affect the optical and

electrical properties of deposited AZO thin films. This is the reason for the significant disparities

of best resistivity values reported in literature, which can be as high as several orders of

magnitude.

The magnetron is a critical part of the experimental set up. The magnetic Hield created by

the magnetron confines electrons near the surface of the target and consequently keeps the

plasma close to the target surface instead of evenly distributed between the target and substrate.

This reduces the damage done to AZO thin films through bombardment and improves the

electrical properties of sputtered AZO thin fi1m.

Zinc is very reactive with oxygen and if oxygen is included in the sputtering gas, its

amount has to be small and accurately controlled. Otherwise the resistivity would be

significantly increased. Generally for the Perkin Elmer sputtering system used, no oxygen is

used during normal sputtering processes.

Deposition parameters can also affect the optical and electrical properties of AZO.

Examples of such parameters are deposition power, working pressure and starting base pressure.

Generally, increasing the deposition power would increase the sputtering rate, but it tend to

reduce the optical transmission if no oxygen is present in the sputtering gas. This effect is done

through the modification of the zinc and oxygen ratio at the target surface. When the deposition

power is too high, the surface of the target can be reduced to a zinc rich mode if no oxygen is

present in the sputtering gas. Depending on the degree of this reduced condition, optical

transmission of AZO films deposited can drop to as low as zero. If the target is in such reduced










mode, a regeneration using argon and oxygen mixture gas is required. To prevent the target from

reducing, a deposition power below 400W is suggested. Working pressure can also affect the

sputtering rate and resistivity of AZO thin films and an operation under the optimized working

pressure is recommended.

Experimental results showed a significant decrease of the resistivity with introducing

hydrogen into the sputtering gas. It is also found that AZO resistivity is affected by the starting

base pressure, which is related with the amount of residue water vapor, a source of hydrogen.

These results support the theoretical predictions in literature that hydrogen acts as a doping

source for zinc oxide thin films.

Device Fabrication

Window layer of thin film zinc oxide is critical for CIGS solar cells. It affects the solar

cell performance through two mechanisms.

First, the thickness of zinc oxide affects the amount of photons that reaches the CIGS

absorber layer. This is directly related with the optical transmission of the zinc oxide layer.

When there is no antireflection layer present, the thickness of zinc oxide window layer can lead

to different optical interference, which could be constructive interference or destructive ones.

When constructive interference happens, a significant amount of light is direct back into the air

without being able to enter the device. This could dramatically reduce the solar cell efficiency.

Second, the electrical property of zinc oxide layer is also very important. The sheet

resistance of zinc oxide affects the series resistance of the device. It is known that the fill factor

of solar cell is very sensitive with the series resistance. A slight increase of the series resistance

could lead to a significant efficiency loss. Therefore, depositing zinc oxide with sufficient low

sheet resistance is required for a good solar cell performance.









Defects in CIGS Absorbers

CIGS absorbers can be deposited using various methods including both vacuum and

non-vacuum techniques. Performance of these CIGS absorbers can vary significantly. It has

been a great interest to discover a theory to explain the disparities among the performance of

these CIGS absorbers. Defect distribution is believed to be an important factor.

CIGS is not intentionally doped. The p-type conductivity comes from the various defects

existing in the semiconductor. An increase of the shallow acceptor defect concentrations would

increase the carrier concentration and simulation results showed a sharp increase of device

performance .

An inverse modeling process was constructed and it can be utilized to identify the defect

distributions corresponding to different solar cell performance. This could be used as a

theoretical approach of determining defect properties to verify results experimentally obtained

using DLTS.

Future Work

The experimental work on hydrogen doping of AZO thin films can be further investigated

by using different concentrations of hydrogen for the sputtering gas. It is possible that increasing

the hydrogen concentration would lead to even lower resistivity of the AZO thin fi1m. An

optimum concentration of hydrogen should be determined. It is of benefit to reduce the

resistivity so that the same sheet resistance could be achieved with less thickness. This would

reduce the material cost and also the processing time. Also thinner AZO window layer could

reduce the photon absorption and optical interference if appropriate thickness is chosen.

For good uniformity on large size substrate, the AZO thickness variation should be

carefully studied. A sputtering process that would yield good uniformity for large-scale industry









PV applications is desired. This could be achieved by choosing the right shape of the target and

better arrangement of the magnetic field distribution.

The defect study on CIGS using inverse modeling should be accompanied by DLTS

measurements. A combination of experimental and numerical tools could be of help to both

individual techniques.









LIST OF REFERENCES

1. K. Zweibel, in H~arnessing Solar Power: The Photovoltaics Challenge, (Plenum Press,
New York, 1990).

2. L.L. Kazmerski, International Material Review 34, 185 (1989).

3. L. Reith. Sputter deposition ofZnO thin films, Ph.D. Dissertation, University of Florida.
2001.

4. L. Kerr. Gi r on thr and thernzodynantic modeling of absorber and transparent conductive
oxide for copper indium diselenide material based solar cells, Ph.D. Dissertation,
University of Florida. 2004.

5. J. Mazer, in Solar Cells: An hItroduction to Crystalline Photvoltaic Technology, (Kluwer
Academic Publishers, Boston, 1997).

6. P. Wurfel, in Physics ofSolar Cells, f~onsprinciples to new concepts, (Wiley-Vch, 2005).

7. D. M. Chapin, C. S. Fuller, and G. L. Pearson, Journal of Applied Physics 25, 676 (1954).

8. J. L. Shay, S. Wagner, and H. M. Kasper, Applied Physics Letters 27, 89 (1975).

9. L. L. Kazmerski, F. R. White, and G. K. Morgan, Applied Physics Letters 29, 268 (1976).

10. W. Chen et al., Conference Record of the Twenty Third IEEE Photovoltaic Specialists
Conference,422 (1993).

11. K. Ramanathan et al., Progress in Photovoltaics 11, 225 (2003).

12. S. Li, in Semiconductor Physical Electronics, (Plenum Press, New York, 1993).

13. M. B. Prince, Journal of Applied Physics 26, 534 (1955).

14. E. Bucher, Applied Physics 17, 1 (1978).

15. T. Koyanagi, Conference Proceeding 12th IEEE Photovoltaic Specialist Conference, 627
(1976).

16. A. Goetzberger and V. U. Hoffmann, in Photovoltaic Solar Energy Generation, (Springer,
2005).

17. A. W. Blakers et al., Applied Physics Letters 55, 1363 (1989).

18. J. Dietl, D. Helmreich, and E. Sirtl, in Crystals: Giron thrl, Properties and Applications,
(Springer, 1981).

19. P. Stradins, Solar Energy Materials and Solar Cells 78, 349 (2003).









20. D. L. Staebler and C. R. Wronski, Applied Physics Letters 31, 292 (1977).

21. J. Muller, B. Rech, J. Springer, and M. Vanecek, Solar Energy 77, 917 (2004).

22. S. Bose and A. K. Barua, Journal of Physics D-Applied Physics 32, 213 (1999).

23. O. Kluth et al., Thin Solid Films 351, 247 (1999).

24. M. Yamaguchi, Journal of Applied Physics 78, 1476 (1995).

25. J. Song et at., NCPV and Solar Program Review Meeting 874 (2003).

26. R. G. Gordon, MRS Bulletin 25, 52 (2000).

27. T. Minami, MRS Bulletin 25, 38 (2000).

28. P. S. Nayar and A. Catalano, Applied Physics Letters 39, 105 (1981).

29. X. Jiang, F. L. Wong, M. K. Fung, and S. T. Lee, Applied Physics Letters 83, 1875 (2003).

30. D. H. Xu et al., Physics Letters A 346, 148 (2005).

31. J. Q. Zhao et at., Synthetic Metals 114, 251 (2000).

32. I. D. Kim et at., Applied Physics Letters 89, 022905(2006).

33. P. F. Carcia, R. S. Mclean, M. H. Reilly, and G. Nunes, Applied Physics Letters 82, 1117
(2003).

34. H. C. Cheng, C. F. Chen, and C. Y. Tsay, Applied Physics Letters 90, 012113 (2007).

35. J. Aranovich, A. Ortiz, and R. H. Bube, Journal of Vacuum Science & Technology 16, 994
(1979).

36. M. S. Tomar and F. J. Garcia, Thin Solid Films 90, 419 (1982).

37. M. G. Ambia, M. N. Islam, and M. O. Hakim, Journal of Materials Science 29, 6575
(1994).

38. W. S. Lau and S. J. Fonash, Journal of Electronic Materials 16, 141 (1987).

39. D. C. Agarwal et at., Journal of Applied Physics 99, 123 105 (2006).

40. O. A. Fouad, A. A. Ismail, Z. I. Zaki, and R. M. Mohamed, Applied Catalysis B-
Environmental 62, 144 (2006).

41. J. H. Hu and R. G. Gordon, Solar Cells 30, 437 (1991).

42. S. K. Tiku, C. K. Lau, and K. M. Lakin, Applied Physics Letters 36, 318 (1980).










43. J. H. Hu and R. G. Gordon, Journal of Applied Physics 71, 880 (1992).

44. T. Hada, K. Wasa, and S. Hayakawa, Thin Solid Films 7, 135 (1971).

45. J. B. Webb, D. F. Williams, and M. Buchanan, Applied Physics Letters 39, 640 (1981).

46. T. Minami, H. Nanto, and S. Takata, Applied Physics Letters 41, 958 (1982).

47. D. M. Mattox, in H~andbook of Physical Vapor Deposition (PVD) Processing, (Noyes
Publications, Westwood, New Jersey, 1998).

48. K. Ellmer, Journal of Physics D-Applied Physics 33, R17 (2000).

49. K. C. Park, D. Y. Ma, and K. H. Kim, Thin Solid Films 305, 201 (1997).

50. C. R. Aita, R. J. Lad, and T. C. Tisone, Journal of Applied Physics 51, 6405 (1980).

51. S. H. Jeong and J. H. Boo, Thin Solid Films 447, 105 (2004).

52. A. Gupta and A. D. Compaan, Applied Physics Letters 85, 684 (2004).

53. V. Sittinger et al., Thin Solid Films 496, 16 (2006).

54. M. A. Contreras et al., Progress in Photovoltaics 7, 311 (1999).

55. C. Agashe et al., Journal of Applied Physics 95, 1911 (2004).

56. K. K. Kim et al., Applied Physics Letters 83, 63 (2003).

57. Lake Shore 7500/9500 Series Hall System User's Manual (2000).

58. G. B. Hoflund, in H~andbook of Surface and hiterface Analysis, Edited by J. C. Riviere and
S. Myhra (Marcel Dekker,Inc., New York, 1998), Chap. 4.

59. R. Jenkins and R. L. Snyder, in hItroduction to X-ray Powder Diffa~ctometry, (Wiley-
Interscience, 1996).

60. H. P. Klug and L. E. Alexander, in X-Ray Diffiaction Procedures, 2 ed., (Wiley, New
York, 1974).

61. J. I. Pankove and N. M. Johnson, Semiconductors and Semimetals 34, 1 (1991).

62. C. G. Van de Walle, Physical Review Letters 85, 1012 (2000).

63. D. W. Hamby et al., Nuclear Instruments & Methods in Physics Research Section B-Beam
Interactions with Materials and Atoms 249, 196 (2006).

64. A. Rockett et al., Thin Solid Films 372, 212 (2000).










65. A. F. da Cunha, F. Kurdzesau, and P. M. P. Salome, Advanced Materials Forum lii, Pts 1
and 2 514-516, 93 (2006).

66. J. R. Tuttle et al., Progress in Photovoltaics 3, 383 (1995).

67. M. A. Contreras, M. J. Romero, and R. Noufi, Thin Solid Films 511, 51 (2006).

68. R. Caballero, C. Guillen, M. T. Gutierrez, and C. A. Kaufmann, Progress in Photovoltaics
14, 145 (2006).

69. C. M. Xu et at., Chinese Physics Letters 23, 2259 (2006).

70. W. Li, Y. Sun, W. Liu, and L. Zhou, Solar Energy 80, 191 (2006).

71. T. Wada et al., Physica Status Solidi A-Applications and Materials Science 203, 2593
(2006).

72. C. Eberspacher, C. Fredric, K. Pauls, and J. Serra, Thin Solid Films 387, 18 (2001).

73. V. K. Kapur, A. Bansal, P. Le, and O. I. Asensio, Thin Solid Films 431, 53 (2003).

74. B. J. Stanbery, Heteroepitaxy and nucleation control for the growth of metal
chalcogenides using activated reactant sources, Ph.D. Dissertation, University of Florida,
2001.

75. Kincal S., Modeling and control of multiple thermal effusion sources and substrate
temperature in molecular beam epitaxy reactors, Ph.D. Dissertation, University of Florida,
2002.

76. J. Song, Development, characterization and modeling of CGS/CIGS thin film tandem solar
cells, Ph.D. Dissertation, University of Florida, 2006.

77. W. Chen, Cadmium Zinc Sulfide by Solution Growth, US patent 5, 112,410 (1992).

78. K. Ramanathan, J. Keane, and R. Noufi, the 31st IEEE Photovoltaics Specialists
Conference and Exhibition 195 (2005).

79. S. Ishizuka et at., Solar Energy Materials and Solar Cells 87, 541 (2005).

80. D. L. Young et al., Progress in Photovoltaics 11, 535 (2003).

81. N. Dhere et at., NCPV and Solar Program Review Meeting, 853 (2003).
















BIOGRAPHICAL SKETCH

Wei Liu was born in Hebei, China in April, 1978 and attended elementary school there.

He moved with his family to Shandong, China in 1992 and spent his high school time in Yantai,

Shandong. He attended Tsinghua University in 1996 and graduated with his BS of Chemical

Engineering degree in year 2000. He then was recommended into the master program in the

same department and received his MS of Chemical Engineering degree in 2002. During his

study in the master program, he did part of his research proj ects in the Chemical Engineering

department at University of Manchester Institute of Science and Technology (UMIST),

Manchester, U. K. After obtaining his MS degree, he attended University of Florida and started

pursuing his PhD degree in the Chemical Engineering department. His research focused on

investigation of CulnSe2/CdS/ZnO material system and thin film photovoltaic device fabrication

and characterization.





PAGE 1

1 ZINC OXIDE SPUTTER DEPOSITION AND MODELING OF COPPER-INDIUM-GALLIUM-DISELENIDEBASED THIN FILM SOLAR CELLS By WEI LIU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Wei Liu

PAGE 3

3 To my beloved parents and family

PAGE 4

4 ACKNOWLEDGMENTS Dr. Crisalle is the chair of my superv isory committee and I express my sincere appreciation for his guidance and constant encouragement through the last five years. I am also obliged to thank Dr. Hoflund, Dr. Svoronos and Dr. Eisenstadt for their valuable discussions and serving as my research committee members. Of course my gratitude should be given to Dr. Holloway for his supports and suggestions base d on his profound knowledge on vacuum science. Dr. Anderson, Dr. Li also provided a lot of support during the last five years and I would like to express my appreciation. I also thank Dr. Craci un for his valuable discussions on thin film characterization. This work would not have been achieved wit hout the collaborations from my colleagues working on the CIGS solar cell projects at University of Florida. In particular, I hereby extend my thanks to Xuege Wang, Ryan Kaczynski, Le i Li, Ryan Acher and Andre Baran for their valuable discussions and collaborations on devi ce fabrication. Dr. Mark Davidson, Evan Law and Dr. Lei Qian from Dr. Holloways group also provided their kind helps at different points of this work. I would like to show my sincere thanks to them. Of course the acknowledgement would never be complete without expressing my gratitude to my parents. I own everything to them and the word is never enough to express my appreciation. Without th eir support and encouragement, I would not have made it to this point of my life.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES.........................................................................................................................9 ABSTRACT...................................................................................................................................12 CHAP TER 1 INTRODUCTION..................................................................................................................14 2 LITERATURE REVIEW.......................................................................................................18 Solar Light..............................................................................................................................18 The Photovoltaic Effect..........................................................................................................19 History of Photovoltaic Technology....................................................................................... 19 Device Physics........................................................................................................................20 Solar Cells using Di fferent Materials .....................................................................................22 Crystalline Silicon Solar Cells......................................................................................... 23 Single crystalline silicon.........................................................................................23 Polycrystalline silicon..............................................................................................24 Thin Film Solar Cells...................................................................................................... 25 Amorphous silicon...................................................................................................25 CuInSe2 and CuGaSe2..............................................................................................27 Transparent Conductive Oxide............................................................................................... 28 Sputter Deposition............................................................................................................. .....30 Introduction................................................................................................................... ..30 Advantages of Sputter Deposition................................................................................... 31 Sputter Deposition of ZnO.............................................................................................. 32 Summary.................................................................................................................................33 3 EXPERIMENTAL METHODS.............................................................................................40 ZnO Sputtering System...........................................................................................................40 Substrate Cleaning..................................................................................................................42 Characterization Techniques..................................................................................................43 Electrical Property Characterization................................................................................ 43 Four-point probe.......................................................................................................43 Hall measurement..................................................................................................... 44 Optical Spectrophotometer Characterization.................................................................. 45 Other Characterizations................................................................................................... 46 Profilometry............................................................................................................. 46 X-Ray Diffraction....................................................................................................47

PAGE 6

6 Atomic Force Microscopy........................................................................................ 49 Auger Electron Spectroscopy...................................................................................51 4 GROWTH AND CHARACTERIZATION OF ZINC OXIDE.............................................. 56 Optimization of Sputter Depositi on of Alum inum Doped ZnO............................................. 56 Experiments at Different Base Pressures........................................................................ 56 Optical property........................................................................................................ 57 Electrical property.................................................................................................... 58 Growth rate............................................................................................................... 58 Power Effect....................................................................................................................59 Working Pressure Effect.................................................................................................. 61 Characterization......................................................................................................................62 Auger Electron Spectroscopy Analysis...........................................................................62 Surface scan.............................................................................................................. 63 Al sensitivity............................................................................................................64 Depth profile............................................................................................................65 Atomic Force Microscopy............................................................................................... 65 5 EFFECT OF HYDROGEN ON SPUTTERED ZINC OXIDE .............................................. 78 Motivation from the Base Pressure Effect.............................................................................. 78 Experiments and Results........................................................................................................ .78 Hall Characterization....................................................................................................... 79 AFM and SEM Characterization..................................................................................... 80 Film Uniformity Study.......................................................................................................... .81 Summary.................................................................................................................................83 6 DEVICE FABRICATION A ND CHARACTERIZATION ................................................... 92 Device Fabrication Procedure.................................................................................................92 Substrate..........................................................................................................................92 Back Contact Sputtering..................................................................................................92 Absorber Growth.............................................................................................................93 Buffer Layer Deposition.................................................................................................. 94 Window Layer Deposition..............................................................................................95 Metallization....................................................................................................................96 Device Characterization System............................................................................................. 96 Devices with Absorbers from UF PMEE Reactor.................................................................. 97 CIGS Devices..................................................................................................................97 CGS Devices...................................................................................................................98 Effect of Zinc Oxide Window Layer on Device Performance............................................... 99 Effect of Window Layer Transmission......................................................................... 100 Effect of Sheet Resistance.............................................................................................101 Impact of the Intrinsic ZnO Layer................................................................................. 102 Summary...............................................................................................................................102

PAGE 7

7 7 INVERSE MODELING OF CIGS SOLAR CELLS ............................................................ 110 Definition of Inverse Modeling............................................................................................ 110 Inverse Modeling for Photovoltaic Cells.............................................................................. 111 Application of Inverse Modeling Perform ance Optimization......................................... 112 Approach....................................................................................................................... ........112 Forward Modeling.........................................................................................................113 Cell structure generation........................................................................................ 113 Input parameters..................................................................................................... 114 Example of a forward modeling............................................................................. 114 Modeling of NREL champion device....................................................................116 Inverse Modeling........................................................................................................... 116 Inverse Modeling of NREL champion device........................................................ 117 8 CONCLUSION..................................................................................................................... 130 Parameter Effects on Sputtered AZO................................................................................... 130 Device Fabrication................................................................................................................131 Defects in CIGS Absorbers..................................................................................................132 Future Work..........................................................................................................................132 LIST OF REFERENCES.............................................................................................................134 BIOGRAPHICAL SKETCH.......................................................................................................138

PAGE 8

8 LIST OF TABLES Table page 4-1 Estimated grain size using XRD data obtai ned for AZO deposited at different power. .... 77 4-2 Parameters used for collecting the A uger depth profile of AZO thin film s....................... 77 4-3 AFM surface roughness evolution with working pressure................................................77 5-1 Growth parameters comparison for AZO and H-AZO thin films...................................... 91 5-2 Deposition condition comparis on between two AZO thin films shown in Figure 5-5. ..... 91 5-3 Thickness, Rq and estimated grain si ze com parison between two AZO thin films shown in Figure 5-5........................................................................................................... 91 5-4 Roughness data evolution with increasing film thickness................................................. 91 5-5 Hall measurements results on two films with different positions under the target. ........... 91 5-6 Comparison of Hall and four-poi nt-probe m easurement results........................................ 91 6-2 Recipe for CdS buffer layer deposition........................................................................... 104 6-3 Composition and thickness da ta for UF CIGS absorbers. ............................................... 104 6-4 Calibration of the reference cell....................................................................................... 104 6-5 I-V character of the UF CIGS devices............................................................................. 105 6-6 Growth recipe and composition for UF CGS absorbers.................................................. 105 6-7 I-V character of the UF CGS devices............................................................................. 105 6-8 Performance of CIGS solar cells wi th different window layer transm ission................... 105 6-9 Performance of CIGS solar cells with and without i-ZnO layer ...................................... 105 7-1 Typical input parameters for Medici................................................................................ 129

PAGE 9

9 LIST OF FIGURES Figure page 2-1 Energy current density per wavelength of AM0 spectrum (heavy line) compared with blackbody radiation at 5800K(thin line)............................................................................ 34 2-2 Schematic of the Air Mass definition................................................................................ 34 2-3 Energy current density per photon energy of AM1.5 spectrum (heavy line) com pared with blackbody radiation at 5800K (thin line)..................................................35 2-4 Formation of built-in pot ential between p-n junction. ....................................................... 35 2-5 An equivalent circuit of a typical solar cell. ...................................................................... 36 2-6 Effect of series resistance and shunt resistance on sola r cell perform ance........................ 36 2-7 Effect of series resistan ce on solar cell perform ance......................................................... 37 2-8 World PV market share by technology in 2002................................................................. 38 2-9 Development history of thin f ilm solar cells. (Courtesy of NREL)................................... 38 2-10 Thin film CIGS solar cell structure.................................................................................... 39 3-1 Perkin-Elmer 4400 sputtering system................................................................................ 53 3-2 Schematic of the ZnO Sputte ring System (Perkin-Elmer 4400)........................................ 53 3-3 Schematics of four-point probe measurement................................................................... 54 3-4 Square shaped van der Pauw sample geometry................................................................. 54 3-5 Generation of Cu K radiation. ......................................................................................... 55 3-6 Braggs condition.......................................................................................................... .....55 4-1 The effect of base pressure on the op tical transm ission of sputtered AZO films.............. 68 4-2 Effect of base pressure on the resistivity of sputtered AZO fil ms..................................... 68 4-3 The effect of base pressure on the grow th rate of ZnO:Al thin films (Fixed conditions: Dts = 5 cm, P = 4 mTorr, Power = 400W)....................................................... 69 4-4 The effect of power on the resistivity of as grown ZnO:Al thin film s (Fixed conditions: Dts = 5 cm, Pb = 8.0 10-7 Torr, and P = 4 mTorr)........................................... 69

PAGE 10

10 4-5 The effect of power on the growth rate of ZnO:Al thin film s (Fixed conditions: Dts = 5 cm, Pb = 8.0 10-7 Torr, and P = 4 mTorr)......................................................................70 4-6 X-ray diffraction pattern of thin films deposited from a reduced ZnO target................... 70 4-7 X-ray diffraction pattern of thin films deposited from an oxidized ZnO target................ 71 4-8 High resolution XRD pattern for AZO film s deposited using different power (Fixed conditions: Dts = 5 cm, Pb=8.0 10-7 Torr, Pressure = 4mTorr)......................................... 71 4-9 The effect of working pressure on the resistivity of ZnO:Al thin film s ( Fixed conditions: Dts = 5 cm, Pb = 8.0 10-7 Torr, and Power = 500 W )....................................72 4-10 The effect of working power on the growth rate of ZnO:Al thin film s ( Fixed conditions: Dts = 5 cm, Pb8.0 10-7 Torr, and Power = 500 W ).....................................72 4-11 High resolution XRD pattern of AZO thin film s deposited at different working pressures (Fixed conditions: Dts = 5 cm, Pb=4.0 10-7 Torr, Power = 500 W)................... 73 4-12 AES survey scan pattern for an AZO th in film with organi c contamination on the surface................................................................................................................................73 4-13 AES scan results after sputtering for 3 minutes................................................................. 74 4-14 Al Auger peak shown by using increased primary beam energy....................................... 74 4-15 Depth profile for each elemen t of AZO obtained with AES. ............................................ 75 4-16 AFM graph of AZO thin films deposited with different working pressure. a) 2m Torr, Rq=2.2nm; b) 3mTorr, Rq=2.8nm; c) 4mTorr, Rq=3.6nm.................................................76 5-1 Effect of injecting hydrogen gas on th e resistivity of s puttered AZO film s...................... 85 5-2 Hall measurement results for aluminum-z inc-oxide thin film s deposited using only argon as the working gas (sample labeled as AZO) and using a mixture of argon with 0.1 wt% of hydrogen (sampled labeled H-AZO).......................................................85 5-3 AFM image of two AZO films deposited us ing different sputtering gas. a) AZO b) H-AZO ...............................................................................................................................86 5-4 AFM image of H-AZO thin films w ith increasing thickness. a)200nm b)400nm c)600nm d)800nm.............................................................................................................. 87 5-6 Variation of AZO thickness with different positions along th e axial direction. ................ 89 5-7 Variation of sheet resistance with diffe rent positions along th e axial direction. ............... 89 5-8 Variation of resistivity with differe nt position s along the axial direction......................... 90

PAGE 11

11 6-1 Schematic of the PMEE reactor....................................................................................... 106 6-2 Photo of CIGS solar cells fabricated on a 2 1 soda-lim e glass substrate.................... 106 6-3 Cross section SEM image of absorber #582.................................................................... 107 6-4 Cross-section SEM image of absorber #588.................................................................... 107 6-5 Cross-section SEM image of CGS device #640.............................................................. 108 6-6 AZO window layer optical transmi ssion of three CIGS solar cells. ................................ 109 6-7 CIGS cell efficiency versus the AZO layer sheet resistance. .......................................... 109 7-1 Schematic diagram illustrating the forward modeling and inverse modeling operations for a given physical system S ........................................................................120 7-2 Parameter classification with exam ples for solar devices................................................ 121 7-3 Procedure chart for so lar cell inverse m odeling.............................................................. 122 7-4 A typical CIGS cell structure........................................................................................... 123 7-5 Cell structure described in Medici................................................................................... 123 7-7 A typical I-V curve generated by Medici. .......................................................................124 7-8 Medici simulation result fo r an NREL cham pion device................................................ 125 7-9 Iteration of inverse modeling process for there so lar cell perfor mance variables: (a) Isc, (b) Voc, and (c) FF. (Case 1).......................................................................................126 7-10 Iteration of inverse modeling process for there so lar cell perfor mance variables: (a) Isc, (b) Voc, and (c) FF. (Case 2).......................................................................................127 7-11 Iteration of inverse modeling process for there so lar cell perfor mance variables: (a) Isc, (b) Voc, and (c) FF. (Case 3).......................................................................................128

PAGE 12

12 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ZINC OXIDE SPUTTER DEPOSITION AND MODELING OF COPPER-INDIUM-GALLIUM-DISELENIDEBASED THIN FILM SOLAR CELLS By Wei Liu December 2007 Chair: Oscar D. Crisalle Major: Chemical Engineering Highly transparent and conductiv e aluminum doped zinc oxid e (AZO) thin films were successfully obtained through RF magnetron sputtering using argon as the sputtering gas. Thin film AZO grown under different conditions such as base pressure, deposition power and working pressure was characterized. A correlation betw een various operational parameters and the optical and electrical properties of AZO was developed and used to suggest optimum of operational conditions. The sputtering process was improved by a dding a small amount of hydrogen into the sputtering gas. A significant improvement of th e conductivity of AZO thin films was observed. This observation together with results obtained under different base pressures supports the literature prediction that hydrogen plays a favorable role in n-type zinc oxide thin films. Hydrogen participates in the doping process and causes increased ca rrier concentration. The AZO thin films sputtered with the Ar and H2 mixture gas were in corporated in copperindium-gallium diselenide (CIGS) solar cells and devices with a thin film layer structure of the form ZnO/CdS/CIGS/Mo were fabri cated. It is found that the sh eet resistance and thickness of the AZO layer can be an important factor determ ining the performance of CIGS solar cells. The

PAGE 13

13 devices were measured under AM 1.5 radiation and a conversion efficiency of approximately 9% was achieved. To better understand solar cell devices fabric ated using chalcopyrite semiconductors such as CuInSe2 and CuGaSe2, an inverse modeling process was developed. The inverse modeling process was realized by integra ting two software tools, namely Matlab and Medici, under a Linux environment. Matlab provides the values of candidate input parame ters to Medici, which in turn outputs device simulation results that are fe d back to Matlab. This process continues until the outputs reach the target values. The inve rse modeling process successfully identified the defect distribution in a CIGS layer that matc hed a known cell performance. In an ensuing optimization step, optimal defect concentrations were found, revealing th at the shallow acceptor defect concentrations are favorable for increasing the efficiency.

PAGE 14

14 CHAPTER 1 INTRODUCTION With the decline of the traditional fossil fu el resources in the world, renewable energy sources are emerging as the future solution to maintain a sustainable development of human society. Among various kinds of renewable energy solutions including hydroelectric, geothermal, wind, and solar (photovoltaic and thermal), photovoltaic te chnology has attracted more and more attentions for its advantages over other options. Photovoltaic technology has the following advantages. The first advantage is the abundant energy source. The sun is a reliable energy source and it constantly radiates a trem endous amount of energy toward the earth. The second advantage is clean and environmentally friendly. Photovoltaic devices, commonly known as solar cells, convert the energy from sun light directly into el ectricity. There is no emission, noise or pollution of any kind during th e operation. With more and more concerns about the greenhouse gases and their detrimental eff ect on our planet, with the awareness of the possibility of irreversible environmental po llution that could be br ought by nuclear power, the value of PV technology as a clean, renewable energy source has been increasingly appreciated. The third advantag e is reliability. PV technolog y was originally developed for space applications where solar cells were used to power satellites. With no moving parts, solar cells can operate reliably and require little mainte nance. With all these advantages, solar energy is emerging as a promising alternat ive to traditional energy sources. The amount of light energy available annually in a particular landmass on earth depends on the climate and the latitude. In areas closer to the equator, available energy per unit area is larger. On average between 2 and 3 megawatthours (MWh) of solar energy falls on a square meter of land in the United States annually. Accordingly, computed annual solar energy available over the total land area of the United States is about 2.4 1016kWh, which is 10,000

PAGE 15

15 times larger than the annual electricity consumption of the United States [1]. The sun is a huge energy reservoir that could provide the in creasing need for energy consumption. Solar cells have been successfully applied in various filed to provide electricity. Examples of these applications are pow ering devices on satellites for space applications, providing electricity for remote areas where the power infrastr ucture is not available. Although cost is not a big concern in these situations, it becomes a ma jor problem for terrestrial applications where power grid is available. When compared with fossil fuels, the relatively high costs of semiconductor materials and processing are a major barrier for the large-s cale implementation of solar devices. The high material and fabrication cost keeps the price of solar panels at a high level. As a result, the huge investment on thes e solar panels may keep consumers from choosing solar electricity considering the fact that it ma y require tens of years to return the initial investment and start saving consumers money. Until low cost, highly efficient photovoltaic devices are ready to be fabricated, the application of solar elect ricity is unlikely to replace the traditional energy source and become the mainstream. Therefore, it is imperative to develop low cost processes and improve performance of PV materials to bring down the cost per watt for solar electricity. In order to reduce the cost of PV systems, which is usually evaluated in dollars per kilowatt hour ($/kWh), the current R&D is concentrated on both improving the conversion efficiency and reducing the semiconductor material cost as well as the processing cost. The average residential price for utility-generated electric energy in the U.S. in 1994 was US$0.079/kWh. In order to reach a cost target as US$0.06/kWh for electricity from a PV plant operating for 30 years, the module efficiencies are re quired to be in the range of 15% to 20% for a flat plate panel system and 25% to 30% for a system operating under co ncentrated sunlight.

PAGE 16

16 These correspond to module area costs of $45 to $80/m2 and $60 to $100/m2, respectively [2]. Although the current fabrication cost is still beyond these values, with the favorable green energy policy from governments, such as government so lar rebates, more and more solar panels are installed to provide clean utility energy. Single crystal silicon solar cells have achieve d good performance with efficiencies beyond 20%. However, due to the high cost nature of the Czochralski process used to produce single crystal silicon, the pr ice for silicon solar panels remains high. Direct bandgap thin film materials have much higher light absorption coefficient th an silicon and require a much less material thickness to absorb the light. Th ey are possible alternatives that could significantly lower the material consumption and consequently the cost Three typical thin film PV materials are amorphous silicon, cadmium telluride, and copper ch alcogenides. Among these thin film PV materials, copper indium gallium di-selenide (CuInxGa1-xSe2 or simply CIGS) is very attractive for its high performance, long-term stability and rela tively little toxicity. The work in the scope of this dissertation includes both theoretical analysis of CIGS solar cell device physics through simulation and experimental fabrication and char acterization of this kind of solar cells. A process to grow device quality zinc oxide window layer using an industry sputtering system was developed and optimized. Aluminum doped zinc oxide (AZO) thin films with good optical transmission and resistivity between 1 10-3 cm and 2 10-3 cm were repeatedly obtained. It is also observed th at residue water vapor in the sputtering chambe r is beneficial to the electrical conductivity of aluminum doped zi nc oxide. Motivated by this observation, a modified process using hydrogen and argon mixtur e as the sputtering ga s was implemented. Resulted zinc oxide thin films show a lower re sistivity. Hall measurement shows that this lowered resistivity was achieved through the incr ease of carrier concentr ation in the film.

PAGE 17

17 This process was then successfully applied in CIGS photovoltaic device fabrication. Efficiency of approximately 8% and 9% was achieved on CIGS from Energy Photovoltaic Inc. (EPV) and those from the PMEE reactor at the Univ ersity of Florida, respectively. This is the first time that CIGS photovoltaic devices are comp letely fabricated in house, achieving the goal of two previous attempts [3], [4]. Although laboratory scale CIGS devices with efficiency of 19.5% have been achieved experimentally, this material system is not fully understood and many issues remain unknown. For example, the defect distribution in this la yer has big impact on the performance of the final device and it is of great interest to investigate the relation between defects and cell performance. In order to do this, software Medici from Synopsys Inc was utilized in simulation to predict solar cell performance under different conditions. An inverse modeling was implemented to identify the defect distribution and their imp act on CIGS devices. Solar cell performance was also optimized using the inverse modeling program.

PAGE 18

18 CHAPTER 2 LITERATURE REVIEW Solar Light The sun is a hot sphere of gas whose intern al temperatures reach over 20 million degrees Kelvin due to nuclear fusion reactions at the sun's core converting hydrogen to helium. The radiation from the inner core is strongly abso rbed by a layer of hydrogen closer to the sun's surface. Heat transfers through this layer by co nvection. The surface of the sun, called the photosphere, is at a temperature of about 6000K a nd can be represented as a blackbody radiation source. An immense amount of solar energy radiates to ward the earth at any time. The radiation power density at the suns surface is about 5.96 106 mW/cm2 and it decreases with increasing the distance from the suns surface as the radiat ion sphere gets larger. When reaching the atmosphere of the earth, the sunlight car ries a power density of about 135.3mW/cm2 [5]. This spectral distribution is referred to as Air Mass Zero (AM0) since it has not passed through the air. Figure 2-1 shows a comparison between AM0 radiation and a bl ackbody radiation at 5800K [6]. During the process of going through the eart h atmosphere, sunlight will be absorbed and scattered and its energy is reduced. Different incident angles will ca use different amount of energy loss and this is reflected by Air Mass. In the situation shown in Figure 2-2, Air Mass is defined by AM= P P 1 2 1 cos (2-1) where P1 is the actual path length that light takes through the atmosphere and P2 is the shortest possible path length, i.e., the normal inci dent distance. For example, AM1.5 refers to sunlight passes through 1.5 atmosphe res with a solar zenith angle ( in Figure 2-2) of 48.19o.

PAGE 19

19 AM1.5 is usually used as incident light for testin g terrestrial solar device s. Figure 2-3 shows the comparison between AM1.5 radiation a nd a blackbody radiation at 5800K [6]. The Photovoltaic Effect Photovoltaic effect is the crea tion of an electromotive force by the absorption of photons in a semiconductor device. When photons with ener gy higher than the energy bandgap of a specific material are radiated on the material, they can be absorbed and the photon energy will excite electrons from the valance band to the conduction band and consequently generate electron hole pairs. To make a solar cell, the electron hole pa irs need to be separated before they recombine with each other. Semiconductor p-n junctions were successfully applied to realize this process. When a p-type semiconductor contacts with an ntype semiconductor, a built in electric filed will be generated. This electric field drives any electrons generated in the p-type semiconductor across the junction to the n-type semiconductor. Thus the electron-hole pairs are separated. History of Photovoltaic Technology The photovoltaic effect was first discovered by Frenchman Edmund Becquerel in 1839. He observed a special effect while experimenting with an electrolytic ce ll made up of two metal electrodes. When the metal plates immersed in a suitable electrolyte ar e exposed under sunlight, a small voltage and current were produced. Th e first solid-state materials that showed a significant light-dependent voltage were sele nium in 1876 and later cuprous oxide, which indicated that semiconductors would eventually become the most promising materials for photovoltaic energy conversion. PV technological development began with th e build up of a silicon p-n junction by a group of researchers at Bell laborato ry in 1954 [7]. But the powerful driving force for the PV technological development did not come until th e advent of space exploration. The urgent demand for a reliable, long lasting energy source to power the spacecraft became the motivation

PAGE 20

20 of solar cell development. By 1958, the first silicon solar cells were applied on spacecraft. Interest in solar cells as a cl ean, renewable terrestrial energy s ource arose after the oil embargoes in 1973 and 1979. Shay et al .[8] at Bell Labs first discovered that the I-III-VI2 semiconductor (CuInSe2) can be effective used as a photovoltaic mate rial in the 1970s. A polycrystalline CuInSe2 cell of nearly 6% efficiency was developed at the Univ ersity of Maine in 1976 [9]. Over a period of more than one decade starting from early 1980s a research group at Boeing was able to consistently push up the efficiency of polycrystalline chalcopyrite solar cells to a higher value when it reached 13.7% in 1993 [10]. The Boeing CIS project was ceased not long after that. The research of CIGS technology are later led by researchers at the National Renewable Energy Lab (NREL), who reported an conversion efficiency beyond 19% in 2003 [11]. CIGS technology has also been transferred to industr y pilot production both in Europe and the United States. It is becoming a promising thin f ilm technology that can dramatically lower the production cost. Device Physics Most solar cells rely on p-n junction to se parate the photon generated electron and hole pairs. The p-n junctions are formed by joining a p-type and n-type semiconductor together. Once connected, diffusion happens due to the con centration difference of electrons and holes between these two types of semiconductors. Si nce the n-type semiconductor has a high electron concentration and the p-type a high hole concentr ation, electrons diffuse from the n-type region to the p-type region. Similarly, holes flow by diffusion from the p-type region to the n-type region. Because electrons and holes are charged, their diffusion causes ch arge redistribution in the junction area and an internal electric filed is formed across the junction. The region where the electric field is present is called space charge region. All free carr iers in the space charge

PAGE 21

21 region are quickly swept away by the electric fiel d and therefore this layer is also called as depletion region due to the unavailability of free carriers. As a result of the electric field, a built-in potential is present across the junction. This process can also be explained by the Fermi level alignment between the p-type and n-type re gion due to the diffusion of electrons and holes, which is shown in Figure 2-4. Under thermal equilibrium, both diffusion and drift happens and their current balance and there is no net current flow through the junction. When the p-n junction is illuminated by photons whose energy is larger than the bandgap, electron and hole pairs are generated. These free carriers generally have very short lifetime and they will recombine with each other if they are not separated in a timely fashion. However, the built-in potential across the junction will swif tly sweep the electrons and holes across the junction so that photon generated electrons reaches the n-type region and holes the p-type region, where they become the so called majority carriers and no longer su ffer a fast recombination. If the p-n junction is connected with external circuit, the light ge nerated current flows out and the semiconductor p-n junction becomes a solar cell. If the p-n junction is not ex ternally connected, i.e., under open circuit condition, the photon genera ted current will produce a forward bias across the diode, which reduces the barrier for diffusion and promotes the diffusion current to balance out the photon generated current. At equilibri um, the net current still remains zero. This forward bias created is referre d to as open circuit voltage (Voc). Figure 2-5 shows an equivalent ci rcuit of a solar cell [12]. Rs represents series resistance. Series resistance comes from the re sistance of the material that c onstitutes the solar cell and the contact resistance between the body mate rial and external circuit. Rsh represents shunt resistance, which is the resistance in parallel with a solar cell that re sults in current leakage losses within the cell. An ideal solar cell woul d have zero series resistance and infinite large

PAGE 22

22 shunt resistance. The effects of series resist ance and shunt resistance on solar cell performance can be found in Figure 2-6 [13]. From the calculated results show n in the graph, the solar cell current-voltage (I-V) curve ( i.e. performance) is strongly affected by the series resistance while shunt resistance plays a less important role. The results show that the I-V curve changed significantly when the series resistance increased from 0 to 5 Instead when the shunt resistance reduced from infinite large to 100 there is no major change for the I-V curve. Thus series resistance is a dominating factor in determ ining the power available from a solar cell. The performance deterioration mainly comes from the rapid drop of fill factor (FF), a response of a slight increase of series resist ance. Figure 2-7 shows how quickly an increasing series resistance affects the performance of a solar cell. This result shows that it is critical to obtain a low series resistance, i.e. to reduce the resistance of the internal material that consti tutes the device. This is the motivation for the efforts to deposit transparent zinc oxide with extrem ely low resistivity, which is part of the work in the scope of this dissertation. Solar Cells using Different Materials Among the various materials that have been tested as absorber layer for PV devices, several are quite successful in their applications Examples of them are crystalline silicon, amorphous silicon and a wide variety of compoun d semiconductors such as GaAs, InP, CuInSe2 and CdTe. These materials can be generally divided into two categorie s based on their energy band structure. One is called direct band gap mate rial and the other is referred to as indirect band gap material. For direct ba nd gap materials, the maximum of the valence band energy and the minimum of the conduction band energy show at the same crystal momentum values. The indirect band gap material does not have this pr operty. This band gap structure difference leads

PAGE 23

23 to the fact that direct band gap materials have a much higher absorption co efficient than indirect materials. Consequently, for indi rect band gap material such as crystalline silicon, it requires more thickness of the material and more critical preparation techniqu e to produce a qualified absorber layer with reasonable efficiency. Fo r example, a silicon f ilm of thickness about 100 microns is required to absorb 95% of the photons in the solar spectrum [14] while for a direct band gap material, a film of less than 1 micron th ickness is capable of absorbing more than 90% of the incident sunlight. Because of the large thickness of material required, the silicon film has to be ultra-high pure for the generated minority ca rriers to travel the large distance towards the junctions without being captured by defects. Single crystal silicon used in PV applications is usually fabricated by the Czoc hralski (CZ) proces s [15], which signifi cantly elevates the fabrication costs. Nevertheless, silicon technology still dominates the current PV market as a mature technology. Figure 2-8 shows the techno logy market share in 20 02 [16]. Some of the important PV materials are briefly in troduced in the following sections. Crystalline Silicon Solar Cells Silicon is the most extensively studied semiconductor and was the first commercially used PV cell material. It is a cheap and abundant element that can be found in sand and other minerals. Silicon processing is very mature through the continuous development of integrated circuit processing technique si nce the 1950s. This provides great background theory and technical support to silicon solar cell research. Single crystalline silicon High purity single crystal silicon solar cells are the first commercially developed solar cells. Efficiency of single crystal silicon has be en beyond 20% as early as in the 1980s [17]. Commercial single crystalline silicon cel ls usually have lower efficiency.

PAGE 24

24 Major drawback of single crystalline silicon so lar cells are the high cost and silicon supply problem. Silicon wafers need to be cut off fr om ingots produced by Czochralski (CZ) process, which is a costly method. As an indirect bandgap material, it requires thicker silicon consequently more amount of mate rials than thin films to build a high efficiency cell. About half of the cost of a silicon module comes from the co st of single crystal silicon wafers. With the high demanding form the semiconductor industry, pr ice of silicon wafers remains high. The PV industry has in the past used re jected silicon from the semiconduc tor industry that was available at a lower cost. This creates a dependence on th e source materials and re stricts the development scale of this technology. The high material cost and processing cost together with the competition of wafer consumptions from the inte gration circuit industry led to the development of alternative solar cell materials. Polycrystalline silicon Polycrystalline silicon is typically produced by block casting and solidification of silicon[18], which avoids the costly pulling proc ess used by the CZ method, resulting a process much more economic compared with the strict CZ process. Silicon is melted and poured into a square graphite crucible to create silicon bloc ks with a square shape, which yields better utilization of the wafer when la ter assembled into modules. With controlled cooling process, polycrystalline silicon block with a large crysta l grain structure can be produced. The grain size can be obtained is about millimeters or centimeters. The silicon blocks are then sawn into wafers. A waste of about 50% of the silicon ca n happen during the process of sawing. Polycrystalline silicon is not used in integration circuit i ndustry and is only produced for fabrication solar cells. Although it typically has a lower solar cell efficien cy, it is cheaper than single crystal silicon. Another advantage is that the square shaped wafers can easily be manufactured into square sola r cells compared to round shaped pulled silicon crystals. Silicon

PAGE 25

25 based PV technology is the most advanced and they share the largest perc entage of the current market. Thin Film Solar Cells Materials used for thin film solar cells are direct band gap semiconductors with high absorption coefficients. With a thickness less than a few microns, they are able to absorb most of the incident sunlight. Because the film thickness is small, generated electron-hole pairs only need to travel a small distance to be gathered in th e collector. The loss due to recombination is greatly reduced compared with crystalline silicon. Thus, th e purity requirement is not as strict as that for silicon, which may lead to a significant drop fo r the manufacturing costs. Amorphous silicon, CdTe and CuInSe2 (or its alloys) are three type s of most promising material systems for thin film solar cells because of a good combination of reas onable efficiency, reliability and low production costs. The development history of these three thin film materials is shown in Figure 2-9. Amorphous Silicon (a-Si:H) is the first thin-film material that has been commercially fabricated into photovoltaic cells and it holds the largest market sh are in thin film photovoltaic applications. It is also the mo st commercially developed thin film technology. CdTe and CIS technology have also been commercially develo ped recently. A brief introduction of amorphous Si and CIS technology is introduced next. Amorphous silicon Amorphous silicon is a disordered cousin of crystalline silicon. Although each silicon atom in amorphous Si has the same number of nearest neighbors as crystalline s ilicon, the bond angles connecting it to the neighbors are unequal. In stea d of the 109 angles in crystalline silicon, the bond angles of a-Silicon are not constant. This disordered structure leads to a significant difference between amorphous Si and crystalline silicon. For example, instead of having an indirect band gap of 1.1eV (as for crystalline si licon), it has a direct ba nd gap of about 1.75eV.

PAGE 26

26 The average absorption coefficient of a-Silicon is much larger than that of single crystalline silicon over the visible light spec trum, which makes it possible to absorb the incident sunlight with a much smaller film thickness. As stated before, a-Si has the most commer cially advanced technology among all thin film materials. It is actually the only thin film mate rial system that has an impact on the current PV market. But there is a major problem with aSi solar cell system which holds back the development of reliable high efficiency a-Si m odules. It is the performance degradation under illumination caused by light-induced photoconduc tivity degradation [19], a phenomenon known as Staebler-Wronski effect [20]. To improve the performance of a-Si solar cell, a critical technique called light trapping [21] is utilized to increase the light path length in a-Si, thus enhance the amount of photons to be absorbed. Transparent Conductive Oxide (TCO) plays a critical role in rea lizing light trapping. TCO can be deposited or post etched to form a rough surface with features serving as reflection mirrors. Light reaching these surface features can be scattered back into the device. By applying this technique appropriately, short circuit curr ent can be significantly increased. Traditional TCOs like Indium Tin Oxide (ITO) and SnO2 were originally applie d to serve light trapping functions. But for superstrate structured a-Si solar cells, a-Si is typically deposited on top of TCO using PECVD (Plasma Enhanced CVD) pro cess, during which the TCO is exposed to hydrogen plasma. It is found that the electrical conduc tivity and optical transmission of these traditional TCOs deteriorate when exposed to hydrogen plasma while ZnO as a new TCO material shows more resistance to hydrogen plasma [22]. Radio frequency magnetron sputtering can be used to deposit highly conductive and transparent ZnO:Al films to serve the light tra pping purpose. But films grown by this technique

PAGE 27

27 generally are smooth and do not have a light trap ping structure. A wet etching process using hydrochloric acid is typically used to create surfac e roughness [23]. Amorphous Si based solar cells can be fabric ated in two configurations: substrate and superstrate structure. Substrate type modules ar e built on stainless steel foil which has been first covered with a textured back re flector and superstrate structured devices are built on glass coated with TCOs. Tandem structured Si solar cells with a-Si as top cell are also fabricated to further increase the total conversion efficiency. CuInSe2 and CuGaSe2 In the mid-1970s, a small group at Bell Labora tories in New Jersey began to work on several compound semiconductors for advanced p hotovoltaic applications. Among the materials studied, there was a category calle d chalcopyrite such as CuInSe2 (CIS), CuGaSe2 (CGS) and CuInS2. They were successful to produce a singe crystal CuInSe2 based cell which reached 12% efficiency in 1975. This spurred the interest with the exploration of ternary and quaternary compound semiconductors on the app lication in photovoltaic filed. CIS is a direct band gap material and is th e most light absorbing semiconductor known. The absorbance can be as high as 105 cm-1. Half a micron of CIS is sufficient to absorb 90% of the incoming photons, making it possible to achieve high efficiency with much less material consumption than the silicon PV technology. Be sides, this material system also possesses excellent stability and it is espe cially radiation hardne ss [24], which leads to a long life-time with consistent performance. Figure 2-10 shows a typical CIGS solar cell stru cture. The typical substrate is glass. Flexible and light weight substrate such as me tal foil can also be used. A thin layer of molybdenum is followed as the back contact. Then p-type CIGS comp ound is grown on top of Mo. After CIGS is a thin laye r of cadmium sulfide (CdS) with n-type conductivity. Next to CdS

PAGE 28

28 are the double transparent layers of intrinsic and doped ZnO. Optional MgF2 antireflection layer can be evaporated on top of doped ZnO. Finall y, it is nickel and aluminum dual metal layer serving as the top contact of the device, wher e nickel serves as a diffusion barrier between aluminum and zinc oxide. CIS has a bandgap of 1.04 eV. This is smaller than the bandgap of silicon. Devices made out of CIS typically have a small open circuit voltage and a large short circuit current. This leads to a relatively large series resistance loss and compromise cell performance. Gallium can be added to increase the band gap of the compound and form an alloy typically referred to as CIGS. Sulfur can also be added to increase the ba nd gap. High performance CIGS based devices generally require very high temperature to pr oduce the absorber layer and the fundamental reason for that is not clear. With a relatively large bandgap of 1.68 eV, ternary compound CGS is also a very interested material under investigation. La rge bandgap CGS can be used as top cell for multijunction solar cells (tandem cells) and it responds to the short wavelength part of the solar spectrum and allows long wavelength photos to pass through itself and reach the bottom cell. It is estimated that CGS/CIS tandem structure cells can reach efficiency beyond 30%. For CGS/CIGS tandem structure, modeling work pr edicted 25% conversion efficiency can be achieved [25]. Transparent Conductive Oxide Because transparent electrode is an essential ingredient in most optical-electronic devices, extensive research has been done on both material selection and fabricat ion techniques. The most popular materials used to develop highly tr ansparent and conductive thin films are n-type semiconductors consisting of metal oxides.

PAGE 29

29 Transparent and electric ally conductive films have been pr epared from a wide variety of materials, including semi-conducting oxides of tin, indium, zinc and cadmium nitride [26], and they are commonly referred to as Transparent Conductive Oxide (TCO). Among these, the TCO thin films composed of binary compounds such as SnO2 and In2O3 have been developed and doped SnO2 (Sb or F doped SnO2, e.g. SnO2:Sb or SnO2:F) and In2O3:Sn (indium tin oxide, ITO) are in practical use. In additi on, ternary compounds such as Cd2SnO4, CdSnO3 and CdIn2O4 have also been developed but have not yet been used widely [27]. One advantage of binary compound TCOs is that their chemical composition during th e film deposition is relatively easier to control than that of ternary compounds and multi-com ponent oxides. Typical undoped binary compound TCO films such as SnO2, In2O3, ZnO, and CdO are n-type degenerate semiconductors, with electron concentrations of the order of 1020 cm-3. To further lower the resistivity, doping materials are added into the films to enhance the free electron concentration. The lowest resistivity obtained is on the order of 10-5 cm, based on CdO:In films prepared for flat-panel displays. However, these were not practically used because of the toxicity of Cd. In practical films, the lowest resistivity comes from ITO films with a value of approximately 1-4 -cm. Indium is an expensive material because the amount of Indium in nature is very small. There are no mines intentionally built to obtain indium simply because of the trivial amount available. In fact, indium is only obtained as a byproduct from mines for Zn and other metals. To obtain even lower resistivity than that of ITO and find an alternative to ITO, doped ZnO films were developed in the 1980s. Unlike In, Zn is an in expensive, abundant and harmless material. Recently, ZnO:Al and ZnO:Ga doped thin films, with a resistivity of the order of 1-4 cm, have attracted much attention for deployment as the transparent electrode for thin film solar cells. It is believed that doped ZnO is a promising altern ative to ITO.

PAGE 30

30 Thin film ZnO has an energy bandgap greate r than 3eV and can be deposited with high transparency. If heavily doped with Al or Ga it can be very conductive as well. The high transparency allows the light to go through, reaching the absorb er layer, where electron-hole pairs are generated. Highly doped conductive ZnO films easily form an ohmic contact with Al, which is essential for collecting the current generated from the cell. Intensive research work on ZnO has been done for solar cell applic ations [28], organic light-emitting diodes [29-31] and transistors [32-34]. Thin films of zinc oxide can be prepared using various techniques such as spray pyrolysis [35-37], evapor ation [38-40], chemical vapor deposition (CVD) [41-43], and sputtering [44-46 ]. The sputtering technique can produce asdeposited, highly conductive ZnO films without a ny post deposition treatment. Key advantages of sputtering are discussed in the next section. Sputter Deposition Introduction Sputter deposition of films, which is often re ferred to as sputtering, was first reported by Wright in 1877. Edison patented a sputtering deposition processes to deposit silver on wax photograph cylinders in 1904[47]. Sputter deposition is a physical vaporizat ion of atoms from a surface by momentum transfer from bombarding energetic particles, which generally ar e ions generated from a glow discharge or a plasma process. The ions develo ped in the plasma are accelerated in an electric filed and bombard at a target, which is the source of the deposited material. After ion impact, the eroded material then condenses from the gas phase onto the substrate wher e the film is grown. To attract the positive ions, the target is applied a negative electric filed. Therefore the sputter deposition source is also refe rred to as the cathode.

PAGE 31

31 Sputter deposition can be used deposit films of either element or compound material. When sputtering a compound thin film, the target can be either compound or elemental. In the later case, a reactiv e gas is incorporated in the sputtering gas and the compound is formed during the sputtering process. In some cases, sputtering a compound target can re sult a loss of some of the more volatile material (such as oxygen from ZnO) and the stoichiometr y of the film may be different from that of the target. By adjusti ng the deposition parameters or add the volatile component into the sputtering gas to compensate the loss, resulted films can reach a desired stoichiometry, which is important fo r transparent conductive zinc oxide. Advantages of Sputter Deposition Sputter deposition is a flexible thin film growth process. It can be done under different vacuum conditions varying from below 10-5Torr up to approximately 50mTorr. High vacuum sputter deposition using ion beams can reduce the damage of substrate from the plasma environment. Sputter deposition can deposit film s of element, alloy a nd compound from either conductive or insulating target. There are also some of other advant ages can be achieved through sputter deposition. One of the major differences between sputtering as a plasma process and thermally excited thin-film preparation methods such as evaporatio n, chemical vapor deposition is the much higher energy input into the growing film. The rela tively high kinetic energy of the adatoms with values of a few electron voltage can be achieved in contrast to around 100 meV for evaporation process. The high kinetic energy increases the thermal surface mobility of the adatoms, allowing more rearrangement of the atoms on the surface, c onsequently yielding improved film properties. The target is generally cooled and the ener gy of ejected particles comes from momentum transfer form the bombarding ions. The cold surf ace presents little radiant heating to the vacuum system, which is an advantage over evaporati on, where radiant heating can be significant,

PAGE 32

32 especially in the case of long period intensive evaporation. The small amount of heating makes sputter deposition an appropria te process to deposit films on heat sensitive substrates. The sputtering target can provide a stable, la rge area, solid vaporization source. Because the vaporization comes from a solid surface, the sputter geometry can be up, down or sideways, in contrast to evaporation from a melted source inside a crucible. Sputter Deposition of ZnO Thin film ZnO can be produced with different sputtering techniques such as radio frequency sputtering, direct curr ent reactive sputtering and middle frequency reactive sputtering. The properties of sputtered ZnO films are depend ent on many factors. Comprehensive studies on parameters such as deposition power, workin g pressure, substrate temperature, oxygen concentration, and target dopi ng concentration have been carried on [48,49]. Individual parameters such as RF power [50], target to s ubstrate distance [51] are also investigated. Although the research on sputtered ZnO is extensive, the reports fr om different authors are quite different. It seems that the optimum operation co ndition is equipment specific and there is not a universal recipe that guarantees ZnO films with good electrical and optical properties. Therefore, it is essential to determine the optimum parame ter set for the UF oxide sputtering system. Sputtered ZnO thin films have been successfully used for various optoelectronic devices such as organic light emitting diodes (OLED) and so lar cells. Particularly fo r thin film solar cells, ZnO works well for different absorber materials such as CdTe [52], a-Si [53], CIGS [54]. There is also effort made to reduce the visible and pa rticularly nearinfrared optical absorption of ZnO films to further improve the perf ormance of solar cells [55]. Recently, producing p-type ZnO has been a popular research area due to the interest in fabricating homojuction wide ba ndgap ZnO devices for optoelectronic applications. Sputtering

PAGE 33

33 has been successfully demonstrat ed to produce p-type ZnO [56]. The thermal stability of p-type ZnO needs to be improved for reliable applications. Summary Thin film ZnO has broad applications in optoelectronic devices including light emitting diodes, lasers and solar cells. Sputter deposition is an effective method in producing n-type ZnO. The optical and electrical prope rties of sputtered ZnO can be affected by various deposition parameters. An optimization of the sputte ring condition is necessa ry to produce highly conductive yet transparent ZnO films that are required for photovoltaic applications.

PAGE 34

34 Figure 2-1. Energy current density per wavelength of AM0 spectrum(heavy line) compared with blackbody radiation at 5800K(thin line). Figure 2-2. Schematic of the Air Mass definition. P1 P2

PAGE 35

35 Figure 2-3. Energy current density per photon en ergy of AM1.5 spectrum (heavy line) compared with blackbody radiation at 5800K (thin line). Figure 2-4. Formation of built-in potential between p-n junction.

PAGE 36

36 Figure 2-5. An equivalent circ uit of a typical solar cell. Figure 2-6. Effect of series resistance and shunt resistance on solar cell performance.

PAGE 37

37 (a) (b) Figure 2-7. Effect of series re sistance on solar cell performance.

PAGE 38

38 Figure 2-8. World PV market share by technology in 2002. Efficiency (%)CuInSe2CdTe Amorphous silicon (stabilized)Univ. of Maine Boeing Boeing Boeing Boeing ARCO AMETEK Photon Energy NREL Boeing EuroCIS Univ. of So. Florida Univ. of So. FL BP Solar Kodak Kodak Monosolar Matsushita 12 8 4 0 2000 1995 1990 1985 1980 1975 United Solar 16 20 NRELThe Best One-of-a-Kind Laboratory Cell Efficiencies for Thin Films (Standard Conditions) 2005 RCA ECD NREL Figure 2-9. Development hi story of thin film solar cells. (Courtesy of NREL)

PAGE 39

39 Figure 2-10. Thin film CIGS solar cell structure CdS ( 2.4eV ) CuIn1-xGaxSe2 M g F2 Mo ZnO:Al (3.3eV) 0.1 m 0.05 m 0.05 m 1.5-2.0 m 0.5-1.5 m 0.5-1.5 m i-Zn O Ni/Al Glass 0.05/0.5 m

PAGE 40

40 CHAPTER 3 EXPERIMENTAL METHODS ZnO Sputtering System Sputter deposition was adopted as a room temperature process which could keep the pre-deposited CIS and CdS layers from being destroyed by high temperatures. The equipment used for ZnO sputtering is a Perkin-Elmer 4400 industry sputtering system. A photo and schematic of the equipment are shown in Figure 3-1 and Figure 3-2, respectively. It is equipped with a stainless steel, load-locked vacuum syst em to generate the vacuum required for sputter deposition. Typical base pressure of the system is approximately 6 10-7 Torr. However, it requires long period of pumping to reach this pr essure due to the large size and surface area of the substrate transfer pallet. The system originally has three RF planar sources, each handling an 8 inch target in a sputter down geometry. Targets cu rrently being used are Mo, ZnO:Al2O3 (98wt%:2wt%) and Zn. An 8 inch diameter magnetron was later added on top of the ZnO:Al2O3 target to confine electrons close to the target and increase the sput ter rate. Uniformity shields are integrated 1.6 cm below each target to compensate the short ex posure time of substrates placed at large radial distance due to the higher angular velocity when rotating. These shields modify the flux rate from the target (blocking flux in small radial distance areas) to account for the different exposure time along the radial direction of the platen, thus generate a uniform film when the platen is rotated. Because the plasma distribution fo r magnetron sputtering a nd diode sputtering is different, the shape of the uniformity shields fo r the two cases are different. These uniformity shields are removed for experiment s without substrate motion. The sources and the substrate table are cooled by chilled water. The substrate platen has rotary motion capabilities, allowing deposition of uniform films on up to f ourteen 4 inch wafers

PAGE 41

41 in a single batch process. The motion system for the substrate platen is controlled with a stepping motor, which gives accurate control of the substrate motion. It can generate full rotation in either clockwise or counter clockw ise direction at precise speed and acceleration, designated number of oscillations about a region below the target Although these capabilities can be of utility for complicated procedures, accurate positioning of the substrates below the sources are the primary uses for the motion syst em. Generally motion on th e substrate platen is not adapted during deposition to avoid generating particles, whic h could contaminate the vacuum environment. The throw or substrate to ta rget distance can be set between 5 and 9 cm. The RF power is supplied by a Randex 2kW RF generator which provides a radio frequency of 13.56 MHz. This power is delive red to the target through an impedance matching network. There is a switch avai lable to route the power generate d by the single generator to the appropriate source, or to the subs trate table when sputter etching is used to clean or modify the substrate. A CTI CryoTorr 8 cryopump and a Leybold Triv ac 60cfm rotary vane pump provide the high vacuum and rough pumping, respectively. A throttle valve between the cryopump and the deposition chamber reduces the pumps effective pumping speed by up to a factor of 50 when engaged. Two mass flow controllers (MFCs) are used to regulate introduction of Ar and O2 gas, and are used in conjunction with the throttling valve to genera te a controlled gas ambient over the pressure range of 2-150 mTorr, a range typi cal for sputter deposition. Both MFCs have a range of 0-100 standard cubic centimeters per minute (SCCM), a nd control gases from ultrahigh purity gas bottles. The system operates at a typical base pressure of 6x10-7 Torr. Vacuum gauging is provided by an array of two Baya rd-Alpert ionization ga uges, two Convectron gauges, three thermocouple gauges, and one MKS Baratron capacitance manometer. The

PAGE 42

42 Bayard-Alpert gauges are used for meas urements at high vacuum conditions (<10-5 Torr), and are therefore used to determine the base pressure of the system. The base pressure is a critical factor in controlling contamination in the grow ing film. The other critical gauge is the MKS Baratron gauge, which gives accurate, precise, fast and direct measurement of the pressure for the range of 10-5 to 1 Torr, and therefore spans the pressu re range used for sputter deposition. A direct gauge refers to the fact that the gauge measures the gas pr essure directly, instead of using heat transfer rate to determine the ambient pre ssure. A direct gauges ca libration is therefore independent of the gas being measured. This is critical for generating ambient with controlled partial pressures of a multiple gas species. Substrate Cleaning ZnO thin films were deposited on 1mm th ick soda-lime glass substrates for characterization purpose. The substrates are first rinsed in sprays of de ionized (DI) water to remove any major contaminations attached on the surface. They are then vertically placed in a Teflon substrate holder. The substr ate holder is then put in a hot DI water bath at approximately 80C for 10 minutes with all glass substrates completely immersed in the bath. The substrates are then taken out and dipped in DI water diss olved with Alconox deterg ent. They are then individually taken out of the hol der and physically scrubbed with a clean brush and then put back in the holder. Both the sides and edges are clean ed to remove all possible contaminations. The substrates are rinsed again in sprays of DI wa ter and put in a second hot DI water bath in an ultrasonic cleaner. All substrates are complete ly immersed in the bath and the ultrasonic cleaning takes about 10 minutes. The holder are then removed from the ultrasonic cleaner and placed in a three stage cascade rinser agitated wi th nitrogen bubbles. Finally the substrates are individually dried with nitrogen gun and placed in storage boxes.

PAGE 43

43 For CIS device fabrication, the soda-lime gla ss substrates coated with Mo, CIS, CdS are stored in vacuum sealed bags to prevent possibl e oxidation from air. The bags are only opened before the ZnO deposition and the substrates are qui ckly put into the vacu um chamber after they are exposed in air. Characterization Techniques To fully understand the various aspects of se miconductor property, a number of material characterization techniques are utilized. Th e electrical, optical and other properties are comprehensively investigated us ing the following techniques. Electrical Property Characterization Electrical property characterization includes f our-point probe measurement to acquire the sheet resistance of the thin film and Hall measur ement to obtain carrier type, concentration and mobility information. Four-point probe The four-point probe contains four thin coll inearly placed tungsten wires probes which are made to contact the sample under test. In a configuration shown in Figure 3-3, a constant current I flows between the outer probes, and the voltage V is measured between the two inner probes, ideally without drawi ng any current. The equipment used was an Alessi four point probe. It has a probe head with tungsten carbide tips with a point radius of 0.002, a probe space of 0.05. Current supplier was a Crytronics model 120 current sour ce which can provide currents from 1A to 100mA. Voltage was measured with a Keithley model 181 nanovolt electrometer. Sheet resi stance with units of /square, can be obtained from the f our-point probe using the equation R V I s 4532 (3.1)

PAGE 44

44 where V and I are voltage and current respectively and 4.532 is the geometric factor when the size of the thin film is much larger than the probe spacings. Resistivity with units of cm can be obtained by multiplying the thickness of the film by Rs, shown in the following equation I Vt tRs532.4 (3.2) Hall measurement Hall measurement is used to obtain the resist ivity, carrier mobility and concentration information of zinc oxide thin film. The physical theory behind this characterization technique is the Hall effect. When electric current flows in the presence of a perpendicular applied magnetic field, the Lorentz force deflects the moving char ge carriers to one side of the sample and generates an electric field perpe ndicular to both the current and th e applied magnetic field. This is called the Hall effect. The Hall co efficient is the ratio of the perp endicular electric field to the product of current density and magnetic field, whil e the resistivity is the ratio of the parallel electric field to the current density. A Lakeshore 7507 Hall system was used to c onduct the Hall measurement. The samples were cleaved into squares with dimension 1cm 1cm. They are mounted on Lakeshore Hall sample mount (Part numbe r 671-201) using double stick tape. Care was taken to make sure the sample surface is in parallel to the sample surface. Lakeshore Hall sample wires (Part number 671-260) are used for connection be tween the contact spots on the sample mount and the corners of the sample. The wires are so ldered with small amount of indium which forms Ohmic contact with the zinc oxide thin film. Square shaped van der Pauw geometry is used which is shown in Figure 3-4. In this geometry, the contacts are on the circumference of the sample and the dimension of the contacts

PAGE 45

45 is made as small as possible. For the square van der Pauw geometry, the estimated correction factor for resistivity can be s hown in the following equation[57] 2)( a c (3.3) where c is the dimension of the contact spot and a is the length of the Hall sample. Efforts were made to ensure the dimension of the cont act spot as small as 1 mm. Therefore, a good accuracy of the resistivity can be achieved. Before conducting the variable filed Hall meas urement, a forward and reverse I-V curve measurement is always conducted to verify the Oh mic property of all of the four contacts spots. Optical Spectrophotometer Characterization The optical transmission data was measured from the UV region to near IR region using a Perkin-Elmer Lambda 900 spectrophotometer. The system is equipped with a monochromator and a chopper, which is used to generate a sample and a reference beam. For all the measurements, the reference beam path was le ft open. Background corre ction leaving both the reference and sample beam path open is always performed before taking the transmission data. The samples are cleaned with acetone to remove possible contaminations such as fingerprints and dust particles, which may cause s cattering of the incident light. All transmission data was taken between 300nm and 900nm us ing the normal incidence setup. When an incident light with intensity I0 enters the sample, the absorption process can be expressed as d AeIeII 0 0 (3.4) where I is the remaining inte nsity and A is the absorbance, which is the product of the absorption coefficient and the sample thickness d. The transmission T is defined as

PAGE 46

46 d Aee I I T0 (3.5) The absorption coefficient can be estimated by measuring the transmission of two samples with different thickness d1 and d2. Then the absorption coefficient is derived using the equation )/()ln(12 2 1dd T T (3.6) where T1 and T2 are the transmission of sample 1 and sample 2 respectively. Other Characterizations Profilometry, X-ray diffraction, atomic fo rce microscopy and Auge r electron spectroscopy are used to characterize various material properties for sputte red zinc oxide thin films. Profilometry Because sputtering is a reproducible process, the growth rate can generally be maintained at a repeatable rate by using identical s puttering conditions. Therefore, there is no in-situ film thickness monitor set up in the UF zinc oxide sputtering system. The film thickness is determined using profilometer after the deposition and the growth rate is calculated. With growth rate obtained from prev ious runs, the film thickness can be controlled by choosing appropriate deposition duration. A Tencor P2-long scan profilometer was used to measure the thickness. Profilometer is a measurement device that probes the sample surface by a sweeping stylus. The vertical movement of the stylus is record ed when it runs over the sample surface. This gives the height of the surface as a function of location. To measure the thickness of the deposited films, a trench step needs to be created on the surface of the sample. Generally, two methods were utilized to create the step. The first method

PAGE 47

47 is to cover part of the substrate before deposi tion and remove the cover later. A portion of the substrate was either masked with permanent mark er or covered with a scotch tape. After the deposition, the marker was removed with acetone and a swab or the tape was simply removed by hand, which effectively removed the deposited film on top, yielding a well defined trench step. The second method to create the step is etching. The film area to be preserved are covered with tape and the unwanted area are expos ed on top of a beaker containing aqua regia. The solution is heated to 50C and the vapor etches away the oxide crea ting the step. The profilometer was then used to measure the height of the st ep, and therefore the film thickness. X-Ray Diffraction X-ray is the generic te rm used to describe electroma gnetic radiation with wavelengths between roughly 0.01 nm and 10 nm. This is sh orter than ultraviolet but longer than gamma radiation. The spacing of adjacent atoms in a crys tal structure is typically about 0.3 nm. The wavelengths of X-rays are theref ore of the same order of magnitude as the atomic spacing in a crystal. X-rays are generated by acceler ating electrons toward an anode target under vacuum conditions. The high-energy incide nt electrons interact with the target atoms and knock out some of the core electrons. Then the outer shell electrons fill in the holes left through relaxation process and the energy difference is released ge nerating X-ray to conser ve the energy. The energy or wavelength of the ra diated X-ray is determined by the energy difference between the two electrons and is not related with the acceleration energy of th e incident electrons. The Cu Kradiations are widely used in X-ray diffractomet ers. It is generate d via transition between the Cu L shell electrons and K shell electrons. The intensity ratio between K1and K2 radiation is 2:1. Figure 3-5 shows the schematic how Cu K radiations are generated. Other

PAGE 48

48 series of X-rays such as K and L series are generated at the same time. But they are filtered away and only the Kand Kare used to generate the diffrac tion pattern. The wavelengths of these two radiations are ve ry close and the peaks are overl apped especially at the low 2 angle. There are X-ray pattern processing software having the capabilities of stripping off the Cu Kdiffraction pattern allowing more accurate evaluation of the data. X-Ray Diffraction (XRD) can generally be cate gorized as single crys tal XRD and powder XRD. Single crystal XRD examines a single crysta l whose unit cells are perfectly aligned. It requires careful orientation of the sample in or der to obtain the diffraction pattern. X-ray powder diffraction is more widely used and the genera l term X-ray diffraction refers to the powder diffraction unless specified otherwise. The samples used for powder diffraction can be either powders or thin film samples. The diffraction geometry tells that only the crystallites with planes in parallel to the specime n surface will contribute to the diffraction intensity. An ideal powder sample consists of a co llection of small crystallites randomly distributed with all available orientations contribu ting to the diffraction pattern. The diffraction pattern is a fingerprint of a specific crystalline substance ( phase) and the Internationa l Center of Diffraction Data (ICDD, formerly known as JCPDS) maintain s a diffraction pattern database of various minerals, inorganic and organic standards. By comparing the diffraction pattern of an unknown sample with the data of standards, the po ssible phases available in the unknown may be identified. It is noteworthy that XRD is a technique for phase identification. In most cases, it is still necessary to know the chemical compositions of the sample to accurately identify the phases in exist. For thin film samples, the peak intensity can vary significantly from the standards due to preferred orientation. Be sides phase identification, careful examination of the peak broadening can be used to estima te grain size and stress associated with the sample. Finally,

PAGE 49

49 quantitive analysis of the diffraction pattern an be used to estimate the ratio between different phases in a specimen. The Braggs law is the cornerstone of X -ray diffraction analysis. The relation was discovered in 1912 by W. L. Bragg. Bragg law describes the condition that the scattered rays from two parallel planes interact with each other in such a way as to create constructive interference. As shown in Figure 3-6, the extra distance that ray B must travel is the distance X-Y-Z. Thus X-Y-Z = n to create constructive interference. Notice that X-Y = Y-Z = d sin. Hence, the Braggs law can be expressed with the equation = 2dhkl sin hkl (3.7) where 2 is the diffraction angle, which is the a ngle between the incident and diffracted x rays. In a typical experiment, the diffracted intensity is measured as a function of 2 and the orientation of the specimen, whic h yields the diffraction pattern. Atomic Force Microscopy Atomic Force Microscopy (AFM) and S canning Tunneling Microscopy (STM) are two forms of Scanning Probe Microsc opy (SPM), a technique that can reveal information down to atomic scale by using a sharp probe to scan across the surface of a sample. It is relatively new techniques developed in 1980s. Depending on the different operational conditions of the probe during scanning, there are three operation modes for AFM, namely cont act mode, tapping mode and non-contact mode. During contact mode measurement, a tip mounted at the end of a cantilever scans across the sample while contacting the surface. The deflecti on of the cantilever is monitored using a laser shining on the tip and redirect the laser to a sp lit photodiode, which measures the signal variation

PAGE 50

50 coming from the vertical movement of the can tilever. A feedback control loop maintains a constant deflection of the cantile ver by adjusting the vertical position of either the piezoelectric scanner or the sample. This position change is monitored and converted to the surface morphology of the sample. In tapping mode, the can tilever oscillates at or slightly below its resonance frequency at a position ve ry close to the sample surface. Instead of direct contacting the sample surface, the tip only taps on the sample surface. The root mean square (RMS) of the oscillation signal is maintained by the feedback cont rol loop by adjusting the positions of the scanner of the sample and the position change is converted to morphology. For non-contact mode, the tip oscillates close to the surface but never touches the sample surface. The oscillation amplitude and frequency will be changed by the van der Waals force, which extends a few nanometers above the sample surface. The feedback control loop maintains a constant oscillation amplitude and frequency by vertically moving the sca nners and this movement is converted to surface morphology. Contact mode AFM has high scan speed and good resolution, but the lateral forces exert on the sample surface may damage soft samples. N on-contact mode AFM exer ts little force on the sample surface and is not likely to scrap the samp le, but the scan speed is low and the sample should be extremely hydrophobic to prevent the ti p from getting stuck by the adsorbed fluid layer on the sample surface. Tapping mode AF M has less surface damage than contract mode AFM and can be used on most samples compari ng to non-contact mode AFM. The AFM images shown in this dissertation were obtained using a Veeco Dimension 3100 AFM operated under tapping mode AFM. The tips used are tapping mode etched single cr ystal silicon probes produced by Veeco.

PAGE 51

51 Auger Electron Spectroscopy Auger electron spectrosc opy (AES) is a technique that can identify chemical composition information by analyzing the characteristic energy of Auger electrons from an element. Auger electron spectroscopy has been primarily used to determine the chemical compositions of materials, but it is being increasingly used to ch aracterize chemical state properties. [58] It is applied to sputtered ZnO thin film characterizat ion to evaluate the composition of the film and detect any possible contaminations. When a core level electron is removed leav ing a vacancy, electrons from higher energy level can relax and fall into the left vacancy. The energy difference between these two energy levels is released to balance the total energy. This released energy can be transferred to a third electron, which is excited and es cape from the atom. The third el ectron is called Auger electron. Auger electrons have characteri stic energy relating to the en ergy difference between involved energy levels. Three electrons ar e involved to generate an Auger electron. Therefore, AES can only analyze materials whose atom s have at least three electrons Hydrogen and Helium are not detectable using AES and Lithium is material having the least atomic number required for AES analysis. AES is a surface sensitive analytical techni que because Auger electrons have low energy and they can only escape from areas a few atomic layers under the surface. Nevertheless, combined with ion-beam sputtering, which remove s material from the surf ace, this technique can provide depth information. Depth profile data was taken by alternating between sputter etching the sample and collecting small windows of Auger spectra focused on the peaks for elements of interest. A PHI model 660 Scanning Auger Microprobe (SAM ) was used in this work. It has a LaB6 filament as the thermionic emission electron source. A cylindrical mirror analyzer (CMA)

PAGE 52

52 is used to analyze the kinetic energy of the secondary and Auger electrons. The system is equipped with an ion pump and normally ope rates at a base pressure of about 3-10 Torr. An Ar ion gun is equipped to sputter off material from the samples fo r the purpose of either cleaning the sample surface or depth profiling. During th e sputtering process, the system typically works under a pressure in the range of middle 10-8 Torr.

PAGE 53

53 Figure 3-1. Perkin-Elmer 4400 sputtering system. Substrate Pallet ZnO Zn Mo Matching Network RF Power Generator Motor Throttle Valve Cryopump Stepper Drive Magnetron 8 inch 5-9 cm MFC MFC Ar O2 Figure 3-2. Schematic of the ZnO Sputtering System (Perkin-Elmer 4400).

PAGE 54

54 w L t I V Figure 3-3. Schematics of f our-point probe measurement. Figure 3-4. Square shaped van der Pauw sample geometry.

PAGE 55

55 Figure 3-5. Generation of Cu K radiation. Figure 3-6. Braggs condition.

PAGE 56

56 CHAPTER 4 GROWTH AND CHARACTERIZATION OF ZINC OXIDE Optimization of Sputter Deposi tion of Aluminum Doped ZnO The optical and electrical pr operties of Al doped ZnO (AZO) can be affected by various deposition parameters such as base pressure, working pressure and deposition power. Films obtained under different depositi on conditions are routinely charac terized with f our-point probe and profilometry to obtain the resistivity valu e. Various other techniques such as X-ray diffraction, atomic force micr oscope, Hall measurement and Auger electron spectroscopy are employed to further characterize these thin films. The results will be presented in this chapter. Experiments at Different Base Pressures Due to the large surface area of the substrate platen, it re quires a long period of time for outgassing. The starting base pr essure can vary over a large range depending on the pumping time. An investigation which extend s the base pressure between 6.0-7 Torr and 2-6 Torr is carried out. The base pressures are categorized into three regions to better illustrate the obtained results, namely the low, medium and high base pressure region. The low base pressures represent approximately 6-7 Torr, obtained by adopting a very long period (typically greater than 24 hours) of pumping of the sputtering chambe r before starting the deposition. This ensures that a minimum amount of residual water vapor is present during the sput tering process. The medium base pressure, approximately 9-7 Torr, is associated with a shorter pumping period, under conditions that the chambers pressure sti ll continues to drop when the deposition process starts. In this case, the pumping and outgassing have not reached equilibr ium at the beginning of the deposition, and as a consequence the amount of water vapor present in the chamber is larger than in the low base-pressure region. Finally, the high base-pressure condition, approximately 2-6 Torr, is obtained after a short pumping period, and as a consequence the amount of

PAGE 57

57 residual water vapor is relatively large. All of these three base pressure regions ensure a reasonable clean environment for the sputter-deposition process. Optical property The optical transmission of AZO films deposit ed under the three base-pressure regions is measured using Perkin-Elmer Lambda900 sp ectrophotometer over the wavelength between 300nm and 900nm, the spectrum region where solar li ght has most of its intensity. Figure 4-1 shows plots of the optical tran smission of the resulting films as a function of incident-light wavelength for the three pressure-regimes invest igated. The results clearly demonstrate that increasing the base pressure improves the optical transmission. For the film deposited under low base pressure, the average optical transmi ssion between wavelength 300 nm and 900 nm is below 70%. The two films deposited under medium and high base pressure show better optical property and meet this requirement Generally, for solar cell a pplications, an average optical transmission of 80% is desire d for the visible and near in frared wavelength region. It is known that the optical transmission and c onductivity of zinc oxide is directly related to the stoichiometry, i.e., the ratio between zinc and oxygen cont ent. Higher concentrations of oxygen vacancies or zinc in terstitials in turn provi de higher concentrations of intrinsic carriers and result in a more conductive film. However, under such circumstances the optical transmission would be expected to drop. This is exactly what reflected in Figure 4-1, where the optical transmission indeed increases when the base-pressure favors the presence of water vapor in the chamber. Under the plasma environmen t the residual water vapo r dissociates and the active oxygen species from the water vapor cont ributes the incorporation of oxygen in the sputtered AZO films. This explains why the opti cal transmission increases with an increasing of the base pressure.

PAGE 58

58 Electrical property Figure 4-2 plots the effect of each of the thr ee base-pressure regimes on the resistivity of the sputtered AZO thin films. An arbitrary continuous line is drawn between the experimental points simply to suggest a trend a nd to aid in the discussion of the results. First consider the results obtained at the medium and high base-pre ssure regions. Note that the resistivity decreases as the base pressure is reduced from the high to the medium region. This result is expected because under medium base-pressure c onditions the AZO films are more zinc-rich (i.e., oxygen deficient) because less water vapor is present, leading the more conductive film. Next consider the results obt ained at the low and medium base-pressure regions. In contrast to the previous case, the resistivity now increases as the base pressure is reduced from the medium to the low region. Obviously, in this case a different mechanism must be in playing a significant role to a point where it causes a reversal in the resist ivity trend. The explanation for this observation is based on the fact that less re sidual water vapor is present at the low base pressure, promoting the presence of insufficien t amounts of hydrogen. The observation of this resistivity trend reversal lends evidence indicatin g that hydrogen plays a critical role in the resistivity of sputtered AZO thin films. Details of hydrogen effect on sputtered AZO will be addressed in the next chapter. Growth rate The effect of base pressure on growth rate was investigated. A series of films were deposited under all same deposit ion conditions except starting from different base pressures, which are 8-7 Torr, 1-6 Torr, 2-6 Torr, and 3-6 Torr respectively. From the results shown in Figure 4-3, an increase of the base pressure from 8.0-7 Torr to 3-6 Torr caused a growth rate decrease from 3.4 /s econd to 2 /second. This effect is also due to the existence of water vapor. The oxygen generated in the highly energetic plasma process not only reacts with

PAGE 59

59 the deposited film, it also reacts with the target and increases the oxidation extent of the target. It is known that the sputtering rate for an oxide target is generally much lower th an that of a metal target under the same sputtering conditions. Thus, th e growth rate from a more oxidized target is expected to be smaller than that from a more re duced target, when all other sputtering conditions are held constantly. Power Effect The effect of deposition power on the resi stivity of sputtered ZnO:Al films was investigated by holding all othe r deposition parameters as c onstant except the power. All other deposition conditions are fi xed with target to substrate di stance of 5 cm, base pressure of 8.010-7 Torr, and working gas pressure 4 mTorr. The results show that the power can directly a ffect the resistivity. As shown in Figure 4-4, it is clear that with all other deposition parameters held as constant, the film deposited at 400W shows lower resistivity than those ob tained using other values of power. The relation between growth rate and deposition was also evalua ted and the result is shown in Figure 4-5. It can be seen that the trend is increasing the depositi on power causes the increase of the film growth rate. This is caused by an increase of the sputter yield upon increasing the deposition power. Discharge power is a critical factor that determ ines the properties of sputtered ZnO. When sputtering from a compound target, the more volatile component can be more rapidly lost into the gas phase without being condensed on the substrate. This may result a different stoichiometry of the film from th at of the target. For the case of ZnO sputtering, oxygen is more volatile than zinc. With a careful adjustment of the deposition power, a desired Zn rich stoichiometry can be achieved.

PAGE 60

60 Through experiments, it is observed that the deposition parameters, especially the power, can affect the surface stoichiometry of the ZnO target. For magnetron sputtering of ZnO, when large power density (greater than 2 Watt/cm2) is exerted on the target for a long period of time, the target surface could be significantly reduced. Thin films sputtered from a heavily reduced target can contain strong separated Zn phase. This conclusion is obtained by employing X-ray diffraction characterization on sputtered thin film from a reduced target. The equipment used was a Philips APD 3720 powder diffractometer using Cu K radiation with (K equals 0.154056 nm and (K equals 0.154439 nm. The X-ray generation conditions on the diffractometer were set at 40KV and 20mA. For survey scanning, co ntinuous scans with a 2 range between 10 to 80, a step size of 0.03, and a step time of 0.5 second was used. Figure 4-6 shows the XRD survey pattern of a thin film sputtered from a heavily reduced target The vertical lines indicate the 2 theta angle and relative intensity of Zn taken from JCPDS database. It is evident that there is Zn phase available in the resulted thin film. In contrast, Figure 4-7 shows the XRD survey pattern of a typical ZnO thin film sputtered from an oxidize d target. The pattern shows a strong ZnO (002) peak indicating the sputtered film has preferred or ientation. These result s show evidently that the target condition is important determining the property of sputtered ZnO thin film. A control of deposition power with in certain limit is required to maintain the oxidization status of the ZnO target. High resolution continuous scans were also implemented using the following parameters: 2 ranging from 33.5 to 35.5; step size of 0.01 ; and 1.0 second per step. Figure 4-8 shows the high resolution scan pattern of three AZO th in films deposited at 300 W, 400 W and 500 W,

PAGE 61

61 respectively. The Full Width at Half Maximum (FWHM) values were obtained from the high resolution scan and their values are list in table 4-1. Table 4-1 also contains grain size information. The mean grain size was evaluated using Scherrer equation [59] : cosK (4-1) where K is a constant called shape factor, whose typical value is 0.9[60]. The term represents the line broadening due to the effect of small crystallites. Note it is the additional broadening in diffraction peaks beyo nd the inherent peak widths due to the instrument effect. The instrument broadening contribution can be obtained by measuring a standard material that has no significant stress and grain size effect. Here for approximation, only the measured FWHM in radians was used. The other two terms and are the wavelength of the X-ray and the diffraction angle. With this simplification, the estimated grain sizes are 36.3, 32.3, and 30.7 nm for films deposited using 300W, 400W and 500W, respectively. Working Pressure Effect Similarly, the relationship between working pres sure and film resistivity is investigated by changing the working pressure while fixing all other deposition parameters. Experiments are done under four working pressure evenly distribu ted between 2 mTorr and 5 mTorr. The results are shown in Figure 4-9. It is clear that 4 mTorr working pr essure gives the lowest film resistivity for this specific experiment set. The film growth rate is also dependent on work ing pressure. As shown in Figure 4-10, the general trend is increasing the wo rking pressure would lead to an increase of the film growth rate. This effect is especially strong at the relatively low work ing pressure region with pressure

PAGE 62

62 changing from 2 mTorr to 3 mTorr. The curve st art to flat out at high pressure region and only a slight increment was observed when increasing the pressure from 3 mTorr to 5 mTorr. High resolution XRD scans of films deposited with different working pressures are shown in Figure 4-11. The three working pressures used are 2 mTorr, 3 mTorr and 4 mTorr. All other deposition conditions are fixed with target to substrate distance of 5 cm, base pressure of 4.0107 Torr, and deposition power of 500W. It is f ound that the measured FW HM of the three XRD rocking curves decreases with the increased work ing pressure, indicating la rger grain size can be achieved by using a higher working pressure. Characterization Besides four-point-probe, profilometry and X-ray diffraction, ot her characterization techniques were used in analyz ing sputtered ZnO thin films. Auger electron spectroscopy was applied for chemical composition and depth profile analysis. Atomic force microscopy was used to evaluate surface morphology properties. Auger Electron Spectroscopy Analysis Auger electron spectroscopy was utilized to verify the chemical composition of the deposited films and to detect any unexpected impu rities that may exist in the deposited films. Results reveal that the films contain only Zn, O and Al in the bulk, with C contamination only in the surface. The source of the carbon is organi c contaminations residing on the surface of the sample. Depth profile analysis shows a uniform distribution of Zn and O through the film, indicating good stoich iometry control during the deposition. Auger electron spectroscopy is a surface sensitive analytical technique because the Auger electron, generated with a small energy, can only escape from a few atomic layers on the top. Like SEM, Auger electron spectroscopy uses a el ectron beam hitting the sample to generate secondary and Auger electrons. The resulted secondary elec trons can be collected and

PAGE 63

63 transferred into secondary electron images, wh ich is just like what a scanning electron microscopy does. Beyond that, the Auger electrons are analyzed by a cylindrical mirror analyzer (CMA) and transferred to information re garding the chemical composition. The distance between the CMA and the sample surface was always calibrated using the elastic scattering peak from a 3K eV electron beam before taking the survey spectrum. Typical energy of the electron beam used for survey spectrum was varied between 5KeV to 10KeV. The spectrum of counted electrons at an energy (n(E)) versus kinetic energy (KE) were taken with an energy step size of 1.0 eV, and a dwell time of 30ms at each ener gy step. Ten sweeps of this spectrum were taken over the energy range of 50 to 2050 eV and the average value was used for numerical processing. This spectrum is then num erically differentiated using algorithm from the computer acquisition software (AugerScan) to yield the spectrum in terms of d(n(E))/d(E) versus kinetic energy. The differentiated spectrum is then used to compare standard results and consequently determine the elements present on the samples surface. Surface scan Figure 4-12 shows a survey spectrum of an AZO thin film. The differentiated energy distribution d(n(E))/d(E) was plotted against the ki netic energy. The key instrument parameters used for this survey were as follows: e-beam acceleration voltage 5kV, probe current 50nA, and sample tilting 30 degrees. The Auger peaks of C, O, and Zn are identified and labeled in the figure. The result verifies the film composition as Zn, O and it also suggests that there is carbon contamination at the surface of the ZnO fi lm. The source of carbon is from organic contaminations on the sample surface. These contaminations can be reduced by careful handle and storage of the sample afte r it is taken out of vacuum. Al peak was not distinguishable from Figure 412. Various factors ar e responsible for it. First of all, the noise signal level is quite signi ficant due to the surface contamination. Secondly,

PAGE 64

64 Al only constitutes less than 2% of the target, th erefore, it is estimated the film would only contain a relatively small amount of Al. The thir d reason is Al has a sm all relative sensitivity factor Sx (~0.09, 10keV) compared with Zn (~0.16, 10keV) and O (~0.4, 10keV). Ar ion beam was used to sputter the sample surface for 3 minutes to remove the surface contamination. Another survey spectrum was taken using identical su rvey conditions and the result is shown in Figure 4-13. Compared with result shown in Figrue 4-12, it is clear that the carbon peak is no longer present. This verifies the previous assumption that carbon is only a surface contamination and not part of the film constitution. Ar peak is detected indicating implanted Ar from the ion beam exist on the surf ace of the sample. Co mparing with previous results, the noise signal in the second survey (sho wn in Figure 4-13) is much smaller than that in the previous survey (shown in Figure 4-12). Cons idering the fact that tw o scans used identical survey parameters, the noise reduction wa s achieved through the removal of surface contaminations. Semi-quantitive data of the co ncentration of each species detected are also labeled in the diagram. Al sensitivity Because of the reasons presented earlier, the al uminum peak was not clear in either of the previous cases. To identify the aluminum peak and verify its existence, another scan was conducted using higher e-beam energy. Increa sing the acceleration vol tage to 10kV, the new survey result is shown in Figure 4-14. In this spectrum, Aluminum Auger peak at approximately 1375 eV kinetic energy was clearl y distinguished. The increa sed beam energy caused more excitation of Al Auger electrons, which increased the size of the peak, making it distinguishable from the noise signal.

PAGE 65

65 Depth profile Depth profile was measured to examine the distribution of those main elements along the depth of the sample. Depth profile analysis can be performed us ing the window method by alternatively sputter etching the sample and co llecting small energy windows of Auger spectrum focused on the peaks for elements of interest. The energy window is defined using the 3-point method, which uses two background energy position and one peak energy position to define the area of the peak. The system uses the diffe rence between the peak value and the background values to evaluate the intensity of the peak. Th is intensity is then plotted as a function of the sputter etching time, giving a dist ribution of the element distributi on along the depth direction. Parameters used for collecting de pth profiles are presented in Tabl e 4-2. The sputtering rate can be controlled by adjusting the ion beam paramete rs. With the Ar ion beam energy of 3KeV, beam current of 90 nA and raster pattern of 3x3 mm, the sputter etch rate is estimated to be approximately 100/min. Figure 4-15 shows the Auger depth profile of a ZnO thin film. 30 sputtering cycles was used in the particular measurement and during each sputtering cycle, the sample is sputtered for 6 seconds. Therefore, the estimated s putter etch rate is approximately 10/cycle and a total depth of approximately 300 into the bulk of the film was characterized. The results show that carbon contamination is highest on the surface, and th at it stablizes at nosie level at the position approximately 100 below the surface. The depth prof ile analysis also shows that the concentration of Zn and O inside the bulk of the f ilm is uniform. Atomic Force Microscopy Atomic force microscopy was applied to ex amine the surface morphology of sputtered AZO thin films. A Veeco DI Dimension 3100 syst em was utilized for this purpose. The surface

PAGE 66

66 of the sample was carefully cleaned using acetone to remove attached contaminations before the examination. The AFM image was digitally processed before any analysis is performed to eliminate the effects coming from artifacts such as tilt and bow. Tilt happens when the sample is not perfectly leveled and bow is generally caus ed by large area scans using a large scanner. These unwanted effects need to be eliminated before su rface roughness analysis is performed. A Flatten command from the diNanoScope Soft ware 7.0 was first executed to eliminate unwanted features from scan lines. A third order flatten is used. The software calculates a third order least-squares fit th en subtracting it from the scan line This effectively removes the Z offset between scan lines and also eliminates sample tilt and bow from each scan line. After flattening, Plane Fit command which computes a polynomial of a selectable order for an image and subtracts it from the image was used. A third order Plane Fit command was applied on both X and Y direction. Arithmetic roughness Ra and root mean square roughness Rq values are then obtained from the roughness analysis. Ra is the arithmetic average of the absolute value of the surface height deviation and Rq is the standard deviation of the surface height value. They can be expressed using the following equations: N zz Ri a || (4-1) N zz Ri q 2)( (4-2) where zi is the height value of a specific posi tion and N is the tota l number of positions evaluated. Figure 4-16 shows the AFM micrograph of AZO thin films deposited using identical deposition parameter except the working pressure Sample a,b, and c are grown are Corning

PAGE 67

67 1417 glass substrates under worki ng pressure 2mTorr, 3mTorr and 4mTorr, respectively. Care was taken to ensure the samples have similar thickness to eliminate the thickness effect on surface morphology. It is found that with the increase of the work ing pressure, larger grain size is found on the sample surface. Generally, larger grain size is favorable for polycrystalline AZO because it can be beneficial to reduce grain boundary effect and therefore increase the electron mobility, which then improves the conductivity of the film. The surface roughness also increases along with increasing the working pressure. With a working pre ssure change from 2mTorr to 4mTorr, the root mean square roughness increas ed from 2.2nm to 3.6nm. Details of the roughness data change with the working pr essure is shown in Table 4-3. The AFM analysis shows that 4mTorr is a favora ble working pressure fo r larger grain sizes. Since larger grain size tends to yield a better conductivity, this is consistent with the resistivity characterization presented earlier, which inde ed showed that AZO sputtered using 4mTorr working pressure ha s lower resistivity.

PAGE 68

68 0 20 40 60 80 100 300400500600700800900 Wavelength (nm)Transmission ( %) Low base pressure Medium base pressure High base pressure Figure 4-1. The effect of base pressure on the optical transmission of sputtered AZO films. 0 2 4 6 8 10 LowMediumHigh Base pressure Resistivity (10 -3 cm) Figure 4-2. Effect of base pressure on the resistivity of sputtered AZO films.

PAGE 69

69 1.5 2 2.5 3 3.5 4 5.0010.0015.0020.0025.0030.0035.00 Base pressure (10-7Torr)Growth rate (/s ) Figure 4-3. The effect of base pressure on the growth rate of ZnO:Al thin films (Fixed conditions: Dts = 5 cm, P = 4 mTorr, Power = 400W). 1.0 1.1 1.2 1.3 1.4 1.5 300400500 Power (W)Resistivity (10 -3 cm) Figure 4-4. The effect of power on the re sistivity of as grown ZnO:Al thin films (Fixed conditions: Dts = 5 cm, Pb = 8.010-7 Torr, and P = 4 mTorr).

PAGE 70

70 0 1 2 3 4 5 300 400 500 Power (W)Growth rate (/s) Figure 4-5. The effect of power on th e growth rate of ZnO:Al thin films (Fixed conditions: Dts = 5 cm, Pb = 8.010-7 Torr, and P = 4 mTorr). Figure 4-6. X-ray diffraction pa ttern of thin films deposited from a reduced ZnO target

PAGE 71

71 Figure 4-7. X-ray diffraction patte rn of thin films deposited from an oxidized ZnO target 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 33.5 34.0 34.5 35.0 35.52Intensit y 300W 400W 500W Figure 4-8. High resolution XRD pattern for AZO films deposited using different power (Fixed conditions: Dts = 5 cm, Pb=8.010-7 Torr, Pressure = 4mTorr).

PAGE 72

72 1.0 1.1 1.2 1.3 1.4 2345 Working pressure (mTorr)Resistivity (10 -3 cm) Figure 4-9. The effect of working pressure on the resist ivity of ZnO:Al thin films ( Fixed conditions: Dts = 5 cm, Pb = 8.010-7 Torr, and Power = 500 W ) 0 1 2 3 4 5 6 2345 Working pressure (mTorr)Growth rate (/s ) Figure 4-10. The effect of working power on the growth ra te of ZnO:Al thin films ( Fixed conditions: Dts = 5 cm, Pb8.010-7 Torr, and Power = 500 W )

PAGE 73

73 0 2000 4000 6000 8000 10000 12000 14000 16000 33.5 34.0 34.5 35.0 35.52 Intensit y 2 mTorr 3 mTorr 4 mTorr Figure 4-11. High resolution XRD pattern of AZ O thin films deposited at different working pressures (Fixed conditions: Dts = 5 cm, Pb=4.010-7 Torr, Power = 500 W). Figure 4-12. AES survey scan pattern for an AZ O thin film with organic contamination on the surface.

PAGE 74

74 Figure 4-13. AES scan results after sputtering for 3 minutes. Figure 4-14. Al Auger peak shown by using increased primary beam energy

PAGE 75

75 Figure 4-15. Depth profile for each element of AZO obtained with AES.

PAGE 76

76 a) b) c) Figure 4-16. AFM graph of AZO thin films de posited with different wo rking pressure. a) 2mTorr, Rq=2.2nm; b) 3mTorr, Rq=2.8nm; c) 4mTorr, Rq=3.6nm.

PAGE 77

77 Table 4-1. Estimated grain size using XRD data obtained for AZO deposite d at different power. Deposition power FWHM Grain size 300 W 0.229 36.3 nm 400 W 0.257 32.3 nm 500 W 0.270 30.7 nm Table 4-2. Parameters used for collecting the Auger depth profile of AZO thin films. Peaks of interest Peak energy (eV) Background E1 (eV) Background E2 (eV) Dwell time (mS) Number of sweeps O1 506.0 523.0 479.0 250 2 Zn1 985.0 1017.0 975 250 2 C1 264.0 298.0 200.0 250 2 Table 4-3. AFM surface roughness e volution with working pressure. Working pressure (mTorr) Rq (nm) Ra(nm) 2 2.2 1.7 3 2.8 2.3 4 3.6 2.9

PAGE 78

78 CHAPTER 5 EFFECT OF HYDROGEN ON SPUTTERED ZINC OXIDE Motivation from the Base Pressure Effect The observation of base pressures effect on AZO resistivity indicates that residue water vapor plays a critical role in de termining the resistivity of sputtered zinc oxide thin film. Under the plasma environment during sputter depositio n, the water vapor d ecomposes into hydrogen and oxygen. With the base pressure variation in the order of 10-7 Torr, the change of oxygen coming from H2O is quite small and would not affect the amount of oxygen in the resulted film in dramatic fashion. Therefore, it is highly suspected that hydrogen, the other ingredient from H2O plays certain roles in sputtered ZnO. In most semiconductors interst itial hydrogen presents itself as an amphoteric impurity [61], and acts in a fashion that count eracts the prevailing conductivity. This behavior makes hydrogen an unusual candidate for semiconductor doping. Ho wever, in the case of zinc oxide, Van de Walle [62] predicted that inte rstitial hydrogen can provide a donor level ju st below the conduction band and that therefor e it could contribute to enhan ce n-type conductivity in this material. Experimentally, hydrogen implantation experiments showed the change of carrier concentration and mobility in ZnO films [63]. However, hydrogen doping of ZnO needs to be experimentally implemented in our Perkin-Elmer sputtering system to see if a more conductive film can be obtained. Experiments and Results To directly investigate the effect of hydrogen on the electrical properties of sputtered AZO thin films, two groups of AZO films were deposited on soda-lime glass using identical conditions, except the composition of the sputtering gas. One group was deposited using UHP argon only, and the second group using a gas mixtur e of argon and 0.1% hydrogen gas. Special

PAGE 79

79 care was taken trying to maintain an identical deposition conditions. This includes identical sputter parameter, deposition duration, substrate position, etc. The re levant experimental conditions are given in Ta ble 5-1. The soda-lime glass subs trates were thor oughly cleaned using the method described in Chapter 3 befo re depositing with the AZO films. The resistivity obtained through four-point-probe and profilo metry measurement of thin films from the two groups is plotted in the gra ph presented in Figure 5-1. AZO stands for the thin films sputtered using pure argon and H-AZ O represents those grew with argon and hydrogen mixture. The data reported in Figure 5-1 clearly demonstrates that the use of an argon-hydrogen gas mixture significantly decreases the resistivity of the sputtered AZO thin film. This observation is consistent with the claims made previously rega rding the beneficial effect of the presence of hydrogen from the viewpoint of enhancement of the film conductivity. Hen ce, the data provides a strong experimental evidence to justify the claims of hydrogens beneficial effect on sputtered AZO thin films in terms of lowering the resistivity. Hall Characterization To obtain an in-depth unde rstanding of the mechanism on this improvement of conductivity, Hall measurements ar e done to evaluate the carrier concentration and mobility change between the AZO films deposited with pure Ar and Ar/H2 gas mixture. Special care was taken to make sure the samples were taken from areas with identical posit ions under the target. This ensures the elimination of possible eff ects coming from the lo cation difference. Figure 5-2 shows Hall measurement results obt ained for two AZO films. The sample denoted as AZO in the figure was deposited using only argon as the working gas, and the sample denoted as H-AZO was deposited using 0.1% hydrogen in the work ing gas. The results show that the carrier concentra tion increased from 2.51020cm-3 for the AZO film to 4.720cm-3 for

PAGE 80

80 the hydrogen doped H-AZO film while the hall mobility dropped from 15.1cm2/(Vs) to 12.7 cm2/(Vs). As a result, the resist ivity decreased from 1.67-3cm for the AZO sample to 1.05-3 cm for the H-AZO sample. The resistivity value is consistent with that obtained using four-point probe measurement. It is evident that hydrogen improved the conductivity through increasing the carrier concen tration inside the AZO film, indi cating that it either acts as a dopant or increases the doping e fficiency of aluminum for these sputtered ZnO thin films. AFM and SEM Characterization Atomic force microscopy was used to analyze the surface morphology of ZnO sputtered using different sputtering gas. A scan area of 11m was used. Like mentioned in Chapter 4, the Flatten and Plane Fit commands from the diNanoScope Software 7.0 were applied to eliminate unwanted features such as tilt and bow. The root mean square roughness Rq and the arithmetic roughness Ra of the sample surface were then eval uated. The definition of these two parameters can be found in Chapter 4. Figure 5-3 shows a comparison of surface mo rphology between AZO sputtered with Ar (referred to as AZO in the figure) and Ar+0.1%H2 (referred to as H-AZO in the figure). They are grown on soda-lime glass cleaned using the pr ocedure described in Chapter 3. Both films have similar thickness of 600nm and they were deposited using identical deposition conditions except with different sputtering gas. Details of the deposition conditi ons are listed in Table 5-2. The result shows that the AZO sputtered using Ar and H2 mixture shows a much larger grain size compared to the film sputte red using Ar only. The RMS roughness Rq and Ra of these two films are compared in Table 5-3. It can be seen that the root mean square (RMS) roughness Rq of H-AZO film almost doubled th at of the AZO film. This indi cates that a much larger grain size can be achieved through using the Ar and H2 mixture gas.

PAGE 81

81 The thickness effect on surface morphology of sputtered ZnO was also investigated. H-AZO films with different thic kness deposited using Ar and H2 mixture gas were characterized and the resulted AFM micrographs are shown in Figur e 5-4. It is clear that increasing the film thickness would tend to yield larger grain size. The Rq and Ra values of ZnO films with different thickness are shown in Table 5-4. The trend is that the surface roughne ss increases when the film gets thicker. Figure 5-5 shows a typical cross-section SE M image of a H-AZO th in film grown on iZnO substrate. It can be seen that the film contains large column shaped grains grown perpendicular to the substrate. Film Uniformity Study To investigate the uniformity related issue, aluminum doped zinc oxide thin films were deposited onto large size glass substrate covering areas up to 3 inches away from the center axis of the target. The deposition duration was 30 minutes. The thickness of the film was measured every 0. 5 inch along the radius and the results are shown in Figure 5-6. Sheet resistance was measur ed using four-point probe at identical positions where the thickness was measured and the result is shown in Figure 5-7. Resistivity was then calculated and its variati on with position is shown in Figure 58. In these figures, a position of 0 inch indicates the position directly below the ta rget center. The number in front of a position indicates the distance between this position and the 0 in ch center point. The positive or negative sign of this number indicates th e position locates at the right hand side of the center point or the left hand side, respectively. For example, a posi tion with a value of -1 inch denotes the location one inch away from the center point sitting on the left hand side. From Figure 5-6, it is clear that the film thic kness is the lowest at the center point and it increases with an extended distance from the cen ter point until it reaches its maximum value of

PAGE 82

82 about 600nm at the location approximately 2.5 inch away from the center point. After passing this maximum point, the thickness starts to decrease with increasing the distance from the center point. Figure 5-7 and Figure 5-8 indicate that the sheet resistance and film resistivity show similar distribution along the axial direction. The resistivity at the center point is the highest and it decreases along the direction away from the center. This trend persists until a minimum value is reached at the position approximately 2-2.5 inch away from the center point. After this point, the resistivity starts to increase again. To further investigate the m echanism behind the resistivity distribution, Hall measurement was utilized to reveal the carrier concentrati on and mobility variation on films with different positions under the target. Two samples were cleav ed off the large substrate from positions with the highest and lowest resistivity indicated by four-point-probe measurement results. The comparative results are shown in Table 5-5. It is found that the film de posited directly under the center of the target has a much lower electron mob ility and relatively higher carrier concentration comparing to that deposited 2.5 inch away from the center. As a result, the lower electron mobility directly leads to a higher resistivity. The lower electron mobility of the film deposited at the center point could be caused by the hea vy bombardment of the ions from the plasma during the sputtering deposition pro cess. It is noteworthy that for hall measurement, the measured carrier concentration is directly related to the van der Pauw sample thickness, which is a manual input before the measurement. Due to the nature of non-unifo rmly distributed film thickness along the Hall sample, it is difficult to use an accurate description of the Hall sample thickness. Thus, some errors could exist on the carrier concentration results.

PAGE 83

83 The resistivity obtained by using Hall and four-point-probe m easurements is compared in Table 5-6. It can be found that the resistivity values obtained from different methods fall in a reasonable close range. The difference between results obtained using these two measurement techniques could come from the uncertainty of the position where the Hall samples were cleaved from. Summary The experiments on base pressure effect on sp uttered AZO thin films are extended to very low base pressure region. The resistivity obtaine d at this region show different trend from those at higher region presented in Chapter 4. Results suggest that hydrogen emanating from residual water vapor affects in a favorable fashion the resi stivity. This supports Van de Walles theory on hydrogens doping effect on zinc oxide thin films. To direct investigate hydrogens effect, a different sputtering gas consisted of UHP argon and 0.1% hydrogen was utilized to produce AZO thin films. Properties of resulted AZO thin films were compared with control sample. Hall measurements show an increase of both carrier concentration and mobility, resulting a much more conductive thin film. This directly provides a second piece of evidence confirming the claimed role of hydrogen. The results are particularly relevant fr om practical viewpoint given that it is demonstrated that improved optic al and electrical properties ca n be obtained by the addition of hydrogen into the working gas and/or the selection of a relatively high base-pressure that permits the presence of residual water vapor in the ch amber. This identifies a path towards the development of a more consistent method for depositing good-quality transparent AZO films via magnetron sputter deposition method. Further wo rk needed to achieve this goal include experimental studies aimed at characterizing the optimal ra nges of hydrogen content in the working gas as well as the optimal range of ope rating base-pressures. This is a feasible goal

PAGE 84

84 given that the experimental ev idence provided here conclusively demonstrates that the two variables in question have a significant effect on the electrical and optical film properties. The film uniformity distribution was investig ated on AZO thin films sputtered using Ar and 0.1% H2 gas. Non-uniform distribution of the film thickness and resistivity was observed over a range covering up to 3 inch away from the center point under the target It is found that the center point has the lowest film thickness and highest resistivity. The difference between the highest and lowest resistivity is wi thin a factor of two for freshly deposited films. This result is helpful for determining the feasibility of large area AZO coating using this equipment for large size solar module applications.

PAGE 85

85 0 5 10 15 20 25 AZO H-AZOResistivity (10 -4 cm) Figure 5-1. Effect of injec ting hydrogen gas on the resistiv ity of sputtered AZO films. Figure 5-2. Hall measurement results for alumin um-zinc-oxide thin films deposited using only argon as the working gas (sample labeled as AZO) and using a mixture of argon with 0.1 wt% of hydrogen (sampled labeled H-AZO).

PAGE 86

86 a) b) Figure 5-3. AFM image of two AZO films deposite d using different sputte ring gas. a) AZO b) H-AZO

PAGE 87

87 a) b) c) d) Figure 5-4. AFM image of H-AZO thin films with increasing thickne ss. a)200nm b)400nm c)600nm d)800nm

PAGE 88

88 Figure 5-5. Cross-section SEM image of H-AZO thin films grown on i-ZnO substrate showing large column shaped grain structure.

PAGE 89

89 300 350 400 450 500 550 600 -2-101234 Position (inch)Thickness (nm ) Figure 5-6. Variation of AZO thickness with di fferent positions along th e axial direction. 0 10 20 30 40 50 -2-101234 Position (inch)Sheet Resistance ( / ) Figure 5-7. Variation of sheet resistance with different positio ns along the axial direction.

PAGE 90

90 0 5 10 15 20 -2-101234 Position (inch)Resistivity (10 -4 cm) Figure 5-8. Variation of resistivity with different positions along the axial direction.

PAGE 91

91 Table 5-1. Growth parameters comp arison for AZO and H-AZO thin films. Film composition Base pressure Working pressure Power density Sputtering gas AZO 1.010-6 Torr 4 mTorr 350W Ar only H-AZO 1.010-6 Torr 4 mTorr 350W Ar + 0.1% H2 Table 5-2. Deposition condition comparison between two AZO thin films shown in Figure 5-5. Film Power Working pressure Deposition duration AZO 350 W 4mTorr 30min H-AZO 350 W 4mTorr 30min Table 5-3. Thickness, Rq and estimated grai n size comparison between two AZO thin films shown in Figure 5-5. Film Thickness Rq Ra AZO ~600 nm 2.89 nm 2.34 nm H-AZO ~600 nm 5.85 nm 4.63 nm Table 5-4. Roughness data evolution with increasing film thickness. Thickness Rq Ra ~200 nm 2.38 nm 1.89 nm ~400 nm 4.15 nm 3.38 nm ~600 nm 5.85 nm 4.63 nm ~800 nm 8.18 nm 6.51 nm Table 5-5. Hall measurements results on two fi lms with different posit ions under the target. Position Carrier concentration Hall mobility Resistivity center 6.51020 (cm-3) 4.79cm2/(Vs) 2.0010-3cm 2.5 inch from the center 4.31020 (cm-3) 14.2cm2/(Vs) 1.0310-3cm Table 5-6. Comparison of Hall and f our-point-probe measurement results. Position Resistivity obtained from Hall measurement Resistivity obtained from four-point-probe measurement center 2.0010-3cm 1.8610-3cm 2.5 inch from the center 1.0310-3cm 9.7410-4cm

PAGE 92

92 CHAPTER 6 DEVICE FABRICATION AND CHARACTERIZATION Device Fabrication Procedure Substrate The fabrication of CIGS solar cells starts from choosing the right substrate. Soda-lime glass (SLG) is generally preferre d due to its chemical inactivity, thermal stability, and vacuum compatibility. Soda-lime glass contains 15.6 wt % of sodium in the form of Na2O. At the high processing temperatures used for absorber growth, soda-lime glass starts softening. The sodium ions can diffuse through the molybde num back contact entering the ab sorber layer. It is also found that an optimum amount of diffused sodium is beneficial to improve the performance of the devices [64]. It is notewo rthy that CIGS solar cells can al so be fabricated on flexible substrates such as metal foil and polymer substr ates. Sodium doping may be required to achieve high efficiency from devices fabricat ed on these sodium free substrates. Back Contact Sputtering Sputtering is generally used for coating the thin layer of molybdenum back contact. Molybdenum is chosen because it forms good ohmic contact with CIGS. It is resistant to react with the elements contained in CIGS, and st ays stable in the highly corrosive Se vapor environment during the absorber growth. The work ing pressure used during sputtering of Mo is found to be related to the diffusi vity of sodium. Increasing th e sputtering pressure generally increases the amount of sodium that can diffuse through. Since there is an optimum amount of sodium diffusion for best device performance, the Mo sputtering pressure needs to be adjusted well. All the UF CIGS devices are fabricated on Mo coated soda-lime glass provided by Shell Solar Industries (SSI). Small amount of Mo wa s sputtered using the Pe rkin-Elmer system for

PAGE 93

93 test purpose. For diode RF s puttering (non magnetron), with th e uniformity shield blocking part of the flux, a growth rate between 10 to 15 nm/min can be achieved using 4mTorr working pressure and 400W deposition power. The substrat e is heavily heated during sputtering and the substrate temperature is elev ated beyond room temperature a lthough the substrate platen is constantly cooled with chilled water. The resulted Mo f ilm generally has good adhesion on soda-lime glass. Resistivity in the low 10-5cm region was achieved on 150nm thick film. Absorber Growth The next step of fabrication is to deposit th e CIGS layer. Vacuum processes are typically adopted for better device performance. Such vacuum processes are generally physical vapor depositions (PVD) including co -evaporation and sputtering [ 65,66]. The current champion CIGS devices are fabricated using absorber depos ited through 3-stage process, which is basically a co-evaporation process [67]. A lternative pathways are selenization of metal precursors within a selenium environment [68-70]. Non-vacuum processes [71-73] are also re ported to be feasible in producing the CIGS layer. The non-vacuum processes generally produce devices with less efficiency, but the production cost is lower than it s vacuum counterparts. They could prove to be a promising way of producing low cost solar cells for terrestrial applications. Here in the University of Florida, the CIGS layer is deposited using a Plasma Migration Enhanced Epitaxy (PMEE) system [74,75]. Figure 2-1 shows a schematic of the PMEE reactor. The system is a variant of traditional Molecula r Beam Epitaxy (MBE) system. The substrate is put on a rotating platen and it sequentially passe s under each source. The metal components, i.e., copper, indium, gallium are sequentially deposi ted during the exposure. The substrate then rotates into the selenium zone a nd encounters a surplus of selenium vapor. At the high substrate temperature (up to 490C) used during the deposition, the co mponents reacts and polycrystalline

PAGE 94

94 CIGS is formed. The substrate is only exposed to each source once during every rotation cycle and the system has a very low throughput compared with other PVD processes such as co-evaporation. It is not econom ical for large scale production usi ng the system but the epitaxial ability makes the system potentially capable fo r exploration into the single crystal regime. Single crystal growth of CIGS can be performed on GaAs substrates. Buffer Layer Deposition Following the absorber layer is the deposition of buffer layer. A wet chemical process called chemical bath deposition (CBD) was utilized to deposit a thin layer of n-type buffer layer on top of CIGS. CBD is a technique of controlled precipitati on of a compound from its aqueous solution. It is a relatively economic way to deposit thin films because no vacuum or high temperature is required. Traditional buffer layer for CIGS solar cell is cadmium sulfide (CdS). For CdS deposition, solutions containing certain concentr ations of ammonia hydroxide, CdSO4 and thiourea are mixed in a water circulated 1 Liter jacket beak er and reaction took place in a controlled batch under a preferred temperature of 65C. CdS wi th approximately 50nm thickness is deposited on to the surface of the substrat e during a 15minute deposition tim e. Table 6-1 shows detailed recipe for traditional CBD deposition of CdS buffer layer. Due to the toxic nature of Cd, there have been efforts to remove or at least reduce the toxic cadmium component and produce more environmentally friendly CIGS solar cells. For the result presented in this dissertation, an alternative buffer layer of Cd1-xZnxS was used. The targeted stoichiometry is a ratio between Cd and Zn as 7:3, which gives a good device performance [76]. The recipe used is similar to that in a patent from Beoing company [77]. It is basically an aqueous solution contains CdCl2 2(1/2)H2O, NH4Cl, thiourea (H2NCSNH2), ZnCl2, and NH4OH

PAGE 95

95 was used for the UF CBD deposition. Detailed concentration of each ingredient is documented in Table 6-2. The chemical bath is formed in a 1 liter jack et beaker heated by circulating hot water and the temperature is maintained at a target of 85 C at all times during film growth. The bath is vigorously stirred and typical deposition duration is 45min, whic h gives a film thickness of approximately 50nm. Window Layer Deposition After depositing the buffer layer, an intrinsic ZnO and an Al doped ZnO (AZO) layer are sequentially deposited using a Perkin Elmer 4400 sputtering system Details of the sputtering system are introduced in Chapter 3. It is known that both the optical tran smission and electrical resistance of the window layer have strong impact on the performance of CIGS solar cells. The transmission determines the amount photons enteri ng the absorber layer wh ile the resistance is related with the series re sistance of the device. Both of thes e factors are associ ated with window layer thickness in a conflicting way. Decr easing ZnO thickness would help to improve transmission but it also causes the increase of sh eet resistance, which may result a deterioration of device performance. Therefore, the thickness of the AZO layer is adjusted to obtain a desirable sheet resistance while maintain ing a reasonable optical transmission. The intrinsic ZnO layer is deposited from a fully oxided target and the growth rate is low. It will dramatically improve the throughput if this layer can be omitted. Ramanathan et al. [78] reported that the conversion efficiency of CIGS solar cell does not change using a simplified device structure without the intrin sic ZnO layer. This could be a major simplification of the fabrication process and could signi ficantly reduce the cost and turn around time. However, there is contradictory report from Ishizuka et al. [79]showing different in trinsic ZnO thickness would

PAGE 96

96 affect the performance of CIGS solar cell. An investigation on this issue is addressed in the work of this dissertation. Metallization To finish up the device, a dual metal layer cont act grid consisting of nickel and aluminum is deposited onto ZnO:Al. A shadow mask is used to define the contact finger pattern. Nickel acts as a diffusion barrier to keep aluminum oxide from entering th e zinc oxide layer. The high melting temperature of Nickel is out of the reach of typical thermal evaporators and makes it necessary to use e-beam evaporation. Th e thickness of Ni and Al are 50nm and 1m respectively. The grid coverage should be small (approximately 5% ) to allow light passing. To improve the cell performance, an MgF2 anti-reflection layer can be ev aporated to reduce the light reflection and allow maximum amount of light ente ring the device. For all the results shown in the scope of this work, no anti-reflection layer was used. Figure 7-2 shows the photo of a CIGS device completed at the University of Florida. Device Characterization System The device current-voltage (I-V) measurement was conducted in our optoelectronic lab. The I-V system uses an ELH (tungsten-halogen) lamp with an integral dich roic rear reflector as the solar simulator. It is classified as a class C solar simulator due to a lack of energy below the wavelength of 0.5 m and excess energy in the wa velength between 0.6 and 0.7 m compared to AM1.5G radiation. The radiation intensity from the solar simula tor was calibrated using the reference cell method. Basically, a calibration cell with NREL certified character ization report was used as the reference cell to calibrate the system. This calibration cell, namely S2049-A1 fabricated in November of 2002, with a total cell area of 0.408 cm2 was measured on 3/22/04 at NREL with

PAGE 97

97 the following parameters for cell #2: Voc = 0.661 V, Jsc = 33.73 mA/cm2, FF =77.47 %, and = 17.28 %. The distance between th e light source and the calibration cell was adjusted to obtain a calibrated position where the measured Jsc matches the certified Jsc value. The temperature of the device tends to increa se under elongated radi ation from the solar simulator. Water cooling system and a thermal couple is embedded in the test stage to control the temperature. Efforts were made to mainta in the temperature of th e test cell at 25 1C during the characterization. Devices with Absorbers from UF PMEE Reactor Devices were first time completely fabricated in house. Two kinds of absorbers, namely CIGS and CGS, were grown in the UF PMEE r eactor. These absorbers were etched in 10% potassium cyanide solution to remove any possibl e Cu rich phases before putting on the buffer layer. All the buffer layers are CdZnS appr oximately 50nm thick. No intrinsic ZnO was deposited so that the structure is simplified and fewer variables are involv ed. AZO window layer of approximately 600nm was deposited to ensure a sheet resistance well below 50/, although later optimization on EPV absorbers indicates th at thinner AZO layer would be sufficient and should have improved the perf ormance of these devices. CIGS Devices For CuIn1-xGaxSe, absorbers with different x value a nd different Cu to group III metal ratio were grown and devices were fabricated usi ng the method described ear lier. The substrate temperature during absorber gr owth was approximately 490C. All metal fluxes were kept constant during the growth. Therefore there wa s no bandgap grading for these absorbers. Table 6-3 shows the film properties of the CIGS absorbers used for devi ce fabrication. These are Cu

PAGE 98

98 poor films with x values vary ar ound 0.3. Film thickness is around 1m. Only one thick sample with thickness of 1.9 m was grown due to the lo w throughput of the system. The I-V properties of these CIGS devices were measured using the I-V system. The solar simulator intensity was calibrated using the reference cell before and after the lengthy measurement to ensure there is no major paramete r drift during the process. The calibration cell performance parameter is the short circuit current. Table 6-4 shows the NREL certified parameters of cell #2 on S2049-A1 and the calibrated parameters of this cell before and after the measurement. It can be seen that the calibrated results are conservative to avoid an exaggeration of the device performance. Table 6-5 lists the measured I-V character of the UF CIGS devices. Device 582, the CIGS with zero Ga content (essentially CuInSe2), showed the highest efficiency of 8.9%. It had a relatively high open circuit voltage and the highe st short circuit current among all the CIGS devices. The highest CIGS device, #588, has good short circuit current and fill factor, but the lower open circuit volta ge compared with device 582 made its efficiency lower. The two best performance absorbers were evaluated under high resolution scanning electron microscope. Figure 6-3 and Figure 6-4 are the cross-section SEM images of absorber #582 and #588 respectively. It can be seen that these two films have similar thickness. Both of these two absorbers have relativ ely large grain size, which is important for high performance. CGS Devices High efficiency CuGaSe2 devices are required as the top cell for high performance tandem chalcopyrite structure. Currently the efficiency of CGS is lagging far behind that of CIGS [80]. Much still needs to be done to improve the efficiency of CGS devices.

PAGE 99

99 Some devices using CGS grown from the PMEE reactor were fabricated. The CGS growth parameters are shown in Table 66. Generally, two growth r ecipes were involved in producing these CGS absorbers. One recipe is an emulated NREL 3-stage growth process trying to create bandgap grading within the absorber layer. The ot her recipe uses constant metal flux to obtain a uniform band structure in the CGS thin films. All of these films were Ga rich with copper to gallium ratio varied slightly be tween 0.91 and 0.95. Table 6-7 shows the device performances of these CGS films. The highest efficiency of 4.9% was achieved on Device 640grown via the emulated 3-stage process. Cross-section SEM image was taken on the UF champion CIGS device and the result is shown in Figure 6-5. Each layer constituting the device can be clearly seen except there is no significant boundary between the CdS and ZnO layers This device has a fairly thick absorber with large grain size in the orde r of microns. Good attachment between the absorber and the underneath Mo layer indicating a g ood stress control on the relatively thin Mo back contact. Effect of Zinc Oxide Window Layer on Device Performance CIGS devices are also fabricated using the absorbers provided by Energy Photovotaics Inc. These absorbers are grown on large sheet of substrates then cut into smaller 22 pieces and shipped from EPV to UF. Therefore, it is reas onable to assume that the absorber layer has similar performance and can be used for the window layer optimization. The effect of optical transmission and sheet resistance of the AZO laye r was investigated. Another issue addressed is the effect of the intrinsic ZnO layer. For experiments conducted to investigate the transmission and sheet resistance, only the AZO thin film layer was deposited and the intr insic ZnO layer was skipped. Doing so would

PAGE 100

100 simplify the solar cell structure and the fabrica tion process, reduce the possibility of effects brought in by an extra step and thus focus on the issue under inve stigation. The impact of intrinsic ZnO layer on solar cell performance was then assessed. The motivation is to evaluate the necessity of this intrinsic ZnO layer. The deposition rate for this layer is generally low because it is sputtered from a fully oxidi zed target. Although only a very thin layer of intrinsic ZnO is required, it generally consumes comparable time as that for sputtering the AZO layer due to the much smaller de position rate. If this layer can be proved to be unnecessary, it would significantly reduce the complication of the fabrication process and greatly improve the process throughp ut. In literature there are contradictory reports regarding to the effect of this layer and furt her investigation needs to be conduc ted. Devices fabricated with and without the intrinsic Zn O layer are compared and the results were analyzed. Effect of Window Layer Transmission The AZO transmission experiments were conducted using CIGS absorbers from EPV. Efforts were made trying to eliminate the possi ble variations coming from other layers. These include cutting the absorbers off from the same 22 CIGS substrate and depositing the ZnCdS buffer layer during the same ba tch of CBD process. A 22 substrate was cut into four 1 1 pieces and three of them were then coated w ith buffer layer during one single batch of bath deposition. The maximum number of substrat es can be processed in one batch of CBD deposition is three due to the number of substrate holders avai lable. The recipe used was targeted to a stoichiometry of Cd1-xZnxS with x equals 0.3. AZO thin films with different optical transmission were then deposited on top of the buffer layer. A dummy gl ass was placed beside each sample to monitor the optical transmission and sheet resistance. Finally, the metal top contact fingers are evaporated to finish the device.

PAGE 101

101 Figure 6-6 shows the window layer optical tran smission of the three aforementioned CIGS devices. For easy comparison, the transmission is categorized into three levels: low, medium and high. The current-voltage (I-V) properties of these solar cells were characterized and important device parameters are lis ted in Table 6-8. The results showed that by increasing the window layer optical transmission, both the open circuit voltage (Voc) and short circuit current (Jsc) were increased. The Voc improved from 0.436V to 0.477V and Jsc increased from 24.12mA/cm2 to 27.50mA/cm2 when the transmission improved fr om low to high region. As a result, the solar cell efficiency was improved from 4.59% for a low transmission window to 6.21% for a high transmission window. This is approximately 35% improvement of the efficiency. The results from AZO transmission experime nts showed that the window layer optical property is critical to solar cell performance. It is a very eff ective way to improve CIGS solar cell performance by enhancing the optical transmi ssion of the AZO window layer. It is also known that improvement of transmission by applying MgF2 antireflection layer can further increase the device performance. Effect of Sheet Resistance As mentioned before, the sheet resistance affect s the series resistance of a solar cell and according the theory showed in Chapter2, it woul d affect the performance of a device. To evaluate this effect, a number of devices with various sheet resi stances were fabricated on CIGS provided by EPV and the device performan ce was evaluated via I-V measurement. Figure 6-7 shows the relationship between AZO layer sheet resistance and CIGS solar cell efficiency. Many data po ints were taken around 50cm region, a value adopted by literature report[81]. But exploration of sh eet resistance to regions much higher than this value also

PAGE 102

102 yielded devices with good efficiency Generally, the device performa nce is very tolerant to sheet resistance over a fairly large range. Reasona ble performance was obtained from devices with window layer as thin as 100nm. A good tolerance to sheet resistance yet bei ng sensitive to transmission indicate device performance can be improved by reducing the window layer thickness to achieve better transmission while maintaining a reasonable sheet resistance. Since many factors coming from different layers of the device can affect the efficiency, it is diffi cult to perform an optimization with limited amount of EPV absorb ers available. Indeed, there is no straight forward trend observed from these performance data. Impact of the Intrinsic ZnO Layer Three pairs of devices were fabricated using CIGS from EPV to evaluate the effect of the intrinsic ZnO layer. Within each pair, one device is fabricated with the intrinsic ZnO layer and the other without. These devices were charac terized using I-V measurement system and the result is shown in Table 6-9. It is evident th at devices with the intrinsic ZnO layer unanimously showed an improvement of Voc. The Jsc value is also higher for the devices with the intrinsic ZnO layer compared to its counterpart, w ith the first pair having very close Jsc values though. As the result, all devices with the intrinsi c ZnO layer showed improved efficiency. These data show that intrinsic ZnO is necessary in terms of achieving high efficiency, at least for the case of EPV absorber s. But from the standpoint of industry manufacturing, this improvement has to be balanced with the cost of an extra step of deposition. A careful evaluation of the benefit and cost need to be made. This is out of the scope of this dissertation. Summary Devices were fabricated usi ng UF CIGS and CGS absorbers. Reasonable performance was achieved on these devices indicating device qua lity ZnO has been achieved. Further

PAGE 103

103 optimization of the window layer to improve device performance was conducted using EPV absorbers. Results showed that the devices are to lerant to the sheet resi stance over a large range. Optical transmission plays a more direct role affecting the efficiency. Device efficiency can be increased by using thinner ZnO as the window laye r. The necessity of the intrinsic ZnO layer was investigated. The results showed that this layer is beneficial to improve the device efficiency. It is necessary to have this layer to achieve best efficiency devices.

PAGE 104

104 Table 6-1. Recipe for CdS buffer layer deposition. Constitute Concentration Volume DI water N/A 365mL Thiourea 1.5mol/L 25mL Ammonia hydroxide 28.0-30.0wt% 62.5mL CdSO4 0.015mol/L 50mL Table 6-2. Recipe for CdS buffer layer deposition. Constitute Concentration Volume DI water N/A 490mL Thiourea 0.0834mol/L 100mL Ammonia hydroxide 3.5-3.75wt% 0.25mL ZnCl2 0.254mol/L 1.4mL CdCl2 2.5H2O 0.0084mol/L 121.9mL Table 6-3. Composition and thickness data for UF CIGS absorbers. Film ID Cu/III ratio Ga/(In+Ga) ratio Thickness ( m) 569 0.97 0.30 0.8 575 0.97 0.33 0.85 578 0.97 0.32 1.3 579 1.00 0.30 1.9 582 0.91 0 1.0 586 0.90 0.25 0.9 587 0.99 0.41 0.85 588 0.98 0.21 0.9 Table 6-4. Calibration of the reference cell. Data Source VOC (V) JSC (mA/cm2) FF (%) Eff. (%) NREL certified value 0.661 33.73 77.47 17.28 Calibration result before measurement 0.642 33.62 76.09 16.42 Calibration result after measurement 0.643 33.56 76.24 16.24

PAGE 105

105 Table 6-5. I-V character of the UF CIGS devices. Device ID VOC (V) JSC (mA/cm2) FF (%) Eff. (%) 569 0.457 24.3 39.4 4.4 575 0.476 26.3 55.5 7.0 578 0.325 26.6 34.2 3.0 582 0.457 30.5 64.0 8.9 586 0.317 21.3 26.8 1.8 587 0.308 18.1 41.7 2.3 588 0.437 29.3 67.0 8.6 Note: Device 579 gave no I-V characteristics Table 6-6. Growth recipe and composition for UF CGS absorbers. Film ID Process Cu/Ga ratio 640 3-stage 0.91 647 Constant flux 0.92 649 3-stage 0.95 655 Constant flux 0.95 Table 6-7. I-V character of the UF CGS devices. Device ID VOC (V) JSC (mA/cm2) FF (%) Efficiency (%) 640 0.502 16.9 57.4 4.9 647 0.480 13.7 55.7 3.7 649 0.426 16.8 46.3 3.3 655 0.516 14.2 57.8 4.2 Table 6-8. Performance of CIGS solar cells with different window layer transmission. Device ID Transmission Voc (V) Jsc (mA/cm2) FF(%) Efficiency (%) EPV5-2 Low 0.436 24.12 43.65 4.59 EPV5-1 Medium 0.467 25.79 47.89 5.77 EPV5-3 High 0.477 27.50 47.35 6.21 Table 6-9. Performance of CIGS sola r cells with and without i-ZnO layer Pair number Device ID i-ZnO Voc (V) Jsc (mA/cm2) FF(%) Efficiency (%) EPV3-1 Yes 0.536 29.64 47.18 7.49 Pair 1 EPV2-2 No 0.488 29.58 47.00 6.78 EPV5-4 Yes 0.494 26.78 45.51 6.02 Pair 2 EPV5-1 No 0.467 25.79 47.89 5.77 EPV 6-1 Yes 0.498 28.75 50.63 7.25 Pair 3 EPV 5-3 No 0.477 27.50 47.35 6.21

PAGE 106

106 Figure 6-1. Schematic of the PMEE reactor. Figure 6-2. Photo of CIGS solar cells fabricated on a 21 soda-lime glass substrate.

PAGE 107

107 Figure 6-3. Cross section SEM image of absorber #582. Figure 6-4. Cross-section SEM image of absorber #588.

PAGE 108

108 Figure 6-5. Cross-section SEM image of CGS device #640.

PAGE 109

109 0 10 20 30 40 50 60 70 80 90 100 300400500600700800 WavelengthTransmission(% ) EPV5-1 EPV5-2 EPV5-3 Figure 6-6. AZO window la yer optical transmission of three CIGS solar cells. 0 1 2 3 4 5 6 7 8 050100150200250300350Sheet resistance ( cm)Efficiency (%) Figure 6-7. CIGS cell efficiency ve rsus the AZO layer sheet resistance.

PAGE 110

110 CHAPTER 7 INVERSE MODELING OF CIGS SOLAR CELLS CIGS solar cell structure c onsists of several layers of compound semiconductors. Although todays material character ization techniques are highly a dvanced, there is still much information remains unknown. For example, the de fect distribution within the CIGS layer has a fundamental impact on the behavior of the absorb er layer and it is critica l to determine the cell conversion efficiency. The work shown in this section is dedicated to develop a convenient way of identifying some of the key unknown factors through numerical simulation. The idea is to construct an inverse problem and solve the unknown parameters through inverse modeling. Details of the procedure are addr essed in the following sections. Definition of Inverse Modeling This section proposes a definition of inverse modeling, the process of so lving an inverse problem. Let a physical system under study be denoted by S. The scientific analysis of system S can be divided into the following three aspects: (i) Parameterization of the system: specification of a minimal set of model parameters whose values completely characterize the system. (ii) Forward modeling: use of the physical laws to make predictions on some observable parameters (output values) basing on some input values. (iii) Inverse modeling: use of the actual results of some measurements of the observable parameters (output values) to infer the actual values of the model parameters. Strong feedback exists among these three steps and a dramatic advance in one of them is usually followed by the other two [20]. A schematic description of the three components discussed above is given in Figure 7-1, which shows a system S with known parameters. The process of feeding through input values to derive output values is called forward modeling. The process of finding the input values that yield given output values is ca lled inverse modeling. A special case of inverse modeling is

PAGE 111

111 system identification, where both input values and output values are known, and the objective is to determine the parameters of system S. Inverse Modeling for Photovoltaic Cells Applying the three steps defined previously to the case of solar cell, the solar cell system can be analyzed under th e following perspectives: (i) Parameterization of the system Examples of input and output parameters of a solar cell system are shown in Figure 7-2. The input parameters include cell configuration pa rameters and material properties parameters. The output parameters may include the open circu it voltage, short circuit current, fill factor, and energy conversion efficiency. (ii) Forward Modeling The physical laws followed in the forward pr ocess are the various physical models used in the device simulation software, such as Medi ci [21]. Basically, th e input parameters are provided to Medici by the user, a nd the software does the calculation to predict the values of the output parameters of interest. (iii) Inverse Modeling Based on specified values of cell performance parameters (o utput values), such as cell efficiency, the inverse modeling method seeks to find the values of a set of input parameters that yield the given output values. It should be noted that for the solar cell case, it is believed that for certain efficiency, there may be more than one set of input parameters which yield the same efficiency value. In other words, different comb inations of the input pa rameters may result in the same output parameter value.

PAGE 112

112 Application of Inverse Modelin g Performance Optimization Numerical simulation has been adapted to optimize the performance of CIGS solar cells [22], but a comprehensive study is far from co mplete. We hope to obtain better understanding and optimization by using inverse modeling. An extension of inverse modeli ng is the endeavor to optimize key performance parameters, such as cell efficiency, to find a set of input parameters that generate the highest possible cell efficiency. This is extremely attractive because improving the cell efficiency has always been the most challenging task for the Photovoltaic community. Approach The inverse problem for solar cell proposed in this work consists of integrating Matlab [23] and Medici. Matlab is responsible for the parameter analysis and for providing the optimization algorithm. Medici is used to carry out the forward modeling part, and to predict the values of the output parameters. The flow chart shown in Figure 7-3 descri bes the procedure for solving the inverse problem. The inverse modeling algorithm is comprised of all the steps between the Start and Stop indicators in the flow chart. The user must first specify the target output parameters. Then the user specifies an initial guess for the input va lues. At this point, an iteration procedure is initiated by Matlab, by feeding the input values to the Medici progr am. In the next step, Medici predicts the corresponding output value for the Ma tlab-specified parameters. At the beginning, a set of initial attempted parameters such as defect energy distribution and concentration were input to Matlab. Then Matlab send this set of parameters to Medici. Medici will proceed to do the forward modeling and generate th e criteria parameters such as open circuit voltage (Voc), short circuit current (Isc) and fill factor (FF). Then values of these criteria parameters were sent back to Matlab, where they were compared w ith the set value. If

PAGE 113

113 the difference between the generate d values and the set values is smaller than a tolerated value (usually a very small value), the inverse process stops and the set of input parameters that will yield the wanted criteria parameters are found. Otherwise, when there is still a significant difference between the generated values and the set values, Matlab will generate a new set of input parameters according the Algorithm used, and send them back to Medici and the process is repeated until the generated values fit the set values. The process can be easily adapted to op timize arbitrary output values, such as Voc, Isc, FF and cell efficiency. The process stops once the op timized parameter reaches its optimum value. For the efficiency case, the process proceeds until the maximum cell efficiency is found. Forward Modeling This section describes details of utilizing Medici for the fo rward modeling. Cell structure is first built in Medici by defi ning each layer of the device. Th en relevant input values for each layer is assigned to describe the properties of these layers. Some examples of the input values are listed. Cell structure generation The cell structure used in the forward modeling is shown in Figure 7-4. At the bottom is a layer of Molybdenum on top of a glass substrat e serving as the ohmic contact for the p type Copper Indium Gallium di-Selenide (CuInxGa1-xSe2 or simply CIGS). Next is the absorber layer p type CIGS. The typical thickness for the CIGS layer is about 2 microns so as to completely absorb the incoming sun light. The OVC (ordered vacancy compound) layer is not intentionally deposited. It is an indium rich n-type layer form ed between the p-type CIGS and the n-type CdS. The existence of this layer was initially suggested by Schimid et al. [24]. The CdS buffer layer is deposited through Chemical Bath Deposition (CBD). On top of CdS, a thin layer of intrinsic ZnO (i-ZnO) is deposited typically by sputtering. Finally, the top of the structure consists of a

PAGE 114

114 layer of Transparent Conductive Oxide (TCO) in the form of heavily Al-doped ZnO, which forms ohmic contact with an Aluminum grid (not shown in the figure) that completes the device. The software Medici uses a mesh concept to describe the property distribution inside the simulated device. It does calculation at each mesh node and then combines these results to form a whole structure view of the complete device. The CIGS cell structure of Figure 7-4 should be first described using mesh concept in order for Medici to proceed with the calculation. As shown in Figure 7-5, the graph on the left shows the mesh generated in Medici and the graph on the right shows each layer. Starting from the bottom to the top, the successive layers are CIGS, OVC, CdS, i-ZnO and Al doped ZnO. Input parameters The simulation temperature is kept at 300K in all cases considered, and the input solar spectrum was introduced from the AM0.dat file provided by Medici. This spectrum generates an input energy density of approximately100 mW/cm2. Some of the important parameters used in the simulation are shown in Table 7-1. The values of these parameters are consistent with the values used in the simulation based on the widely used solar cell simulation software AMPS1D [25]. With all the materials parameters defined in Me dici, the bandgap profile can be generated. An example is shown in Figure 7-6. The layers can be seen in the bandgap profile from left to right are Al-ZnO, i-ZnO, CdS, OVC, and CIGS. Graded CIGS bandgap profile can also be achieved by defining more than one layer of CIGS with different bandgap values. Example of a forward modeling Figure 7-7 shows a typical I-V curve generated by Medici. The input parameters were taken from Table 7-1. Several important device performance parameters can be extracted from

PAGE 115

115 the I-V curve. They are short circuit current (Isc), open circuit voltage (Voc), fill factor (FF) and efficiency (). The short circuit current Isc is defined as the current value at the spot where the I-V curve crosses the vertical axis [7]. From the graph it can be seen that the Isc value is approximately 37 mA/cm2. The open circuit voltage Voc is the voltage value at th e spot where the I-V curve crosses the horizontal axis. Hence, the Voc value is approximately 0.61 V for the case shown in Figure 7-7. Efficiency is defined as inpP Pmax (7.1) where Pmax is the maximum output power of the device and Pinp is the input power of the solar light. Pmax is basically the maximum value of the product of I and V al ong the I-V curve. Pinp is the total input power of the solar light. Its value is approximately 100 mW/cm2 for the input spectrum used in this simulation study. As the maximum power for the case shown in Figure 7-7 is approximately 17.5 mW/cm2, the value of approximately equals to 17.5% for this case. The fill factor (FF) is defined using the expression FF P IVscocmax (7.2) where Isc and Voc are the short circuit current and open circuit volta ge respectively, which can be identified from the I-V curve. Note Pmax can be geometrically taken as the largest rectangular within the IV curve and the product of Isc and Voc is basically the rectangular whose corners are defined by the Isc and Voc spot together with the origin. FF can be geometrically

PAGE 116

116 taken as the area ratio of these two rectangular. It has been di scussed in early chapters that FF is sensitive with the series resistance of the device. Note Isc, Voc, fill factor and efficiency are not four independent parameters. The efficiency can be calculated using Isc, Voc, and fill factor. Therefore, in the inverse process discussed later, the three output parameters used are only Isc, Voc, and fill factor. Modeling of NREL champion device The champion CIGS device in 2003 from NREL was modeled using Medici. This device has an efficiency of 19.2% with Isc = 35.71 mA/cm2, Voc = 0.698 V and FF = 78.12%. Figure 7-8 shows the simulated I-V curv e plotted together with the experimental data obtained by NREL. A reasonable agreement between simula tion and experiment was achieved. This indicates the Medici forward modeling program may be used in the inverse process to automatically determine the appropriate values of some input parameters. Inverse Modeling Using the procedure described earlier, Matlab and Medici can be integrated to fulfill the inverse modeling process and identify important unknow ns in the CIGS absorber layer. Matlab optimization toolbox was utilized for this purpose. The criteria y for optimization is shown in the following equation: 2 0 0 2 0 0 2 0 0)()()( FF FFFF c V VV b I II ayoc ococ sc scsc (7.3) where Isc0, Voc0 and FF0 are the targeted device performance value. Parameters a, b, and c are weighing factors to adjust the convergence preference among the three parameters. From the equation it is clear that during the minimization process of crit eria y, the device performance parameters Isc, Voc, and fill factor approach their targeted values.

PAGE 117

117 Three input variables were adjusted during th e inverse process. They are three defect concentrations in the CIGS absorber layer. The interested defects in th e CIGS layer are shallow level acceptor state, deep level electron trap and deep level hole trap. Their energy levels are assigned with reasonable values within the bandga p which are shown in Table 7-1. The inverse process is supposed to identify a set of input parameters th at would yield certain output parameters. The search range for the three defects is between 217/cm3 and 120/cm3. Note that due to the complication of the forward model (the Medici simulation process), there could be more than one set of input parameters that woul d yield certain output para meters. In other word, the solution may not be unique. Inverse Modeling of NREL champion device The previous mentioned NREL champion device was used as the target for this inverse modeling process. This means the ta rget performance was set so that Isc0 = 35.71 mA/cm2, Voc0 = 0.689 V and FF0 = 78.12%. Inverse modeling seeks a set of input values that yields the same performance values of the champion cell. For the first run case, the star ting values of the three defect concentrations are randomly chosen as 119, 517 and 119/cm3, respectively. This yields a performance parameter set of Voc = 0.500V, Isc = 35.81 mA/cm2, and FF = 72.67%. The inverse process is then executed following the procedures described earlier in Figure 7-3. As shown in Figure 7-9, after only approximately 20 iterations, the th ree performance parameters are converged to values close to the target performance parameters. A set of defect concentrations with values 5.319, 6.619, and 2.618 is identified to give a performance parameter set of Voc = 0.689 V, Isc = 35.7 mA/cm2, and FF = 78.4%. The first run case was successful and very effi cient. This can be contributed to a good initial guess which yields Isc and FF close to their target value. For the second attempt, the initial

PAGE 118

118 guess defect concentrations were taken as 517, 318 and 817/cm3. This reflects a performance parameter set of Voc = 0.357V, Isc = 40.92 mA/cm2, and FF = 53.96%. Note none of these parameters are close to their final target value. The iteration process is shown in Figure 7-10. Compared with the first case, the second inverse process takes mu ch more iterations to converge and stabilize at the target values. The variation of each of the three parameters also covers a much larger range. After approximate ly 80 iterations, a set of defect concentrations with values 5.219, 1.020, and 2.618/cm3 is identified to give a performance parameter set of Voc = 0.688 V, Isc = 35.7 mA/cm2, and FF = 78.4%. Note that for the second inverse process, two defect concentrations converged to the same values that were obtained in the first run. Yet the electron trap concentration converged to another value which had the same order of magnitude. This suggests that the three output parameters are not very sensitive with the electron trap concentration. For the third demonstration cas e, the defect concentrations were chosen at the upper boundary, which corresponds to values of 1.020, 1.020, and 1.020/cm3. This makes the initial guess as far away as possible from the fina l converging point. With this initial guess, the performance parameters are: Voc = 0.500V, Isc = 35.38 mA/cm2, and FF = 69.58%. As seen in the iteration process shown in Figure 7-11, after approximately 40 iterations, the performance parameters converged and stabilized at their targeted values. This yields a defect concentration combination of 5.319, 1.020, and 2.618/cm3, the same point where the second case converged. In conclusion, the inverse proce ss is feasible to identify some input parameters in the CIGS layer. It has been demonstrat ed through different cases, no matter what the initial guess is, the sensitive parameters all converge to the same point. Yet the insensitive parameters can be

PAGE 119

119 randomly located. This process ca n provide an alternative way to determine or at least shine light on some of the sensitive parameters that can not be easily measured through experiments. But numerical simulation can never completely replace experimental work. Theoretical predictions still need to be verifi ed through experimental methods.

PAGE 120

120 Figure 7-1. Schematic diagram illustrating the forward modeling and inverse modeling operations for a given physical system S Input values Output values S (parameters) Forward modeling Inverse modeling

PAGE 121

121 Figure 7-2. Parameter classificati on with examples for solar devices Output values Voc Isc FF Efficiency Input values Cell configuration Cell structure Film thickness Anti reflection coating, etc. Material properties Semiconductor bandgap Electron/hole concentration Electron affinity Electron/hole mobility Density of states in both bands etc.

PAGE 122

122 Figure 7-3. Procedure chart for solar cell inverse modeling. Specify target output parameters Define input parameters (initial guess) Medici predicts performance parameter values (Forward Modeling) Matlab feeds input parameters to Medici Predicted value Specified value? = Yes No Matlab generates another set of input parameters Start iteration = iteration+1 iteration = 0 Stop

PAGE 123

123 Figure 7-4. A typical CIGS cell structure. Figure 7-5. Cell structur e described in Medici. 50nm, Eg=3.3eV 30nm, Eg=2.4eV 30nm, Eg=1.3eV i-ZnO CdS OVC CIGS 2 m, Eg=1.0-1.7eV TCO (ZnO)

PAGE 124

124 Figure 7-6. Bandgap profile of a CIGS device generated by Medici. Figure 7-7. A t ypical I-V curve generated by Medici.

PAGE 125

125 0 10 20 30 40 00.10.20.30.40.50.60.70.8 V (Volt)I (mA/cm2) NREL Champion Cell Medici Simulation Figure 7-8. Medici simulation resu lt for an NREL champion device.

PAGE 126

126 15.0 20.0 25.0 30.0 35.0 40.0Jsc (mA/cm2) 0.20 0.30 0.40 0.50 0.60 0.70 0.80Voc (V) 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 12 14 1 Iteration numberFF Figure 7-9. Iteration of invers e modeling process for there solar cell performance variables: (a) Isc, (b) Voc, and (c) FF. (Case 1) (a) (b) (c)

PAGE 127

127 15.0 20.0 25.0 30.0 35.0 40.0 45.0Jsc (mA/cm2) 0.20 0.30 0.40 0.50 0.60 0.70 0.80Voc (V) 0.40 0.45 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 121416181101 Iteration numberFF Figure 7-10. Iteration of invers e modeling process for there solar cell performance variables: (a) Isc, (b) Voc, and (c) FF. (Case 2) (a) (b) (c)

PAGE 128

128 15.0 20.0 25.0 30.0 35.0 40.0Jsc (mA/cm2) 0.20 0.30 0.40 0.50 0.60 0.70 0.80Voc (V) 0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 12 14 16 18 1 Iteration numberFF Figure 7-11. Iteration of invers e modeling process for there solar cell performance variables: (a) Isc, (b) Voc, and (c) FF. (Case 3) (a) (b) (c)

PAGE 129

129 Table 7-1. Typical inpu t parameters for Medici. Parameters i-ZnO CdS OVC CIGS Thickness (nm) 50 30 30 2000 Electron affinity (eV) 4.0 3.75 3.95 3.8 Eg (eV) 3.3 2.4 1.3 1.12-1.4 Electron mobility (cm2/V-s) 50 6 10 300 Hole mobility (cm2/V-s) 5 3 5 30 Density of States in conduction band Nc (cm-3) 119 119 6.617 318 Density of States in valence band Nv (cm-3) 119 119 1.519 1.519 Doping Concentration (cm-3) 517 616 112 N/A Acceptor State Concentration NAS (cm-3) N/A N/A N/A 518 Electron Trap Concentration NET (cm-3) N/A N/A N/A 119 Hole Trap Concentration NHT (cm-3) N/A N/A N/A 2.818 Acceptor State Energy level (eV) N/A N/A N/A Ev+0.5Eg-0.44 Electron trap Energy level (eV) N/A N/A N/A Ev+0.5Eg+0.2 Hole Trap Energy level (eV) N/A N/A N/A Ev+0.5Eg-0.2

PAGE 130

130 CHAPTER 8 CONCLUSION Parameter Effects on Sputtered AZO The sputtering system setup and deposition pa rameters greatly affect the optical and electrical properties of deposited AZO thin films. This is the reason for th e significant disparities of best resistivity values repor ted in literature, which can be as high as several orders of magnitude. The magnetron is a critical part of the experimental set up. The magnetic field created by the magnetron confines electrons near the surfac e of the target and consequently keeps the plasma close to the target surface instead of evenly distributed between the target and substrate. This reduces the damage done to AZO thin films through bombardment and improves the electrical properties of sputtered AZO thin film. Zinc is very reactive with oxygen and if oxygen is included in the sputtering gas, its amount has to be small and accurately contro lled. Otherwise the resistivity would be significantly increased. Generally for the Pe rkin Elmer sputtering system used, no oxygen is used during normal sputtering processes. Deposition parameters can also affect the op tical and electrical properties of AZO. Examples of such parameters are deposition power working pressure and starting base pressure. Generally, increasing the deposi tion power would increase the s puttering rate, but it tend to reduce the optical transmission if no oxygen is present in the sputteri ng gas. This effect is done through the modification of the zinc and oxygen ratio at the target surface. When the deposition power is too high, the surface of the target can be reduced to a zinc rich mode if no oxygen is present in the sputtering gas. Depending on th e degree of this redu ced condition, optical transmission of AZO films deposited can drop to as lo w as zero. If the target is in such reduced

PAGE 131

131 mode, a regeneration using argon and oxygen mixture ga s is required. To pr event the target from reducing, a deposition power belo w 400W is suggested. Working pressure can also affect the sputtering rate and resistivity of AZO thin f ilms and an operation under the optimized working pressure is recommended. Experimental results showed a significant d ecrease of the resistivity with introducing hydrogen into the sputtering gas. It is also found that AZO resistivity is affected by the starting base pressure, which is related with the amount of residue water vapor, a source of hydrogen. These results support the theore tical predictions in literatur e that hydrogen acts as a doping source for zinc oxide thin films. Device Fabrication Window layer of thin film zinc oxide is critical for CIGS solar cells. It affects the solar cell performance through two mechanisms. First, the thickness of zinc oxide affects th e amount of photons that reaches the CIGS absorber layer. This is directly related with the optical transmission of the zinc oxide layer. When there is no antireflection layer present, th e thickness of zinc oxide window layer can lead to different optical interferences which could be constructive interferences or destructive ones. When constructive interference happ ens, a significant amount of light is direct back into the air without being able to enter the de vice. This could dramatically re duce the solar cell efficiency. Second, the electrical property of zinc oxide layer is also very important. The sheet resistance of zinc oxide affects the series resistance of the device. It is known that the fill factor of solar cell is very sensitive with the series resi stance. A slight increase of the series resistance could lead to a significant efficiency loss. Ther efore, depositing zinc oxide with sufficient low sheet resistance is required for a good solar cell performance.

PAGE 132

132 Defects in CIGS Absorbers CIGS absorbers can be deposited using various methods incl uding both vacuum and non-vacuum techniques. Performance of these CI GS absorbers can vary significantly. It has been a great interest to discove r a theory to explain the disp arities among the performance of these CIGS absorbers. Defect distribution is believed to be an important factor. CIGS is not intentionally doped. The p-type conductivity comes from the various defects existing in the semiconductor. An increase of th e shallow acceptor defect concentrations would increase the carrier concentration and simulatio n results showed a sharp increase of device performance. An inverse modeling process was constructed a nd it can be utilized to identify the defect distributions corresponding to different solar cell perf ormance. This could be used as a theoretical approach of determin ing defect properties to verify results experimentally obtained using DLTS. Future Work The experimental work on hydrogen doping of AZ O thin films can be further investigated by using different concentrations of hydrogen for the sputtering gas. It is possible that increasing the hydrogen concentration would lead to even lo wer resistivity of the AZO thin film. An optimum concentration of hydrogen should be dete rmined. It is of benefit to reduce the resistivity so that the same sheet resistance c ould be achieved with less thickness. This would reduce the material cost and also the processi ng time. Also thinner AZO window layer could reduce the photon absorption and optical interference if appropriate thickness is chosen. For good uniformity on large size substrate, the AZO thickness variation should be carefully studied. A sputtering process that w ould yield good uniformity for large-scale industry

PAGE 133

133 PV applications is desired. This could be ach ieved by choosing the right shape of the target and better arrangement of the ma gnetic field distribution. The defect study on CIGS using inverse modeling should be accompanied by DLTS measurements. A combination of experimental and numerical tools coul d be of help to both individual techniques.

PAGE 134

134 LIST OF REFERENCES 1. K. Zweibel, in Harnessing Solar Power: The Photovoltaics Challenge, (Plenum Press, New York, 1990). 2. L.L. Kazmerski, International Material Review 34, 185 (1989). 3. L. Reith. Sputter deposition of ZnO thin films, Ph.D. Dissertation, University of Florida. 2001. 4. L. Kerr. Growth and thermodynamic modeling of absorber and transparent conductive oxide for copper indium diseleni de material based solar cells, Ph.D. Dissertation, University of Florida. 2004. 5. J. Mazer, in Solar Cells: An Introduction to Crystalline Photvoltaic Technology, (Kluwer Academic Publishers, Boston, 1997). 6. P. Wurfel, in Physics of Solar Cells, from principles to new concepts, (Wiley-Vch, 2005). 7. D. M. Chapin, C. S. Fuller, and G. L. Pearson, Journal of Applied Physics 25, 676 (1954). 8. J. L. Shay, S. Wagner, and H. M. Kasper, Applied Physics Letters 27, 89 (1975). 9. L. L. Kazmerski, F. R. White, and G. K. Morgan, Applied Physics Letters 29, 268 (1976). 10. W. Chen et al., Conference Record of the Twenty Third IEEE Photovoltaic Specialists Conference,422 (1993). 11. K. Ramanathan et al., Progress in Photovoltaics 11, 225 (2003). 12. S. Li, in Semiconductor Physical Electronics, (Plenum Press, New York, 1993). 13. M. B. Prince, Journal of Applied Physics 26, 534 (1955). 14. E. Bucher, Applied Physics 17, 1 (1978). 15. T. Koyanagi, Conference Proceeding 12th IEEE Photovoltaic Specialist Conference, 627 (1976). 16. A. Goetzberger and V. U. Hoffmann, in Photovoltaic Solar Energy Generation, (Springer, 2005). 17. A. W. Blakers et al., Applied Physics Letters 55, 1363 (1989). 18. J. Dietl, D. Helmreich, and E. Sirtl, in Crystals: Growth, Properties and Applications, (Springer, 1981). 19. P. Stradins, Solar Ener gy Materials and Solar Cells 78, 349 (2003).

PAGE 135

135 20. D. L. Staebler and C. R. Wronski, Applied Physics Letters 31, 292 (1977). 21. J. Muller, B. Rech, J. Springer, and M. Vanecek, Solar Energy 77, 917 (2004). 22. S. Bose and A. K. Barua, Journal of Physics D-Applied Physics 32, 213 (1999). 23. O. Kluth et al., Thin Solid Films 351, 247 (1999). 24. M. Yamaguchi, Journal of Applied Physics 78, 1476 (1995). 25. J. Song et al., NCPV and Solar Program Review Meeting 874 (2003). 26. R. G. Gordon, MRS Bulletin 25, 52 (2000). 27. T. Minami, MRS Bulletin 25, 38 (2000). 28. P. S. Nayar and A. Catalano, Applied Physics Letters 39, 105 (1981). 29. X. Jiang, F. L. Wong, M. K. Fung, and S. T. Lee, Applied Physics Letters 83, 1875 (2003). 30. D. H. Xu et al., Physics Letters A 346, 148 (2005). 31. J. Q. Zhao et al., Synthetic Metals 114, 251 (2000). 32. I. D. Kim et al., Applied Physics Letters 89, 022905(2006). 33. P. F. Carcia, R. S. Mclean, M. H. Reilly, and G. Nunes, Applied Physics Letters 82, 1117 (2003). 34. H. C. Cheng, C. F. Chen, and C. Y. Tsay, Applied Physics Letters 90, 012113 (2007). 35. J. Aranovich, A. Ortiz, and R. H. Bube, Journal of Vacuum Science & Technology 16, 994 (1979). 36. M. S. Tomar and F. J. Garcia, Thin Solid Films 90, 419 (1982). 37. M. G. Ambia, M. N. Islam, and M. O. Hakim, Journal of Materials Science 29, 6575 (1994). 38. W. S. Lau and S. J. Fona sh, Journal of Electronic Materials 16, 141 (1987). 39. D. C. Agarwal et al., Journal of Applied Physics 99, 123105 (2006). 40. O. A. Fouad, A. A. Ismail, Z. I. Zaki, and R. M. Mohamed, Applied Catalysis BEnvironmental 62, 144 (2006). 41. J. H. Hu and R. G. Gordon, Solar Cells 30, 437 (1991). 42. S. K. Tiku, C. K. Lau, and K. M. Lakin, Applied Physics Letters 36, 318 (1980).

PAGE 136

136 43. J. H. Hu and R. G. Gordon, Journal of Applied Physics 71, 880 (1992). 44. T. Hada, K. Wasa, and S. Hayakawa, Thin Solid Films 7, 135 (1971). 45. J. B. Webb, D. F. Williams, a nd M. Buchanan, Applied Physics Letters 39, 640 (1981). 46. T. Minami, H. Nanto, and S. Takata, Applied Physics Letters 41, 958 (1982). 47. D. M. Mattox, in Handbook of Physical Vapor Deposition (PVD) Processing, (Noyes Publications, Westwood, New Jersey, 1998). 48. K. Ellmer, Journal of Physics D-Applied Physics 33, R17 (2000). 49. K. C. Park, D. Y. Ma, and K. H. Kim, Thin Solid Films 305, 201 (1997). 50. C. R. Aita, R. J. Lad, and T. C. Tisone, Journal of Applied Physics 51, 6405 (1980). 51. S. H. Jeong and J. H. Boo, Thin Solid Films 447, 105 (2004). 52. A. Gupta and A. D. Co mpaan, Applied Physics Letters 85, 684 (2004). 53. V. Sittinger et al., Thin Solid Films 496, 16 (2006). 54. M. A. Contreras et al., Progress in Photovoltaics 7, 311 (1999). 55. C. Agashe et al., Journal of Applied Physics 95, 1911 (2004). 56. K. K. Kim et al., Applied Physics Letters 83, 63 (2003). 57. Lake Shore 7500/9500 Series Ha ll System User's Manual (2000). 58. G. B. Hoflund, in Handbook of Surface and Interface Analysis, Edited by J. C. Riviere and S. Myhra (Marcel Dekker,Inc., New York, 1998), Chap. 4. 59. R. Jenkins and R. L. Snyder, in Introduction to X-ray Powder Diffractometry, (WileyInterscience, 1996). 60. H. P. Klug and L. E. Alexander, in X-Ray Diffraction Procedures, 2 ed., (Wiley, New York, 1974). 61. J. I. Pankove and N. M. Johnson, Semiconductors and Semimetals 34, 1 (1991). 62. C. G. Van de Walle, Physical Review Letters 85, 1012 (2000). 63. D. W. Hamby et al., Nuclear Instruments & Methods in Physics Research Section B-Beam Interactions with Materials and Atoms 249, 196 (2006). 64. A. Rockett et al., Thin Solid Films 372, 212 (2000).

PAGE 137

137 65. A. F. da Cunha, F. Kurdzesau, and P. M. P. Salome, Advanced Materials Forum Iii, Pts 1 and 2 514-516, 93 (2006). 66. J. R. Tuttle et al., Progress in Photovoltaics 3, 383 (1995). 67. M. A. Contreras, M. J. Romero, and R. Noufi, Thin Solid Films 511, 51 (2006). 68. R. Caballero, C. Guillen, M. T. Gutierrez, and C. A. Kaufmann, Progress in Photovoltaics 14, 145 (2006). 69. C. M. Xu et al., Chinese Physics Letters 23, 2259 (2006). 70. W. Li, Y. Sun, W. Liu, and L. Zhou, Solar Energy 80, 191 (2006). 71. T. Wada et al., Physica Status Solidi A-App lications and Materials Science 203, 2593 (2006). 72. C. Eberspacher, C. Fredric, K. Pauls, and J. Serra, Thin Solid Films 387, 18 (2001). 73. V. K. Kapur, A. Bansal, P. Le, and O. I. Asensio, Thin Solid Films 431, 53 (2003). 74. B. J. Stanbery, Heteroepitaxy a nd nucleation control for the growth of metal chalcogenides using activated reactant sources, Ph.D. Dissertation, University of Florida, 2001. 75. Kincal S., Modeling and control of multiple thermal effusion sources and substrate temperature in molecular beam epitaxy reactors, Ph.D. Dissertation, University of Florida, 2002. 76. J. Song, Development, characterization a nd modeling of CGS/CIGS thin film tandem solar cells, Ph.D. Dissertation, Univ ersity of Florida, 2006. 77. W. Chen, Cadmium Zinc Sulfide by Solution Growth, US patent 5,112,410 (1992). 78. K. Ramanathan, J. Keane, and R. Noufi, the 31st IEEE Photovoltaics Specialists Conference and Exhibition 195 (2005). 79. S. Ishizuka et al., Solar Energy Materials and Solar Cells 87, 541 (2005). 80. D. L. Young et al., Progress in Photovoltaics 11, 535 (2003). 81. N. Dhere et al., NCPV and Solar Program Review Meeting, 853 (2003).

PAGE 138

138 BIOGRAPHICAL SKETCH Wei Liu was born in Hebei, China in April, 1978 and attended elemen tary school there. He moved with his family to Shandong, China in 1992 and spent his high school time in Yantai, Shandong. He attended Tsinghua University in 1996 and graduated with his BS of Chemical Engineering degree in year 2000. He then was recommended into the master program in the same department and received his MS of Chem ical Engineering degree in 2002. During his study in the master program, he did part of his research projects in th e Chemical Engineering department at University of Manchester Institute of Science and Technology (UMIST), Manchester, U. K. After obtaining his MS degree he attended University of Florida and started pursuing his PhD degree in the Chemical Engin eering department. His research focused on investigation of CuInSe2/CdS/ZnO material system and thin film photovoltaic device fabrication and characterization.