USF Libraries
USF Digital Collections

Stability studies of cdte solar cells with varying amounts of cu in the back contact

MISSING IMAGE

Material Information

Title:
Stability studies of cdte solar cells with varying amounts of cu in the back contact
Physical Description:
Book
Language:
English
Creator:
Erra, Swetha
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Copper
Transients
Degradation
Stress testing
Light soaking
Dissertations, Academic -- Electrical Engineering -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Solar energy is one of the abundant, non-polluting renewable energy options in our planet. During the last three decades considerable progress has been achieved in developing technologies to produce electricity from solar radiation, but producing electricity with low cost and low pollution is of concern. The CdTe solar cells are the leading source for the production of cost effective solar cells. The main issue of concern in these CdTe solar cells is degradation observed when stressed at elevated temperatures. The degradation in CdTe solar cells can be attributed to the back contact, which often contains Cu to improve the electronic properties of CdTe (absorber layer) and to enable a quasi ohmic back contact.The main objective of this thesis was to study effect of amount of Cu in the back contact and contact annealing temperature on device stability.
Thesis:
Thesis (M.S.E.E.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Swetha Erra.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 125 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001681068
oclc - 62744298
usfldc doi - E14-SFE0001050
usfldc handle - e14.1050
System ID:
SFS0025371:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Stability Studies of CdTe Solar Cells Fabr icated With Varying Amounts of Cu in the Back Contact by Swetha Erra A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Electrical Engineering Department of Electrical Engineering College of Engineering University of South Florida Major Professor: Christos S. Ferekides, Ph.D. Don L. Morel, Ph.D. Yun L. Chiou, Ph.D. Date of Approval March 22, 2005 Keywords: Copper, Transients, Degradati on, Stress Testing, Light Soaking Copyright 2005, Swetha Erra

PAGE 2

Dedication This thesis is dedicated to my Mother and Father for their struggle and sacrifices to educate their three children Sr avanthi, Praneeth and Myself

PAGE 3

Acknowledgements I owe my gratitude and gratefully acknow ledge Dr. Ferekides for his continual support in every possible way for the past tw o and half years, without his valuable guidance this thesis would not have been possible. I would like to express my special thanks to Dr. Morel for all his suggestions given in the group meetings. I also would like to thank Dr. Chiou and Dr. Morel for their invaluable support in the course work and for serving on the Thesis Committee. I would like to thank all the fellow researchers at the Semiconductors Lab. Special thanks to Mr. Ashok Rangaswamy for training me on stability testing and Mike for his timely help in fixing stability chamber. I would like to thank my friends Vikram, Santosh, Saru, Roja and Sudhakar for all the cherishable moments spent throughout my stay at USF. I would like to thank my sweetest sister and brother for their unconditional love throughout my life. I also would like to ac knowledge my In-laws Mr. A.S.V. Sarma and Mrs. Girija Sarma for their best wishes and blessings whenever I needed them the most. Thanks is not enough to say to my parent s who encouraged me to opt engineering initially and never compromised to provi de me with the very best education. Finally I am extremely thankful to my husband Sriram for his patience, understanding and support in ever y difficult situation since I me t him; he made this world a wonderful place to live in.

PAGE 4

i Table of Contents List of Tables................................................................................................................. ....iv List of Figures................................................................................................................ .....v List of Symbols................................................................................................................ .xii Abstract....................................................................................................................... .....xiii Chapter 1 Introduction........................................................................................................1 1.1 Photovoltaics.............................................................................................................1 1.2 Background of Photovoltaics....................................................................................3 Chapter 2 Solar Cells..........................................................................................................5 2.1 Solar Cells and Solar Energy Materials....................................................................5 2.2 Solar Cell Operation.................................................................................................5 2.3 Solar Cell Parameters................................................................................................6 2.3.1 Short Circuit Current, Isc....................................................................................7 2.3.2 Open Circuit Voltage, Voc..................................................................................7 2.3.3 Fill Factor, FF ....................................................................................................8 2.3.4 Efficiency, .......................................................................................................8 2.3.5 Series and Shunt Resistances.............................................................................9 2.3.5.1 Series Resistance.........................................................................................9 2.3.5.2 Shunt Resistance.......................................................................................10 2.3.6 Diode Factor.....................................................................................................11 2.4 Importance of Thin Film Solar Cells......................................................................11 2.5 CdTe/CdS Solar Cells.............................................................................................11 2.6 CdTe/CdS Solar Cell Structure...............................................................................12 2.6.1 Substrate...........................................................................................................13 2.6.2 Front Electrical Contact...................................................................................13 2.6.3 CdS Window Layer..........................................................................................14 2.6.4 CdTe Absorber Layer......................................................................................15 2.6.5 Back Contact....................................................................................................16 Chapter 3 Literature Review.............................................................................................17 3.1 Development of Efficient and Stable Back Contacts on CdS/CdTe Solar Cells....17 3.2 Effect of Back Contact Copper Concentration on CdTe Cell Operation................19 3.3 Sb2Te3/Mo, Cu/Mo, Sb/Mo Back Contacts............................................................21

PAGE 5

ii3.4 Review on Cu/Au, Sb/Au Back Contacts to CdTe Solar Cells and the Effect of Etching on CdTe Layer with Different Etchants............................................23 3.5 Effect of CdCl2 Treatment......................................................................................26 3.6 Stress Testing..........................................................................................................26 3.6.1 Stability of CdTe/CdS Thin Film Solar Cells..................................................27 3.6.2 Analysis of Stress Induced Degradation..........................................................30 3.6.3 Determination of Cu in CdTe/CdS Devices Before and After Accelerated Stress Testing..................................................................................................33 3.6.4 Thermal Stressing............................................................................................36 Chapter 4 Experimental Techniques.................................................................................38 4.1 Experimental Techniques........................................................................................38 4.2 Current-Voltage Measurements..............................................................................38 4.3 Capacitance-Voltage Measurements.......................................................................39 4.4 Monochromatic J-V Measurements........................................................................40 4.5 Quantum Efficiency Measurements........................................................................40 4.6 Series-Resistance Measurements............................................................................41 4.7 Shunt-Resistance Measurements............................................................................42 4.8 Transient Measurements.........................................................................................42 Chapter 5 Stability Testing...............................................................................................44 5.1 Objective.................................................................................................................44 5.2 Processing Conditions.............................................................................................45 5.3 Stability Testing Procedure.....................................................................................46 5.4 Obstacles During The Stress Period.......................................................................50 Chapter 6 Results and discussion......................................................................................51 6.1 Set 1(10, 20 and 40 of Cu, Contact Annealed at 240oC).....................................52 6.1.1 Initial Performance of Set 1, Measur ed at Room Temperature Using Simulator.........................................................................................................52 6.1.2 Degradation in Voc and FF with Time............................................................53 6.1.3 Variation in Voc and FF at OC and SC Conditions.........................................55 6.1.4 Comparison of Degradation in Voc at 10, 20 and 40..............................57 6.1.5 Comparison of Degradation in FF at 10, 20 and 40................................58 6.1.6 J-V Characteristics for Samples Before and After the Light Soaking.............58 6.1.6.1 Sample with 10 Cu in Back Contact and Contact Annealed at 240oC...59 6.1.6.2 Sample with 20 Cu in Back Contact and Contact Annealed at 240oC...62 6.1.6.3 Sample with 40 Cu in Back Contact and Contact Annealed at 240oC...65 6.2 Analysis for Samples with 10 Cu and Contact Annealing Temperatures of 175oC and 200oC.............................................................................................68 6.2.1 Variation of Voc and FF with Time.................................................................69 6.2.2 J-V Characteristics for Set 2............................................................................73 6.2.3 Variation in Voc and FF at OC and SC Conditions.........................................75 6.3 Effect of Light Soaking on Rse................................................................................78 6.4 Variation in Rsh with Light Soaking........................................................................80

PAGE 6

iii6.4.1 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Dark and Light at OC.....................................................................................................80 6.4.2 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Dark and Light at SC......................................................................................................81 6.5 Transient Behavior..................................................................................................82 6.6 Conclusions.............................................................................................................90 References..................................................................................................................... ....91 Appendix A: Measurement Automation...........................................................................95 Appendix B: Degradation in Solar Cell Parameters.........................................................99 B.1 Variation of Rse.....................................................................................................99 B.1.1 During Dark Cycle..........................................................................................99 B.1.2 During light soaking......................................................................................100 B.2 Variation of Rsh...................................................................................................101 B.2.1 During Dark Cycle at OC..............................................................................101 B.2.2 During Dark Cycle at SC..............................................................................102 B.2.3 During Light Cycle at OC.............................................................................104 B.2.4 During Light Cycle at SC..............................................................................105 B.3 Comparison of Cu Sputtered Thickness with Dark J-V.......................................106 B.4 FF Variation in Monochromatic J-V....................................................................108 B.5 Effect of Contact an nealing Temperature on JV..................................................109

PAGE 7

iv List of Tables Table 1.1 Notable Events in the History of Photovoltaics [20]..........................................2 Table 1.2 Advantages and Disadva ntages of Photovoltaics [20]........................................4 Table 3.1 Composition of the Etch ants Br/MeOH, NP1 and NP2....................................23 Table 3.2 Sheet Resistances with Di fferent Etching Times in NP1.................................24 Table 6.1 List of Devices Sele cted for Stability Analysis................................................51 Table 6.2 Initial Performance of Set 1..............................................................................52 Table 6.3 Initial Performance of Set 2..............................................................................68 Table 6.4 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Dark OC.......80 Table 6.5 Shunt Resistance Variation for 10 20, 40 up to 640hrs in Light at OC..80 Table 6.6 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Dark SC........81 Table 6.7 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Light SC.......81 Table B.1 Variation of Series Resistance for Bo th Open and Short Circuit Conditions with Time for 10, 20 and 40 of Cu Concentration Respectively During Dark Cycle.......99 Table B.2 Variation of Series Resistance fo r Both Open and Short Circuit Conditions with Time for 10, 20 and 40 of Cu Concentration Respectively During Light Cycle Table B.3 Initial FF Comparison of 10, 20 and 40 of Cu Sputtered Back Contact Annealed at 240oC.108

PAGE 8

v List of Figures Figure 2.1 Energy Band Diagram of a p-n Junction in Thermal Equilibrium....................6 Figure 2.2 Ideal Solar Cell..................................................................................................7 Figure 2.3 Dark and Light I-V Characteristic of Solar Cell...............................................8 Figure 2.4 Circuit Model for a Solar Cell...........................................................................9 Figure 2.5 Effect of Series Resistance on the I-V Curve of a Solar Cell..........................10 Figure 2.6 Effect of Shunt Resistance on the I-V Curve of a Solar Cell..........................10 Figure 2.7 Energy Band Diagram of a CdTe-CdS p-n Junction Showing the Drift of Electrons and Holes Under the Influence of Light.........................................12 Figure 2.8 CdTe-CdS Photovoltaic Structure...................................................................13 Figure 2.9 Maximum Conversion Efficiency is Dependence of Bandgap Width for Solar Spectrum AM1.5...................................................................................15 Figure 3.1 SEM Images of the CdTe Morphol ogy after Etching, th e BrMeOH Etch (left) Affects the Morphology less Th an the NP Etch (right) Which Smoothens the Surface and Widens GBs........................................................17 Figure 3.2 Accelerated Stability Tests at 65C (l eft) for Solar Cells with Different Back Contacts and Test of Stable Cells with Mo Metallization at 80C (right)..............................................................................................................18 Figure 3.3 SIMS Depth Profiling of a CdTe /CdS Cell with Cu/Au Back Contact...........19 Figure 3.4 Quantum Efficiency as a Functi on of Wavelength of the Incident Light........20 Figure 3.5 Room-Temperature C-V Curves Show ed Progressive Deterioration in Fill Factor with Reduced Copper..........................................................................20 Figure 3.6 Room-Temperature J-V Curves Showed Progressive Deterioration in Fill Factor with Reduced Copper..........................................................................21

PAGE 9

viFigure 3.7 Doping Profiles of Ce lls with Mo Metallization.............................................22 Figure 3.8 Effect of Etching Time on the Efficiency........................................................24 Figure 3.9 XRD Pattern of the CdTe Layer. The Te (1121) Peak Indicates the Production of Hexagonal Elemental Te at the CdTe Surface with Etching Time................................................................................................................25 Figure 3.10 Effect of Contact A nnealing on Cell Performance........................................25 Figure 3.11 Comparison of Cu SIMS Profiles in Thin Films of Polycrystalline CdTe for Samples with Different Degrees of Crystallinity......................................27 Figure 3.12 SIMS Profile Comparing the Effect of Ni & Ti Metallization......................28 Figure 3.13 SIMS Profiles for Cl, S, Sn and Te...............................................................29 Figure 3.14 Schematic Illustration of Cu and Cl within CdTe/CdS Solar Cells...............29 Figure 3.15 Light and Dark J-V Curves for Cd S/CdTe Solar Cells: Initial; 6 Weeks Stress at 100oC at OC; and Following Recontacting......................................31 Figure 3.16 Relative Change in the Efficiency of Devices with Different Contacts after Stressing at Bias in Light for 10 Days at 100oC.............................................32 Figure 3.17 FF and Voc for Devices with 0 or 6 nm Cu for Stress under FB in the Dark at 60oC for 10 Days................................................................................33 Figure 3.18 SIMS Profile Obtained by Profiling fr om the CdTe (Back) Side of the 0.8% Stressed Device......................................................................................34 Figure 3.19 SIMS Profile Obtained by Profiling Th rough the SnO2, (Front) Side of the Device.......................................................................................................34 Figure 3.20 SIMS Profile of Stressed and Un stressed Devices with Different Cu Concentrations................................................................................................35 Figure 3.21 Comparison of Cu Depth Prof iles from Devices with Different Cu Concentrations in the Contact After Stress.....................................................35 Figure 3.22 Comparison of Cu Depth Profiles from all Unstressed Contacted Devices and Also Stressed Graphite-Only Devices......................................................36 Figure 3.23 SIMS Depth Profiles for CdTe Cells Contacted vs. Uncontacted (Unstressed)....................................................................................................37

PAGE 10

vii Figure 3.24 Stressed vs. Unst ressed Sample (Control).....................................................37 Figure 4.1 I-V Curve for an Ideal Solar Cell....................................................................38 Figure 4.2 A.M 1.5 Condition...........................................................................................39 Figure 4.3 dV/dJ Vs Voltage Plot to Obtain Rs................................................................42 Figure 4.4 dV/dJ Vs Voltage Plot to Obtain Rsh..............................................................42 Figure 4.5 Transient Behaviors in FF (4 Hour Dark Cycle is Not Shown in the Figure).43 Figure 5.1 Different Techniques for Cu Sputtered Samples.............................................44 Figure 5.2 Selected Sample...............................................................................................45 Figure 5.3 Schematic for Light Soaking Setup.................................................................47 Figure 5.4 Top View of a Substrate..................................................................................48 Figure 5.5 Schematic of the Stability Set-Up...................................................................49 Figure 6.1 Variation in Voc and FF with Cu Sputtered Thickness (Back-Contact Annealed at 240oC).........................................................................................52 Figure 6.2 Degradation of Voc and FF at OC & SC for 10 Thick Cu Sputtered Back Contact Annealed at 240oC.............................................................................53 Figure 6.3 Degradation of Voc and FF at OC & SC for 20 Thick Cu Sputtered Back Contact Annealed at 240oC.............................................................................53 Figure 6.4 Degradation of Voc and FF at OC & SC for 40 Thick Cu Sputtered Back Contact Annealed at 240oC.............................................................................54 Figure 6.5 Variation in Voc and FF for 10 of Cu Sputtered, Back Contact Annealed at 240oC...........................................................................................................55 Figure 6.6 Variation in Voc and FF for 20 of Cu Sputtered Back Contact Annealed at 240oC...........................................................................................................56 Figure 6.7 Variation in Voc and FF for 40 of Cu Sputtered Back Contact Annealed at 240oC...........................................................................................................56

PAGE 11

viiiFigure 6.8 Variation in Voc for 10, 20, 40 of Cu Sputtered Back Contact Annealed at 240oC..........................................................................................57 Figure 6.9 Variation in FF for 10, 20, a nd 40 of Cu Sputtered Back Contact Annealed at 240oC..........................................................................................58 Figure 6.10 Light J-V at OC for 10 Thick Cu Sputtere d Back Contact Annealed at 240oC Initial and After 640hrs........................................................................59 Figure 6.11 Light J-V at SC for 10 Thick Cu Sputtere d Back Contact Annealed at 240oC Initial and After 640hrs........................................................................59 Figure 6.12 Dark J-V at OC for 10 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs........................................................................60 Figure 6.13 Dark J-V at SC for 10 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs........................................................................60 Figure 6.14 Dark and Light J-V at OC (Left) and SC (Right)..........................................61 Figure 6.15 Semi-Logarithmic Dark J-V at OC (Left) and SC (Right)...........................61 Figure 6.16 Light J-V at OC for 20 Thick Cu Sputtere d Back Contact Annealed at 240oC Initial and After 640hrs........................................................................62 Figure 6.17 Light J-V at SC for 20 Thick Cu Sputtere d Back Contact Annealed at 240oC Initial and After 640hrs........................................................................62 Figure 6.18 Dark J-V at OC for 20 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs........................................................................63 Figure 6.19 Dark J-V at SC for 20 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs........................................................................63 Figure 6.20 Dark and Light J-V at OC (Left) and SC (Right)..........................................64 Figure 6.21 Semi-Logarithmic Dark J-V at OC (Left) and SC (Right)............................64 Figure 6.22 Light J-V at OC for 40 Thick Cu Sputtere d Back Contact Annealed at 240oC Initial and After 640hrs........................................................................65 Figure 6.23 Light J-V at SC for 40 Thick Cu Sputtere d Back Contact Annealed at 240oC Initial and After 640hrs........................................................................65

PAGE 12

ixFigure 6.24 Dark J-V at OC for 40 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs........................................................................66 Figure 6.25 Dark J-V at SC for 40 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs........................................................................66 Figure 6.26 Dark and Light J-V at OC (Left) and SC (Right)..........................................67 Figure 6.27 Semi-Logarithmic Dark J-V Plot at OC (Left) and SC (Right).....................67 Figure 6.28 Initial Voc (Left) and FF (Right) for 10 Thick Cu Sputtered Back Contact with Varying Contact Annealing Temperature.................................69 Figure 6.29 Degradation of Voc at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 175oC...................................................69 Figure 6.30 Degradation of FF at OC (left) & SC (right) for 10 Thick Cu Sputtered Back Contact Annealed at 175oC....................................................................70 Figure 6.31 Degradation of Voc at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 200oC...................................................70 Figure 6.32 Degradation of FF at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 200oC....................................................................71 Figure 6.33 Degradation of Voc at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 240oC....................................................................71 Figure 6.34 Degradation of FF at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 240oC....................................................................72 Figure 6.35 Dark J-V at OC for 10 Thick Cu Sputtered Back Contact Annealed at 175oC...............................................................................................................73 Figure 6.36 Dark J-V at SC for 10 Thick Cu Sputtered Back Contact Annealed at 175oC...............................................................................................................73 Figure 6.37 Dark J-V at OC for 10 Thick Cu Sputtered Back Contact Annealed at 200oC...............................................................................................................74 Figure 6.38 Dark J-V at SC for 10 Thick Cu Sputtered Back Contact Annealed at 200oC...............................................................................................................74 Figure 6.39 Initial Voc at Different Contact Annealing Temperatures............................75

PAGE 13

xFigure 6.40 %Degradation in Voc at Differe nt Contact Annealing Temperatures...........76 Figure 6.41 Initial FF at Different Contact Annealing Temperatures..............................77 Figure 6.42 %Degradation in FF at Differe nt Contact Annealing Temperatures.............77 Figure 6.43 Rse Variation in Dark Condition...................................................................79 Figure 6.44 Rse Variation under Light Soaking...............................................................79 Figure 6.45 Variation of Voc with Time (4 H our Dark Cycle Transpired in Between is not Shown in the Figure).................................................................................82 Figure 6.46 Variation of Voc at OC and SC (Not Shown in the Figure 4 Hour Dark Cycle)..............................................................................................................83 Figure 6.47 Variation of FF with Time.............................................................................84 Figure 6.48 Variation in FF at OC and SC (4 Hour Dark Cycle Transpired in Between is Not Shown in the Figure)............................................................................85 Figure 6.49 Variation of Voc at OC and SC for 10 Sample..........................................86 Figure 6.50 Temperature Dependence of FF at OC and SC for 10 Sample..................87 Figure 6.51 Variation of Voc and FF with Time for 20 Sample....................................87 Figure 6.52 Temperature Dependence of FF at OC and SC for 20 Sample..................88 Figure 6.53 Variation of Voc and FF with Time for 20 Sample....................................88 Figure 6.54 Temperature Dependence of FF at OC and SC for 40 Sample..................89 Figure 6.55 Variation of Voc and FF with Time for 40 Sample....................................89 Figure A.1 Front Panel of VI Pr ogram for Stability Testing.96 Figure A.2 Snap Shot of Block Diagram...97 Figure A.3 Circuit Configura tion of Hardware Setup...97 Figure A.4 Input Panel for Measurement Interval Specification...

PAGE 14

xi Figure B.1 Variation of Series Resistance for both Open and Short Circuit Conditions with Time for 10, 20 and 40 of Cu Concentration Respectively During Dark Cycle....99 Figure B.2 Variation of Series Resistance for Bo th Open and Short Circuit Conditions with Time for 10, 20 and 40 of Cu Concentration Respectively During Light Cycle.....100 Figure B.3 Variation of Shunt Resi stance at Dark OC for Samples w ith 10, 20, 40 Cu Sputtered Contact......101 Figure B.4 Variation of Shunt Resistance at Dark SC for Samples with 10, 20, 40 Cu Sputtered Contact......102 Figure B.5 Degradation in Rsh in Dark Condition.. Figure B.6 Variation of Shunt Resistance at Li ght SC for Samples with 10, 20, 40 Cu Sputtered Contact.......... Figure B.7 Variation of Shunt Resistance at Li ght OC for Samples with 10, 20, 40 Cu Sputtered Contact......105 Figure B.8 Variation of Shunt Resistance Under Light Soaking.....106 Figure B.9 Dark Analysis of 10, 20 and 40 of Cu Sputtered Back Contact Annealed at 240oC at OC........106 Figure B.10 Dark Analysis of 10, 20 and 40 of Cu Sputtered Back Contact Annealed at 240oC at SC.....107 Figure B.11 Light Analysis of 10, 20 and 40 of Cu Sputtered Back Contact Annealed at 240oC at OC (left) and SC (right)...107 Figure B.12 Degradation of Voc at OC & SC fo r 40 Thick Cu Sputtered Back Contact Annealed at 240oC.....108 Figure B.13 Initial FF Comparison of 10, 20 a nd 40 of Cu Sputtered Back Contact Annealed at 240oC.....109 Figure B.14 Light J-V at OC & SC for Different Contact Annealing Temperatures..109

PAGE 15

xii List of Symbols Rs Series Resistance Rsh Shunt Resistance SC Short Circuit OC Open Circuit Io Reverse Saturation current Wavelength of Light ( ) Absorption Coefficient of Light at given Wavelength IL Load current A Diode Factor Voc Open Circuit Voltage FF Fill Factor Jsc Short Circuit Current Angstrom Ohm

PAGE 16

xiii Stability Studies of CdTe Solar Cells Fabricated With Varying Amounts of Cu in the Back Contact Swetha Erra ABSTRACT Solar energy is one of the abundant, non-polluting renewable energy options in our planet. During the last three decades c onsiderable progress has been achieved in developing technologies to produce electricity from sola r radiation, but producing electricity with low cost and low pollution is of concern. The CdTe solar cells are the leading source for the production of cost effec tive solar cells. The main issue of concern in these CdTe solar cells is degradation obser ved when stressed at elevated temperatures. The degradation in CdTe solar cells can be at tributed to the back contact, which often contains Cu to improve the electronic propertie s of CdTe (absorber layer) and to enable a quasi ohmic back contact. The main objective of this thesis was to st udy effect of amount of Cu in the back contact and contact annealing temperature on device stability. The samples used for this study had varied amounts of Cu sputtered thickne ss in the back contact and varied contact annealing temperatures. The st ress tests were conducted in da rk and light, at elevated temperatures, and under open circuit and short circuit conditions. The stress analysis was done after 700 hours of light so aking. Degradation in specific parameters like Voc, FF, Rse, Rsh has been noted and an attempt has been made to explain these results.

PAGE 17

1 Chapter 1 Introduction 1.1 Photovoltaics Photovoltaics (PV) comprise the technology to convert sunlight directly into electricity. The term photo means light a nd voltaic means elec tricity. A photovoltaic cell also known as solar cell is a semiconduc tor device that generates electricity when light falls on it. When sunlight falls on a PV cell the photons in th e sunlight dislodge the electrons from the atoms of the material, th ese free electrons move across the cell filling and creating holes. This move ment of holes and electrons generates electricity and the physical process of converting sunlight into electricity is call ed photovoltaic effect. Commercialization of photovol taic systems needs a technology, which is reliable, efficient and most importantly low cost. Ev ery semiconducting material under a suitable electronic environment is capable of exhi biting photovoltaic properties, i.e. the generation of an electric current and pot ential difference under illumination. What may be considered surprising is only few material s are able to form photovoltaic devices with sufficient efficiency to make them of potential in terest for practical applications. It is still noteworthy that only the following materials have exhibited a solar efficiency more than 10%: silicon (Si), gallium arse nide (GaAs), indium phosphid e (InP), cadmium telluride (CdTe), copper indium diselenide (CuInSe2) and cuprous sulphide (Cu2S). Of these the most versatile is Si, which can be used to produce efficient cells in single crystal, polycrystalline or amorphous form.

PAGE 18

2 Table 1.1 Notable Events in the History of Photovoltaics [20]

PAGE 19

3 1.2 Background of Photovoltaics In 1839 Edmond Becquerel noticed that By the action of a beam of sun light over two different liquids, chemically interacting and carefully superposed in a glass container, an electric current was deve loped, as indicated by a very sensitive galvanometer connected with two platinum plates dipping in the two different solutions [1] The development of photovoltaic cells can be dated origin ally to French physicist Antoine-Cesar Becquerels discovery. About 40 years later, Adams and Day observed photovoltaic effect in selenium (Se). Solar effi ciencies of about 1% characterized Se and copper oxide (CuO) cells by 1914. The modern era of semiconductor photovoltaics started in 1954 when Chapin, Fuller and Pear son obtained a solar efficiency of 6% for a silicon junction cell, a value that increas ed to 14% by 1958, and to 28% by 1988[20]. The year 1954 Reynolds obtained an efficiency of 6% with first all-thin-f ilm cell composed of a CuxS/CdS junction; later 9% efficiency was obtained by Brag agnolo in 1980, which unfortunately had stability problems. Concer n about energy resources motivated a strong interest in solar cells for terr estrial applications in the earl y 1970s. In spite of the fact that the 1980s and s have been a period in which publ ic and government support for photovoltaics have been underemphasized, c onsiderable activity in research and development of solar cells has continued.

PAGE 20

4 Table 1.2 Advantages and Disadvantages of Photovoltaics [20]

PAGE 21

5 Chapter 2 Solar Cells 2.1 Solar Cells and Solar Energy Materials Solar energy is one of the abundant, non-pol luting renewable ener gy in our planet. During the last three decades considerable progress has been achieved in developing technologies to harness electricity from so lar radiation. The most commonly used solar cell material is silicon with an average effici ency of 24% [2]. But these high efficiency devices are not best suited for terrestrial app lications because they are very expensive. In the last two decades considerable work has b een done in thin film solar cells to replace costly solar cell materials. CdTe and Cu(In, Ga)Se2 are the leading thin film photovoltaic materials with efficiencies exceeding 16.4%[3] and 19.2[4]% resp ectively. But the predicted efficiencies are higher than the reported values; theref ore much work is required to optimize the material properties of the films, the junc tions and also the fa brication techniques. 2.2 Solar Cell Operation Solar cell is the device used to convert solar energy into electrical energy. It works like a semiconducting diode, which is optimized in order to get maximum electrical power when exposed to sunlight. When a photon strikes the cell it is absorbed by an electron in the valence band and theref ore provides energy to the electron elevating it to the electronic state in the conduction band leaving a hol e in the valence band. These electrons and holes act as free charge carri ers and contribute to the current. In the absence of the electric field both electron an d hole randomly move until they recombine.

PAGE 22

6 If during this random movement they get into the space charge regi on of the p-n junction they are separated by the electric field and collected as a part of photocurrent. Figure 2.1 shows the energy band diagram for p-n junction under equilibrium. Figure 2.1 Energy Band Diagram of a p-n Junction in Thermal Equilibrium To maximize the efficiency of the solar ce lls the following conditions have to be met: 1. A major portion of the incident light shoul d be absorbed by th e cell to generate electron-hole pairs 2. The cell structure should provide a m echanism for separating the generated charge carriers. 3. The lifetimes of the minority charge carriers are high to ensure good collection efficiency. 2.3 Solar Cell Parameters

PAGE 23

7 Figure 2.2 shows the equivalent circuit diagram of an ideal solar cell, which includes a current generator representing photo-excitation. Figure 2.2 Ideal Solar Cell The current flowing through the circ uit in the presence of light is L oI AkT qV I I ] 1 ) [exp( 2.1 where, Io is the reverse saturation current, A is ideality factor (discussed later), and IL is the light generated current. We shall now look at each of the parameters quantifying the perfor mance of a solar cell: 2.3.1 Short Circuit Current, Isc Short circuit current is the current when no bias is applied to the cell. Isc can be obtained by substituting V= 0 in the equation 2.2. L scI I 2.2 2.3.2 Open Circuit Voltage, Voc The voltage developed when terminals are op en (Infinite load resistance) is called open circuit voltage, Voc is given by ] 1 ln[ ] [ o l ocI I q kT A V 2.3

PAGE 24

8 Figure 2.3 Dark and Light I-V Characteristic of Solar Cell 2.3.3 Fill Factor, FF From Figure 2.3 Pmax is maximum power that can be obtained where Vmax and Imax are voltage and cu rrent points at Pmax Pmax is area of the rectangle with sides Vmax and Imax. max max maxV I P 2.4 FF is defined as the ratio of the maximum power to the product of Voc and Isc. sc ocI V I V FFmax max 2.5 It is a measure of the squareness of the light curve. 2.3.4 Efficiency, The efficiency of the solar cell is defi ned as the ratio of Pout, electrically generated power to the input power on the cell. in sc ocP FF I V 2.7

PAGE 25

92.3.5 Series and Shunt Resistances The losses in the solar cell can be re presented by adding a series and shunt resistance to the ideal solar cell structure. Figure 2.4 Circuit Model for a Solar Cell 2.3.5.1 Series Resistance The series resistance of a solar cell is mainly due to metal-semiconductor contact resistance, ohmic resistance in metal contact s, and ohmic resistance in semiconductor substrate. The Rs is the resistance the carriers find on their way, it is due to the low conductivity of the wind ow layer and absorber layer and recombination of carriers into the bulk materials. The series resistance d ecreases the maximum achievable output power and hence softens the IV characteristics of a solar cell in the fourth quadrant, eventually degrading fill factor of the solar cell. It is found that the fill factor of a solar cell decreases by about 2.5% for each 0.1 increase in series resistance [5]. In open-circuit condition the current is zer o and the voltage drop across the Rs is also zero, so the Voc is not affected by the Rs. In short-circuit condition, current flows

PAGE 26

10 through the series resistance and the changes th e short circuit current. Variation in I-V characteristic due to change in Rs is shown in Figure 2.5 Figure 2.5 Effect of Series Resistance on the I-V Curve of a Solar Cell 2.3.5.2 Shunt Resistance Shunt resistance is due to le akage currents, pinholes and voids. Shunt resistance degrades the fill factor and Voc. The resistance of the diode is more than the Rsh, which enables current path through the shun t resistance thereby affecting the Voc. Variation in I-V characteristics with Rs h is shown in Figure 2.6 Figure 2.6 Effect of Shunt Resistance on the I-V Curve of a Solar Cell

PAGE 27

112.3.6 Diode Factor The diode factor is the parameter characterizing the perfectness of the junction. The value is dependent on the m echanism of junction transport, e.g. If A=1, the transport is dominated by diffusion; if recombination is the primary transport mechanism then the value of A is closer to 2.This factor is dependent on the process and material and should be close to 1 to get optimum output for sola r cells. Defect and impurity precipitates have been shown to increase the value of A. A can be determined by the slope of the Ln (J)-V curves. 2.4 Importance of Thin Film Solar Cells Crystalline silicon has been the do minant photovoltaic technology since the development of the first so lar cell in the Bell Laborato ries. When talking about drawbacks of Si, we can say that Si is the most weakly absorbing semiconductor used for solar cells because it has an indirect bandga p while most of the other semiconductors have a direct bandgap. Therefor e, at least ten times more cr ystalline Si is needed to absorb a given fraction of sunlight compared to other semiconductors like GaAs, CdTe, Cu(InGa)Se2. Thicker semiconductor material mean s higher material volume and also a higher quality material because of the long pa ths that the high-energy electrons excited by the photons must travel before they are de livered to the external circuit to produce useful work. All this leads to high material cost. 2.5 CdTe/CdS Solar Cells Of the potential photovolta ic technologies, thin f ilm CdTe/CdS photovoltaic devices have marked there prominence. Cd based II-VI semiconductors gained importance in making photovoltaic devices, CdTe is the one that rece ives more interest

PAGE 28

12 because of its low cost and high efficiency in making thin film solar cells. CdTe has a direct bandgap (Eg =1.45 eV at room temperat ure), optimal to absorb the terrestrial solar spectrum. The maximum theoretical efficien cy corresponding to such bandgap is about 27 percent. A few micron thick CdTe film ab sorbs more than 90 percent of the light having photon energy above the bandgap. The sma ll thickness required for an absorbing layer makes the cost of material for the sola r cells relatively low. Figure 2.7 depicts the drift of electrons and holes under the influence of light. Figure 2.7 Energy Band Diagram of a CdTe-CdS p-n Junction Showing the Drift of Electrons and Holes Under the Influence of Light 2.6 CdTe/CdS Solar Cell Structure The CdS/CdTe solar cell structure can be divided into five components: 1) The substrate, 2) the front contact, 3) th e window layer, (4) the absorber layer (5) the back contact. Figure 2.8 shows the pictorial depiction of the different components of a CdS/CdTe thin film solar cell.

PAGE 29

13 Figure 2.8 CdTe-CdS Photovoltaic Structure 2.6.1 Substrate Substrate is very important component of the solar cell; it should withstand the cell fabrication process temperatures and must not contaminate the layers that are grown later. CdS /CdTe solar cells have a superstrat e structure it implies that light enters the substrate first; hence CdTe solar cells require a transparent substrate so that absorption doesnt occur in the substrate, which would be detrimental to the current generation of the cell. The best choice for a transparent su bstrate is glass because it is cheap and withstands high temperatures Common types used are soda-lime glass, which is inexpensive, and borosilicate glass [6]. The latter has a higher softening temperature, and for this reason is often used for the higher te mperature deposition methods, but since it is approximately 10 times expensive, soda-lime glass is generally preferred for low-cost production. 2.6.2 Front Electrical Contact In general transparent conducting oxides (TCO) are used as front contact. In general TCOs are used as the front contact [6 ]. The most widely used materials is tin oxide (SnO2), it is deposited onto the glass either by sputtering or atmospheric pressure

PAGE 30

14 chemical vapor deposition. As tin oxide has too low conductivity for making a good contact, it is often doped either with indi um, forming indium tin oxide (ITO) or with fluorine giving a compound Fluorine tin oxide (FTO). 2.6.3 CdS Window Layer The polycrystalline CdS layer acts as the n part of the p-n junction. It can be deposited by different deposition techniques; In this research work Chemical Bath Deposition (CBD) is used. As a wide bandga p material (Eg=2.42 eV at 300K), it is largely transparent to wave lengths of around 510 nm. Ther efore it also permits the absorber layer to receive the photons of lowe r energy to give rise to electron-hole pairs on illumination with the solar light. Depending on the thickness of the CdS some of the light below the 510 nm can still pass through to the CdTe giving additional current to the device. The reduction of CdS layer thickness improves the gain of photons in CdTe, but on the other hand the uniform coverage of the SnO2 and the consumption of layer into CdTe during annealing treatment imply not to reduce the thickness beyond a limit; otherwise the cell is either shunted or gives low voltage. CdS layer deposited by low temperature techniques generally requires an annealing treat ment for recrystallization prior to the deposition of the CdTe absorber la yer. The structural a nd optical properties of the layer are strongly depending on the depos ition technique (temperature and energies involved have different values), so th e properties must be matched with the characteristics of the absorber layer with which it has to intermix. Chemical Bath Deposition (CBD) grow n CdS is preferred as window layer because of low optical absorption and good coverage properties on the TCO [7].

PAGE 31

152.6.4 CdTe Absorber Layer The polycrystalline CdTe layer should be electrically p-type to form the p-n junction with n-CdS. It has an energy gap of 1.45 eV, which gives the highest theoretical efficiency. The maximum conve rsion efficiency in dependence of bandgap width for solar spectrum AM1.5 is shown in Figure 2.9, bandgap of 1.45eV is sufficient to absorb the useful part of the solar spectrum. Since CdTe has lower carrier concentration than the CdS, the depletion region is mostly within th e CdTe layer and in th is region most of the carrier generation and collection occur [9]. The conductivity and grain size of the layer depend on the deposition technique and pos t-deposition annealing treatment, which affects also the intermixing of CdS with CdTe. Typically thickness of this layer, depending on the growth method, is betwee n 2m and 10m, with grain sizes from 0.5m to 8m. Figure 2.9 Maximum Conversion Efficiency is Dependence of Bandgap Width for Solar Spectrum AM1.5

PAGE 32

162.6.5 Back Contact Copper is commonly used in the back c ontact to improve de vice performance of thin-film n-CdS/p-CdTe solar cells [10]. Since CdTe has a high electron affinity ( ) and bandgap (Eg), many metals will form a Schottky barrier, resulting in a significant limitation to hole transport from the p-CdTe. In a circuit model, this barrier will form a diode of opposite polarity to the primary junction [9], decreases device efficiency by reducing the fill factor and by limiting the Voc.

PAGE 33

17 Chapter 3 Literature Review 3.1 Development of Efficient and Stable Ba ck Contacts on CdS/CdTe Solar Cells In this study [11] solar cells were developed on closed space sublimated (CSS) CdTe/CdS layers with a CdCl2-treatment. Prior to the deposi tion of the back contact (BC) materials the CdTe surface was etched, either in a dilute solution of bromine in methanol (BrMeOH etch) or a mixture of nitric and phosphoric acid in water (NP etch). This etching cleans the CdTe surface from oxides and other contamination. It smoothens the surface and creates a Te-rich la yer which effectively results in a p+-doped surface layer but besides beneficial effect s, the chemical etching also widens the grain boundaries (GB) and makes the GB-surface Te-rich, see Fi gure.3.1. This is de trimental to the cell stability, because the Te-rich GB surface ma y provide a conducting link between the BC and the pn-junction and the widened GB will enhance the metal diffusion into the CdTe down to the front contact. Figure 3.1 SEM Images of the CdTe Morphology afte r Etching, the BrMeOH Etch (left) Affects the Morphology less Than the NP Etch (right) which Smoothens the Surface and Widens GBs

PAGE 34

18 Long term stability tests were carri ed out in a chamber where the nonencapsulated cells were kept in air with 40 65 % relative humidity. The cells are continuously illuminated in open circ uit conditions with 1 sun (1000 W/m2, halogen light similar to AM 1.5) at 80C cell temperature. All tested cells had initial efficiency in the range from 9.5% 10.5%. The efficiencies of cells with Au or Al metallization or Cu buffer layer degraded severely with time. Th e accumulation of Cu/Au is shown in Figure 3.3, it is explained that th ese metals diffuse under thermal and electric fields and contribute to the shunting of the solar cells in conjunction with the Te rich surface of the GB. Cells with Mo metallization and buffer layers of Sb or Sb2Te3 did not degrade, they even showed some improvement after the in itial illumination in the first few days (Figure.3.2).Long term stability tests at elev ated cell temperatures of 80C indicate that Sb/Mo and Sb2Te3/Mo BCs yield stable cells for whic h the efficiency increases over a long period before slow degradation occurs. Figure 3.2 Accelerated Stability Tests at 65C (left) for Solar Cells with Different Back Contacts and Test of Stable Cells with Mo Metallization at 80C (right)

PAGE 35

19 Figure 3.3 SIMS Depth Profiling of a CdTe/CdS Cell with Cu/Au Back Contact 3.2 Effect of Back Contact Copper Concentration on CdTe Cell Operation In this study[12] Five sets of CdTe ce lls, identical except for the back contact, were selected, a copper containing material was deposited to form back contact to the CdTe absorber, followed by an anneal a nd the application of a carbon paste. The deposition times for the copper material were 0, 0.5, 1, 2, and 4 min. (Two minutes, which adds the equivalent of 20 of copper). A common CdCl2 treatment was used for all cells. Three cells with each copper concentr ation were measured in depth. Comparison was done for light (100 mW/cm2) and dark current voltage (J-V), quantum efficiency (QE), and capacitance volta ge (plotted as (A/C) 2 vs. V, where A is the cell area) curves for the five copper concentrations. Except for the zero-copper set of cells, only the best one is shown, there was good consistency among the three cells from each set. The 2-min and 4-min Cu, have an efficiency just ove r 11%. The lower-Cu one s are progressively lower in efficiency and fill factor. Data trends similar to Figure 3.4 are also seen in CdTe cells with ZnTe:Cu contacts of varying thickne ss [13]. The QE curves in Figure.3.4 were nearly identical for all amounts of copper, as one would expect with changes only at the back contact.

PAGE 36

20 Figure 3.4 Quantum Efficiency as a Function of Wavelength of the Incident Light The capacitance of the zero-Cu cell was i ndependent of voltage, which implies a CdTe layer that is fully depleted (1.9 m). The variations in capacitance of the cells with finite Cu, for the voltage range shown, ar e in the CdTe/contact transition region and imply that the transition is less abru pt with Cu present. If the (A/C)2 vs. V curve (Figure 3.5) is extrapolated to zero (infinite capacitance) at a reasonable built-in potential, its slope implies a CdTe hole density in the mid-1014 cm-3 range, typical of all CdTe cells measured by the CSU group. Figure 3.5 Room-Temperature C-V Cur ves Showed Progressive Deterioration in Fill Factor with Reduced Copper

PAGE 37

21 Figure 3.6 Room-Temperature J-V Cur ves Showed Progressive Deterioration in Fill Factor with Reduced Copper The primary effect on J-V curve as shown in the Figure 3.6 is a much lower FF when copper level is low or zero. Physically Cu in the BC region produces a lower barrier contact junction, whethe r it heavily dopes CdTe or forming a physically distinct layer. In any case the C-V measurements impl y addition of Cu leads to a transition region few tenths of a micron thick with much larg er hole density than the bulk of the CdTe. 3.3 Sb2Te3/Mo, Cu/Mo, Sb/Mo Back Contacts In this study Sb2Te3/Mo was used as the back cont act for cell fabrication [14]. The carrier concentration at th e back contact of the cells is responsible for the barrier height and hence the roll-over in the I-V characteristics. Sb2Te3/Mo, Cu/Mo and Sb/Mo back contacts will be discussed in this section based on the carrier concentration. The carrier concentration at the back contact of cells with Sb2Te3/Mo was found to be highest, which corresponds to lower barri er height and less roll-over observed in the IV-characteristics. The cells with Sb2Te3/Mo back contact were found to have highest carrier concentration and low barrier height. The rollover in the IV-characteristics of such

PAGE 38

22 cells is minimal. The doping profiles of ce lls with Mo metallization are as shown in Figure 3.7. Figure 3.7 Doping Profiles of Cells with Mo Metallization The carrier concentration for the Cu/Mo back contact cells is lower than that of the Sb2Te3/Mo cells however the barrier height is low enough to create a quasi-Ohmic contact. Consequently no rollove r is observed in the IV-charact eristics of these cells. In this case the Cu buffer layer was significantly thinner than the Sb2Te3 buffer layer and the Cu diffuses away from the back contact. In contrast, SIMS measurements give no indication of diffusion for Sb2Te3 into the CdTe. The cells with Sb2Te3 buffer exhibit the highest carrier concentration at the back contact. Cells with Sb/Mo back contact have the lowest carrier concentration and the highest ba rrier, at the same ti me a strong rollover is observed in the IV-characteristics. Results with different vacuum deposition bu ffer layers have been reported. Sb and Sb2Te3 back contact buffer layers with Mo metallization yield long-term stable CdTe/CdS solar cells. Minor degradation in efficiency, particularly for Sb/Mo back contacts, could be due to oxidation of the non-encapsul ated back contact. The metallization with Mo is important for the cell stability, as cells with Au of Al

PAGE 39

23 metallization degrade. The chemical etching of the CdTe might be detrimental for the long-term stability as it widens GBs and provides diffusion paths along GBs and enhances shunting by covering the GB-sur face with a Te-rich layer. Preliminary investigations with an evaporated Te laye r replacing the etching treatment have shown encouraging results. Cells with Te/Sb2Te3/Mo back contact are st able with somewhat lower efficiency. 3.4 Review on Cu/Au, Sb/Au Back Contacts to CdTe Solar Cells and the Effect of Etching on CdTe Layer with Different Etchants Typically back contacts to CdTe solar cells are Cu/Au and Cu/Graphite. But these back contacts affect the stability of the cells, which degrade due to diffusion of copper into the junction from the back contact [ 15]. This can be avoided using copper free contacts such as Sb/Au, Sb2Te3/Mo back contacts etc. Maxi mum efficiency of 12.5% was reported using the Sb/Au as a back contact a nd using the mixture of nitric and phosphoric acid as etchants. The composition of the etchants used by this group is given in the Table 3.1 below. Table 3.1 Composition of the Etchants BrMeOH, NP1 and NP2 In the NP etchant the nitric acid is res ponsible for the differential removal of the Cd from the CdTe surface. The concentration of the nitric acid is NP2 is 34% more when compared to the NP1 etchant solution. NP2 is the most aggressive etchant that produces a

PAGE 40

24 Te rich layer in about 10 seconds of dipping time. The optimum etching times reported for NP1 and NP2 are 60 and 10 seconds respectively. The BrMeOH etchant is mild and the influence on cell effici ency is not so pronounced. The Figure 3.8 gives the graph between the efficiency of the cells with respect to etching times. Figure 3.8 Effect of Etching Time on the Efficiency Table 3.2 Sheet Resistances with Different Etching Times in NP1 Table 3.2 gives the sheet resistance report ed with different etching times with NP1 solution. The sheet resistance was measur ed with the 4-point probe technique. From the above table it can be seen that etchi ng produces a Te rich surface with higher conductivity. The sheet resistance of the CdTe surface is very higher but decreased to 53 k-ohm when etched with NP2 for 12 seconds It was also repor ted that for BrMeOH

PAGE 41

25 etched samples the sheet resistance is very hi gh to measure. The fo rmation of a Te layer with etching was measured with XRD measurements (Figure.3.9). Figure 3.9 XRD Pattern of the CdTe Layer. The Te (1121) Peak Indicates the Production of Hexagonal Elemental Te at the CdTe Surface with Etching Time An optimum annealing of the Cu/Au back contact is essential to obtain a good cell. Figure 3.10 shows the eff ect of annealing on the efficien cy of solar cells with Sb/Au (NP2 etched) and Cu/Au (BrMeOH etched) contacts. Figure 3.10 Effect of Contact Annealing on Cell Performance

PAGE 42

263.5 Effect of CdCl2 Treatment The CdCl2 treatment has long been known to improve the efficiency of devices. Numerous studies have shown that it results in changes in crystal texture, grain size and orientation, strain and defect distribution have all been studied [8]. The effects of the CdCl2 treatment are: Induces grain growth and recrystallization (depending on the deposition process). Improves the intermixing of CdS and CdTe, and formation of CdS1-XTeX. Increases minority carrier lifetime. Lowers series resistance. For a reliable and reproducible process it is very important to control the intermixing between the CdS and CdTe and the size of the CdTe grain boundaries. The CdCl2 treatment permits to do this and moreov er increases the conductivity of p-CdTe. The CdCl2 is evaporated on the CdTe and then a nnealed in oxygen at temperature in the range of 400 to 500C. This procedure changes the structural and elec trical properties of the absorber layer, and enables the solar cell to increase its efficiency. The CdCl2 treatment increases the grai n size and also causes the widening of grain boundaries. Wider grain boundary regions are not desired b ecause they enable conduction across the grain boundaries and also cause the shunting of solar cells by providing a conducting link between the top and bottom layers. For high effi ciency and stable cells large grains and compact CdTe is required. 3.6 Stress Testing Stress tests performed as a function of bias, temperature, illumination, and atmospheric conditions attract much interest. Some degradation processes are believed to

PAGE 43

27 be thermally activated, and therefore hi gh temperatures are used to accelerate degradation. Other stress st udies consider the effects of light and varying bias. 3.6.1 Stability of CdTe/CdS Thin Film Solar Cells In this study [10] effect of copper containing cells we re examined from both the diffusion and electronic points of view. SI MS measurements confirmed the diffusion copper into the CdTe from Au /Cu back contacts during the metallization process in which contact annealing is done at 150C in Ar ambient for 90 min. The SIMS results showed a consistent amount of copper near CdTe/CdS interface. There was no significant difference in the copper profile befo re and after post-contact anneal. In attempt to study the Grain boundaries(GB) effect, CdTe films with different grain sizes were tested, SIMS measurements showed the incorporation copper into CdTe film of the conventional cells was more than the large grain size devices whereas the Cu was un detectable in single-crystal CdTe devi ce. This implies that the GBs are the main path for Cu diffusion. Figure 3.11 Comparison of Cu SIMS Profiles in Thin Films of Polycrystalline CdTe for Samples with Different Degrees of Crystallinity

PAGE 44

28 The paper also discusses about comparing the effect of Ni and Ti metallization on Cu diffusion From the Cu:ZnTe contact into the thin film polycrystalline CdTe film, deposited by the CSS method and back contac ts deposited by R.F. magnetron sputtering at ~300oC. The Figure 3.12 shows the SIMS prof ile for copper concentration on a noncontacted CdTe device and sing le crystal CdTe substrate. From the SIMS profile we can see that back contact depletion is noticed with Ni, but not in Ti metallization. Accumulation of Cu near CdS interface can be seen for both back-contacted cells. Figure 3.12 SIMS Profile Comparing the Effect of Ni & Ti Metallization From the SIMS profiles in Figure 3.13 for Cl, S, Sn and Te in a polycrystalline CdTe/CdS/SnO2: F structure, where CdTe is deposited by the CSS method, CdCl2 treatment at 360C for 1 hr (CdCl2 source temperature was 400C) clearly indicate an accumulation of Cl near the CdS interface corre sponding to the increase in S content and decrease in Te content. The amount of Cl accumulated near the CdS interface increased with higher temperatures and longer treatment times, while the Cl content remained constant within the CdTe layer.

PAGE 45

29 Figure 3.13 SIMS Profiles for Cl, S, Sn and Te The Figure 3.14 shows the schematic illustra tion of the behavior of Cu and Cl inside the CdTe/CdS solar cells. We can see the formation of Cu2Te at the back contact/CdTe interf ace, GB diffusion of Cu from Cu2Te to the intermixed region and CdS layers, accumulation of Cu and Cl ions in the GBs in this region and further GB diffusion may enable the ion penetration to the TCO interface. Figure 3.14 Schematic Illustration of Cu and Cl within CdTe/CdS Solar Cells

PAGE 46

30 Diffusion of Cu in polycrystalline CdTe ma y be expected to be faster than in single crystals, with GB proposed to be suitable pathways for rapid diffusion. Generally, diffusion along GB is more than through the bulk of a grain because surface bonds are weaker than bulk ones due to incomplete coordination. In many cases GB diffusion is also easier than diffusion along grain surfaces, because at GB, two surfaces are available for temporary coordination of the diffusing sp ecies, rather than one, along grain surfaces. The ability of Cu to move under the influenc e of electric fields, for example like that present at a p-n junction, may also enhance Cu diffusion within CdS/CdTe cells. After stress the light and da rk J-V curves of cells wi thout Cu showed less rollover at forward bias compared to cells with Cu. Cells when rested in dark at room temperature for 6-12 months showed some recove ry in behavior i.e. increase in Voc and loss of J-V roll-over, but cells without Cu exhibited less recovery in storage. This suggests that under certain circumstances Cu can return to the CdTe surface/back contact interface. During stressing, under illumination at 100oC for 1000 hrs, Voc and FF decreased faster when the Cu content was increased, while the onset of J-V roll-over was faster for cells with lower Cu content. We note that these observations suggest that rollover is indeed an indication of low Cu (accep tor) density near the back contact and that excessive Cu near the junction, or front (CdS side) of the cell, resu lts in degradation of cell performance 3.6.2 Analysis of Stress Induced Degradation In this study[16] accelerated stressing of CdS/CdTe solar cells at elevated temperatures (60-100C) under a range of appl ied bias in light and dark has identified three degradation modes: formation of a blocking contact, increased junction

PAGE 47

31 recombination, and increased dark resistivit y. Devices with Cu-cont acts degrade with a strong bias dependence. The blocking contac t is formed under forward bias. Junction degradation is observed when stressed at hi gher temperature and forward bias. They also showed that re-contacting th e device after stress removes th e blocking contact without change in junction losses. Figure 3.15 Light and Dark J-V Curves for CdS/ CdTe Solar Cells: Initial; 6 Weeks Stress at 100oC at OC; and Following Recontacting Devices without Cu in the contact have much poorer initial performance but degrade nearly independent of bias. The relative change in efficiency after stress as the ratio of efficiency after stress/before stress, for devices stressed at RB, SC, MP or OC for 10 days at 100C in dry air. Devices with wet or dry contact s show similar changes with bias. Stress at SC always shows the least de gradation in performance while stress at OC has the greatest degradation. Devices without Cu degrade near ly independent of bias, and they have much less degrada tion at forward bias compared to devices with Cu. J-V curves for devices biased at RB and SC behave similarly, while those at MP and OC have

PAGE 48

32 a different but common degrad ation mode. We simplify our discussion by considering two bias points, OC and SC. Figure 3.15 shows light and dark JV for devices stressed at OC and SC. Figure 3.16 Relative Change in the Efficiency of Devices with Different Contacts after Stressing at Bias in Light for 10 Days at 100oC Figure 3.17 shows FF and Voc for devices with and without Cu for stress in the dark for 10 days at bias from 0 to +2.5V at 60C. There was negl igible degradation for stress from 0 to 1V but signi ficant change occurs for bias greater than 1V. Voc and FF at 0V stress bias are essentially the same as the initial values. Devices with Cu show a strong bias dependence, with Voc and FF decreasing simila rly under forward bias. In contrast, devices without Cu degraded very little at any bias under these conditions.

PAGE 49

33 Figure 3.17 FF and Voc for Devices with 0 or 6 nm Cu for Stress under FB in the Dark at 60oC for 10 Days 3.6.3 Determination of Cu in CdTe/CdS Devi ces Before and After Accelerated Stress Testing In this study [17] the distribution of Cu before and after accelerated stress testing of CdTe/CdS solar cells has been studied using SIMS analysis. Back contacts were graphite paste with varying amount of Cu added in the form of HgTe:Cu powder. The back contacts were applied after NP etch. In one device no intenti onal Cu was added to the graphite paste. The devi ces were stressed at open circ uit in light at 100C for 1000 hours. SIMS analysis was done on both the stressed samples and the companion samples which where not subjected to stress. The stressed samples had Cu accumulated near the CdS and the amount of copper accumulated was comparable to the amount of Cu present in the graphite paste and the Cu level in CdTe was shown to be relatively constant as shown in Figures 3.18 and 3.19 The data in Figure 3.20 clearly show that after the cells are stressed, there is a significant increase in the Cu concentration in the CdS laye r, with the amount of Cu strongly correlated with the Cu concentration in the back co ntact. A second striking result

PAGE 50

34 obtained from the data is the level of Cu in the CdTe itself is relatively unperturbed by the diffusion of Cu to the CdS. Figure 3.18 SIMS Profile Obtained by Profiling from the CdTe (Back) Side of the 0.8% Stressed Device Figure 3.19 SIMS Profile Obtained by Profiling Through the SnO2, (Front) Side of the Device The data for the stressed Cu contacts a nd the stressed graphite-only contact are shown together on one graph in Figure 3.21. This figure clearly illustrates the dramatic increase in Cu in the CdS and the relativ ely minor increase in the CdTe layer. The amount of Cu in the CdS shows a strong correlation with the Cu concentration. It is clear from Figure 3.22 that a sma ll amount of Cu reaches the CdS layer simply from the steps necessary to process the contact, and that the concentration is proportional to the Cu in the back contact. In all cases, the unstressed devices with Cu in

PAGE 51

35 Figure 3.20 SIMS Profile of Stressed and Unstre ssed Devices with Different Cu Concentrations Figure 3.21 Comparison of Cu Depth Profiles from Devices with Different Cu Concentrations in the Contact After Stress the back-contact had higher Cu levels than the stressed graphite-only device. There are several possible explanations for the presen ce of Cu in the graphite-only contacted

PAGE 52

36 device. First, the CdS may have been contaminated with a small amount of Cu. Second, we have observed that some CdCl2, treatment processes may introduce small amounts of Cu into the device [18]. Third, it is typical that the graph ite paste contains a low level of a Cu contaminant [19] Figure 3.22 Comparison of Cu Depth Profiles from all Unstressed Contacted Devices and Also Stressed Graphite-Only Devices 3.6.4 Thermal Stressing In this study[21] CdTe solar cel ls used in this work had glass/SnO2/CdS/CdTe/back contact structure consis ting of a borosilicate glass substrate, SnO2, approximately 1000 of CdS deposited by the chemical bath deposition process (CBD), and CdTe deposited by close-spaced subl imation (CSS) to a thickness of 4-6 m. After the CdTe deposition, the structures were heat treated in the presence of CdCl2. The back contact was graphite paste doped with HgTe:Cu, applied after a brief etch of the CdTe surface in a 0.1% bromine-methanol solution. Four samples were analyzed at the Na tional Renewable Energy Laboratory using SIMS: (a) an uncontacted CdTe/CdS structure, (b) a contacted but unstressed cell, (c) a cell contacted and stress ed at 70C, and (d) a cell contact ed and stressed at 100C. Both (c) and (d) were stressed for approximately 1500 hours.

PAGE 53

37 Figure 3.23 compares SIMS depth prof iles between uncontacted and contacted devices. From the figure we can say that the level of copper in contacted device is more than uncontacted device indicating penetr ation of Cu from the back contact. Figure 3.23 SIMS Depth Profiles for CdTe Cells Contacted vs. Uncontacted (Unstressed) From the results shown in Fi gure 3.24 in all cases the rise in Cu concentration at approximately 4-5 m coincides with the Cd Te/CdS interface, showing accumulation Cu near CdTe/CdS interface. Figure 3.24 Stressed vs. Unstressed Sample (Control)

PAGE 54

38 Chapter 4 Experimental Techniques 4.1 Experimental Techniques To find out degradation mechanisms in solar cells current voltage (I-V), monochromatic J-V, Capacitance voltage(C -V) and Spectral Response (Q-E) are used. 4.2 Current-Voltage Measurements Solar cells are characterized by currentvoltage measurement in dark and light. Light current-voltage measurements are pe rformed using Keithley 2410 sourcemeter. The measurements are conducted with a curre nt and voltage input and output system controlled by a computer. Figure 4.1 I-V Curve for an Ideal Solar Cell The Current Vs Voltage data in dark and light describe the ba sic properties of the solar cell, which include the open circuit voltage (Voc), short circuit current (Jsc), series Resistance (Rse), Shunt Resistance (Rsh), Maximum Power points (Jm, Vm) and Fill

PAGE 55

39 Factor (FF). The standard condi tion for measuring I-V curves is with the light intensity equal to one sun and air mass of AM1.5 (s hown in Figure 4.2), or with sun at 48.2o from the overhead. Figure 4.2 A.M 1.5 Condition 4.3 Capacitance-Voltage Measurements Capacitance-Voltage (C-V) and Capacitance-frequency (C-F) measurements were done using HP impedance analyzer 4145A. Cap acitance Voltage Technique (C-V) is used to find out the carrier concen tration. Derivation of doping density using Capacitance Voltage technique relies on a junction with large carrier concentration on one side. The depletion region prominently ex tends in the semiconductor of low carrier concentration. As there is a dependency of capacitance on frequency, the operating frequency was chosen such that stray capacitance is avoided. The frequency was kept at 106 Hz. The oscillating voltage was 10 mV. The sweep range was from -2.0 V to 1 V. The capacitance frequency (C-F) measurement is used to iden tify traps that may exist in the device.. As there is a dependency of capacitance on fre quency, the operating frequency was chosen such that stray capacitance is avoided. The frequency was kept at 106 Hz. The oscillating

PAGE 56

40 voltage was 10 mV. The sweep range was from -2.0 V to 1 V. The capacitance frequency (C-F) measurement is used to identify traps that may exist in the device. The frequency was varied from 100 Hz to 106 Hz. 4.4 Monochromatic J-V Measurements Wavelength specific J-V measurements pr ovide valuable information about the collection related problems in solar cells. Th e measurement is carried out in a black box which is shielded from stray light and pa inted black on the inte rior. An EHP lamp is used as the light source. Different wavelengt hs can be chosen by using various band pass filters having a band width of 20 nm. The filters are arranged in order on a rotating disc. The disc is motor driven which in turn in connected to a computer. A linearly graded intensity filter is used to adjust the intens ity of light. The light from the source passes through the intensity filter and then through th e wavelength filter and falls in the device. Proper shielding is provided to eliminate all st ray white light. The intensity filter is also coupled to a computer controlled motor. Rota ting the intensity filter varies the intensity. The Isc of the device for bandwidth of di fferent wavelengths are determined from spectral response measurements. The intensity of light is adjusted for this Isc and the intensity can be varied to simulate different AM conditions. 4.5 Quantum Efficiency Measurements Quantum Efficiency (Q-E) is a measure of the wavelength specific output current from a solar cell, under short circuit conditio ns, per unit incident power. Measurements take place in monochromatic light, and ofte n with additional white, bias, light. The monochromatic light is chopped an d the variation of the current signal is detected with a lock in amplifier. When performed with bias light, the measurement takes place under

PAGE 57

41 conditions similar to actual operation. Howeve r, Q-E measurements are also performed with only monochromatic light present. In any case, Q-E can be thought of as a measurement of the number of el ectrons flowing in the device circuit per incident photon. Spectral analysis for desired bandwidth of wavelengths of photons was done using Oriel Cornerstone monochromator (model 74100) w ith light source of GE 400W/120V quartz line lamp (model 43707). Silicon reference was used to adjust the light intensity prior to measurement. 4.6 Series-Resistance Measurements Solar cell principal electrical parameters su ch as short-circuit current density Jsc, open-circuit voltage Voc, maximum power Pm, fill factor FF and maximum conversion efficiency max are decreasing functions of solar cell series resistance Rs [22]. Hence it is very important to find out reas on for increase in Rs and try to minimize Rs and improve solar cell performance. Proper technique is essential to determine this parameter. Relevant literature proposes va rious techniques to derive Rs [ 22]. In this Thesis work Rs is obtained from the dV/dJ Vs Voltage curv e. Figure 4.3 shows as an example for the curve used to obtain the Rs value. The valu e at high voltages is taken the Rs value.

PAGE 58

42 3-22-A23-6501.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 0.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ Light dV/dJ Figure 4.3 dV/dJ Vs Voltage Plot to Obtain Rs 4.7 Shunt-Resistance Measurements Shunt-Resistance (Rsh) is also calculate d from the dV/dJ Vs Voltage curve and the value is taken at negative bias, slope of th e variation in Rsh is taken as the required value. Figure 4.4 shows an example for the plot to obtain the Rsh value. 3-22-A23-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 1.E+09 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ Light dV/dJ Figure 4.4 dV/dJ Vs Voltage Plot to Obtain Rsh 4.8 Transient Measurements The other topic discussed in the results chapter is Transi ents. This is the variation in Voc and FF within the 4 hour light or dark cycle. Variation in Voc and FF followed a trend while increasing and decreasing within 4 hour dark/light cycle.

PAGE 59

43 Temperature dependence of FF30.0 35.0 40.0 45.0 50.0 55.0 60.0 65.0 1214161820 Time(hours)FF OC sc-b Figure 4.5 Transient Behaviors in FF (4 Hour Dark Cycle is Not Shown in the Figure) Figure 4.5 shows an example of the transient curve. The X-axis of the plot represents number of light soaked hours. Ther e is 4 hour dark cycle transpired in between but not shown in the figure, i. e the light cycle started at 12 hours completed by 16 hours after this light cycle, dark cycle for 4 hours fo llows and then light cycle follows after next 4 hours. The dark cycle is not shown in the figure.

PAGE 60

44 Chapter 5 Stability Testing 5.1 Objective The motivation for this research was to study the degradation in standard high efficiency CdTe/CdS solar cells under light soaking for approximately 700 hours. Back contacts for the cells were varying thickness of sputtered Cu, followed by application of undoped graphite. Different techniques were tested to obtain optimum position for Cu deposition. Figure 5.1 show different cell stru ctures tested in USF solar cell laboratory. Figure 5.1 Different Techniques for Cu Sputtered Samples Figure 5.2 shows the structure of Cu sputtered over CdTe followed by undoped graphite, which was selected for the stability analysis.

PAGE 61

45 Figure 5.2 Selected Sample 5.2 Processing Conditions 1. SnO2 is deposited by MOCVD (8+5+5min) and have a thickness of ~1000 2. CdS is deposited by CBD of 80min @80oC/85oC and have a thickness of ~1000 3. CdTe for all the samples is deposited by CSS for 5 minutes and have thickness of ~5 m. 4. CdCl2 of 8000 is deposited using evaporat ion, and the samples are annealed at 390oC for 25min. 5. Bromine-methanol etch was performed 6. Cu sputtered thickness was varied in the back-contact of the samples. 7. Un-doped graphite was applied. 8. The contact annealing temperature was varied from 175oC to 275oC for different samples. 9. All the samples had 7059 glass/SnO2/CdS (CBD)/CdTe (CSS)/CdCl2 treatment/Sputtered Cu/ un-dope d graphite structure.

PAGE 62

46 10. The contacts were applied af ter Bromine-methanol etch. 5.3 Stability Testing Procedure The major task of the stability testing is to evaluate the impact of the Cu sputtered thickness and contact anneali ng temperatures on the CdTe solar cell stability. The variables in device fabricati on were contact annealing te mperature and Cu sputtered thickness. The objective was to light soak devices. The different contact annealing temperatures varied from 175 to 275oC, whereas the Cu sputtered thickness varied from 5-80. From this group of devi ces two sets have been chosen for analysis, The first set constitutes of varying sputtered Cu thickne ss(10, 20 and 40),cont act annealed at 240oC and the second set constitutes of different contact annealing temperatures(175, 200 and 240oC) with 10 Cu in the back contact. The light soaking set up is shown in Figure. 5.3. The sample bed consisted of a Cu plate that was cooled using city water, temperature variations at different points on the sample bed are controlled by using Cu tube circulating city water underneath the Cu plate. The cells were placed on the copper plate in such a way that the light enters through the 7059 glass (substrate) and the back contact was facing the copper plate. The copper plate was covered with a thermal compound which in turn is covered by two layers of thin insulating film to avoid direct contact between the cells and the thermal compound. The thermal compound is used to tran sfer heat from the de vices to the copper plate and the copper plate is cooled by water circulating in the water line inside the chamber. The cells were pressed firmly agai nst the insulating film using an LOF glass plate. The light intensity was calibrated to approximately AM1.5 conditions (%).

PAGE 63

47 Figure 5.3 Schematic for Light Soaking Setup In order to record the temperature insi de the oven shown in Figure 5.5 CdTe substrates, were bonded to thermocouples and placed at three different locations below LOF glass plate and are used as temperature sensors. The water flow was also controlled by solenoids in turn connected to the se nsors to ON and OFF at pre determined temperatures. (If the temperature reaches 35oC the water line turns ON automatically). The entire set up as depicted in Figure. 5.3 was placed inside a vacuum oven. Prior to the start of the light soaking cycles, the oven was evacuated and back filled several times with N2 and a small constant flow of N2 was maintained through the oven, using a pressure relief valve keeping the oven at slightly positive pressure. The light soaking process was a 4 hour ON/4 hour O FF cycle in order to identify any transient mechanisms. The light intensity varied by approximately 10-15% over the area of the sample bed. The temperatures on the platform could reach 25-70C depending on the location of the thermocouples.

PAGE 64

48 Nine (identical) substrates were stressed. Each substrate had four cells as shown in the Figure 5.4 and were named as a, b, c and d. Cells a and c were held at opencircuit (OC) and cells b and d were held at short-circuit (SC) conditions. Current-voltage measurements (dark and light ) were taken during a se t of pre-determined intervals. Each light cycle of 4 hours was fo llowed by a 4 hour dark cycle. The cells were light soaked for approximately 700 hours (i.e 700 hours under light and 700 hours in the dark). Figure 5.4 Top View of a Substrate The stability measurement is performed in a vacuum chamber which can be evacuated and back filled with ultra high purity N2. Standard pumping and purging procedures were used to reduce gas phase co ntamination such as water vapor from the stress chamber before the stress treatment. The chamber was filled with ultra high purity N2 gas prior to light soaking creating nitrogen ambient for light soaking. Each substrate had 4 cells and using conduc tive silver epoxy treatment for 5 min at approx 90oC leads are connected to front and back contact of each cell. The front and back contact leads of all the cells were connected to the Ke ithley Source Meter using low resistance copper electrical wires. The sample s were stored in vacuum desiccators before starting the light soaking to protect the samples from the humidity.

PAGE 65

49 Figure 5.5 Schematic of the Stability Set-Up The samples where stressed under one s un (AM1.5) illumination intensity. The light and dark measurement was taken in the interval 4 hours each. The light soaking is enabled by 20 MR16 lamps (12V, 71W). Two rows of lamps are connected with 10 lamps in each row. The front and back row la mps are aligned so that they illuminate the samples placed on the Cu plate. A clean sodalime glass was used to press the samples to the Cu plate for better heat transfer. Quartz plates where placed between the samples and the lamps, they act as diffusers and enabled a uniform distribution of light to all the cells. Thermocouples were used to continuously monitor the temp erature inside the chamber and Euro therm controllers are used to cont rol water flow. Specific time intervals were chosen to perform the I-V measurements A Keithley 2400 source meter was used to sweep the voltage from -2V to +2V in step s of 0.01V and current was measured. An automated system was used to collect data from the J-V measurements and Omega 44-M

PAGE 66

50 data logger is used to mon itor temperature and humidity wh ich contains an internal temperature and humidity sensors and two external thermistors. 5.4 Obstacles During The Stress Period Light soaking was interrupted several times due to the failure of lamps, but lamps were fixed immediately and th e experiment was continued. The estimated light soaking period was 1000 hrs but due to major water leak after 700 hrs the experiment was completely stopped.

PAGE 67

51 Chapter 6 Results and Discussion In this chapter results of stress tests have been discussed. The stress tests are performed at Open Circuit (OC) and Short Circuit (SC) cond ition. Each sample had four cells out of which two cells were stressed at OC and other two cells were stressed at SC. The group of devices selected for stability an alysis is listed in the following table. Stability studies have been focused on Cu sputtered thickness and contact annealing temperature. A set of devices w ith contact annealing temperatures varying from 175 275oC and copper sputtered thickness va rying from 5-80 have been fabricated. The cells were light soaked in an oven under illumination of approximately 1sun intensity in inert N2 atmosphere for approximately 700hrs. In Table 6.1 different samples selected fo r stability studies have been listed. Table 6.1 List of Devices Selected for Stability Analysis Contact Annealing Temperature(oC) Cu Sputtered Thickness() Sample 175 5 3 22A 28 175 10 3 22B 28 200 10 3 22A 22 240 10 3 22 A 22 240 20 3 22B 22 240 40 3 22 B27 240 40 3 22A 23 275 15 3 22A 15

PAGE 68

52 Two sets of samples have been chosen to perform analysis First set constitutes of different sputtere d Cu thickness (10, 20 and 40) in the back-contact and contact a nnealing was performed at 240oC. Second set constitutes of samples ann ealed at different contact annealing temperatures (175,200 and 240oC) with 10 of sputtered Cu in the back contact. 6.1 Set 1(10, 20 and 40 of Cu, Contact Annealed at 240oC) 6.1.1 Initial Performance of Se t 1, Measured at Room Temp erature Using Simulator In Table 6.2 initial performance of set 1, measured at room temperature is listed. Table 6.2 Initial Performance of Set 1 contact annealing temperature(oC) Cu () sample# Voc FF Area(cm2) Avg Voc Avg FF 240 10 3 22A 22 830 68.2 0.4897 822.5 66.65 820 68.3 0.4369 820 65.3 0.3904 820 64.8 0.1547 240 20 3 22B 22 810 73.9 0.5247 812.5 67.225 810 66.7 0.5126 820 67.6 0.4938 810 60.7 0.4433 240 40 3 22A 23 790 67.6 0.4608 790 65.65 790 69.1 0.4462 790 56.4 0.4756 790 69.5 0.4251 Initial Voc variation780 790 800 810 820 830 01020304050Cu Sputtered thickness()Voc(mV) Initial Voc Initial FF64 66 68 01020304050 Cu sputtered thickness()FF Initial FF Figure 6.1 Variation in Voc and FF with Cu Sputtered Thickness (Back-Contact Annealed at 240oC)

PAGE 69

53 Initial performance of the device s is shown in the Figure 6.1, we can see that sample with 10 Cu sputtered thickness had better performance initially. 6.1.2 Degradation in Voc and FF with Time Figures 6.2, 6.3 and 6.4 show variation in Voc and FF for set 1, with respect to time. The sample title is listed is in the format contact annealing temperature-Cu sputtered thickness-sample number. Figure 6.2 Degradation of Voc and FF at OC & SC for 10 Thick Cu Sputtered Back Contact Annealed at 240oC Figure 6.3 Degradation of Voc and FF at OC & SC for 20 Thick Cu Sputtered Back Contact Annealed at 240oC

PAGE 70

54 Figure 6.4 Degradation of Voc and FF at OC & SC for 40 Thick Cu Sputtered Back Contact Annealed at 240oC The degradation in the samples with more amount of Cu is more from the Figures 6.2, 6.3 and 6.4.This results are comparable to the results discussed in sections 3.6.1 [10] and 3.6.3 [17] where the degradation varied wi th the amount of Cu in the back contact. According to the literature review in section 3.6.1 and section 3.6.3 the degradation with more amount of Cu in the ba ck contact is more due to the formation of Cu2Te following the deposition of back-contacts, fast grain boundary (GB) diffusion of Cu from the surface telluride resulting in depletion of Cu2Te with stress testing and accumulation of Cu in CdS. It is clearly shown in Figure 3.12 and Figure 3.20 using SIMS analysis that the amount of Cu in CdS region varied propo rtionally with the increase in Cu content in the back contact. This might be the reason for the degradation being more for the samples with 40, as shown in Figure 6.4. Due to increased concentration in the back contact more amount if Cu must have accumulated in CdS and CdS/CdTe junction degrading the junction properties which eventually lead to degradation in Voc and FF.

PAGE 71

556.1.3 Variation in Voc and FF at OC and SC Conditions The stress condition chose in this study ar e OC and SC conditions. Each sample had four cells as shown in Figure 5.4, out of which a and c are stressed at OC and b and d are stressed at SC. Figure 6.5 shows the degradation in Voc and FF at OC and SC for 10 sample. 300 350 400 450 500 550 600 650 700 750 800 Voc Degradation(10A) initial 779.953779.953779.964 0.08 739.971719.986739.965 150 640.031660.061680.049 350 660.066660.06700.048 450 620.021640.047680.067 640 640.033640.035700.041 acb 30 35 40 45 50 55 60 65 70 FF degradation(10A) initial 67.600164.547766.2961 0.08 67.089765.733263.5372 150 58.922158.049556.7196 350 54.361756.213754.4548 450 54.429456.122255.1324 640 52.672855.301848.7916 acb Figure 6.5 Variation in Voc and FF for 10 of Cu Sputtered, Back Contact Annealed at 240oC It clearly indicates that there was decreas e in Voc and FF after 640 hrs. Initial data in the plots indicate the measurement taken in itially inside the stability oven with the desired stress conditions. .08 data in the Figure 6.5 is the measurement taken after 15 min of light soaking. As we can see from the figure within first 15min there was approximately 5% decrease in the Voc of th e devices stressed at OC and SC condition. There was either 1% improvement in the de vices stressed at OC or the FF remained constant. For the cells stressed at SC, FF reduced approximately 3%.

PAGE 72

56 300 400 500 600 700 800 Degradation in Voc(20A Cu) initial 779.947779.954759.938779.959 0.08 739.979719.985719.983719.986 150 559.982579.888680.056539.935 350 579.981600680.053579.979 450 539.941559.966660.057539.947 650 579.979600.01680.059579.981 acbd 30 35 40 45 50 55 60 65 70 Degradation in FF(20A Cu) initial 68.411462.04860.953560.4556 0.08 66.570562.408955.641961.3572 150 53.621444.937451.08755.8946 350 52.446549.998551.792453.9867 450 53.599350.342151.734353.9564 650 49.999847.127349.400351.9397 acbd Figure 6.6 Variation in Voc and FF for 20 of Cu Sputtered Back Contact Annealed at 240oC 300 400 500 600 700 800 Degradation in Voc(40A Cu) initial 759.935759.935759.93779.957 0.08 680.025700.013680.025700.015 150 440.022380.061579.978599.996 350 539.942499.98620.021640.042 450 460.015380.067579.991579.982 650 559.962480.001600.001620.01 acbd 30 40 50 60 70 Degradation in FF(40A Cu) initial 62.2153.3961.4965.69 0.08 52.290853.296843.832645.2713 150 49.698347.606938.574941.9925 350 56.161655.707837.782640.354 450 49.587149.613636.749242.5811 650 57.030555.536935.143138.1684 acbd Figure 6.7 Variation in Voc and FF for 40 of Cu Sputtered Back Contact Annealed at 240oC In Figure 6.6 the Figure 6.7 and degradat ion of Voc and FF in the device with 20 and 40 of Cu in the back contact has been pres ented. It clearly indicates that there was decrease in Voc and FF after 640 hrs. Except for 40 where there was improvement in the FF. Within the first 15 min of light soaking we can see from the Figure 6.7 that there was approximately 10% decrease in the Voc of the devices stressed at OC and SC

PAGE 73

57 condition and there was 30% degradation in FF for the devices stressed at SC. The degradation in Voc at OC was more a nd degradation in FF at SC was more. 6.1.4 Comparison of Degradation in Voc at 10, 20 and 40 0 5 10 15 20 25 30 35 Degradation in Voc 240-10 13.511.15.39 240-20 21.6216.665.5419.44 240-40 17.6631.4311.7711.43 acbd Figure 6.8 Variation in Voc for 10, 20, 40 of Cu Sputtered Back Contact Annealed at 240oC In Figure 6.8 the percentage degradation of Voc in the devices with 10, 20 and 40 of Cu in the back contact stressed at OC and SC has been presented. It clearly indicates that after 640 hour s of light soaking the Voc degraded for 10, 20 and 40 samples stressed at OC and SC and the degrad ation was more at OC. This result was in agreement to the result disc ussed in section 3.6.2[14]. The degradation in Voc may be due to the formation of Cu2Te at the back-contact after the deposition of Cu on the Te rich la yer, formed due to BrMeOH etch performed prior to the Cu deposition. In itial performance of the cell s can be attributed to the formation of Cu2Te layer, hence degradation in Voc mi ght be due to diffusion of Cu into the junction depleting Cu2Te, hence decreasing Voc.

PAGE 74

586.1.5 Comparison of Degradation in FF at 10, 20 and 40 -10 -5 0 5 10 15 20 25 Degradation in FF 240-10 21.4815.8623.2 240-20 24.8924.4811.2115.34 240-40 -9.06-4.219.8215.69 acbd Figure 6.9 Variation in FF for 10, 20, and 40 of Cu Sputtered Back Contact Annealed at 240oC In Figure 6.9 percentage degradation of FF in the devices with 10, 20 and 40 of Cu in the back contact has been presented. It clearly indicates that after 640 hours of light soaking the FF degraded for 10 and 20 samp les stressed at OC whereas improved for 40 sample. For samples stressed at SC degradation was more at OC condition 6.1.6 J-V Characteristics for Samples Before and After the Light Soaking In this section light and dark J-V charac teristics of samples with 10, 20 and 40 Cu in the back contact have been shown. In each J-V plot initial performance and performance after 640 hrs of light soaking has been shown. The OC and SC stress conditions have been shown separately.

PAGE 75

59 6.1.6.1 Sample with 10 Cu in Back Contact and Contact Annealed at 240oC -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs Figure 6.10 Light J-V at OC for 10 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs Figure 6.11 Light J-V at SC for 10 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs Figure 6.10 and 6.11 show the light J-V ch aracteristic of 10 sample, in Figure 6.10 we can see there is slight shift in the Voc this might be due to degradation in the back-contact due to the diffusion of Cu into the junction depleting the Cu2Te hence decreasing Voc. In Figure 6.11 when stresse d at SC, we can see lowering in FF due to collection and increase in series resistance. Bad Sample

PAGE 76

60 -2.E-02 0.E+00 2.E-02 4.E-02 6.E-02 8.E-02 1.E-01-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial after 640hrs -2.E-02 0.E+00 2.E-02 4.E-02 6.E-02 8.E-02 1.E-01-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial after 640hrs Figure 6.12 Dark J-V at OC for 10 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs -2.E-02 0.E+00 2.E-02 4.E-02 6.E-02 8.E-02 1.E-01 -2.0 0 -1.5 0 -1.0 0 -0.5 0 0.000.501.001.502.00 Voltage [Volts]Current Density [A/cm2] Initial after 640hrs Figure 6.13 Dark J-V at SC for 10 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs In Figure 6.12 we can see there is a shift in Voc for samples stressed at OC and from Figure 6.13 we can see that there is incr ease in series resistance for samples stressed at SC. The shift in Voc at OC can be due to diffusion of Cu from the back-contact Bad Sample

PAGE 77

61 deteriorating the back-contact properties. Increa se in series resistance can be seen for samples stressed at SC condition. Figure 6.14 Dark and Light J-V at OC (Left) and SC (Right) Figure 6.15 Semi-Logarithmic Dark J-V at OC (Left) and SC (Right) In Figure 6.14 we can see cross-over between dark and light J-V curves, degradation at SC is less than OC, but de veloped a large cross-ove r. The cross-over can be due the photo-conductivity of CdS. Cd S being a photo-conductive compound is very sensitive to the light and at the same time pr oduces very high electri cal resistance in dark condition. Therefore we can see increase in series resistance during dark cycle resulting

PAGE 78

62 in cross-over between light and dark J-V. Fr om Figure 6.15 when stressed at SC (right), we can see ohmic losses which can due to de terioration in back-c ontact ohmic property resulting in increase in series-resistance. 6.1.6.2 Sample with 20 Cu in Back Co ntact and Contact Annealed at 240oC -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 650hrs -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs Figure 6.16 Light J-V at OC for 20 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs Figure 6.17 Light J-V at SC for 20 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs

PAGE 79

63 -2.E-02 0.E+00 2.E-02 4.E-02 6.E-02 8.E-02 1.E-01-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial after 640hrs -2.E-02 0.E+00 2.E-02 4.E-02 6.E-02 8.E-02 1.E-01-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial after 640hrs Figure 6.18 Dark J-V at OC for 20 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs -2.E-02 0.E+00 2.E-02 4.E-02 6.E-02 8.E-02 1.E-01-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial after 640hrs -2.E-02 0.E+00 2.E-02 4.E-02 6.E-02 8.E-02 1.E-01-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial after 640hrs Figure 6.19 Dark J-V at SC for 20 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs In Figure 6.18 we can see there is a shift in Voc for samples stressed at OC and from Figure 6.19 (left) we can see that there is increase in series resistance for samples stressed at SC. The shift in Voc at OC can be due to diffusion of Cu from the backcontact deteriorating the back-c ontact properties. Increase in series resistance can be seen for samples stressed at SC condition this might due to deterioration of ohmic property of the back-contact.

PAGE 80

64 Figure 6.20 Dark and Light J-V at OC (Left) and SC (Right) Figure 6.21 Semi-Logarithmic Dark J-V at OC (Left) and SC (Right) In Figure 6.20 we can see cross-over between light and dark J-V, this is due to photo-conductivity of CdS. In Figure 6.21 when stressed at SC (right) we can see ohmic losses resulting in increase in series-resistance.

PAGE 81

656.1.6.3 Sample with 40 Cu in Back Co ntact and Contact Annealed at 240oC -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.50 0.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.50 0.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs Figure 6.22 Light J-V at OC for 40 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.50 0.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.50 0.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs Figure 6.23 Light J-V at SC for 40 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs Figure 6.22 and 6.23 represent the light JV characteristics of sample with 40 of Cu in the back-contact, it can be clearly seen form the figures that there is shift in Voc for samples stressed at OC from Figure 6.22 where as for the samples stressed at SC from Figure 6.23 there is more th an one issue unlike 10 and 20 samples. We can see that there is collection problem, which is repres ented by using an arro w in the Figure 6.23.

PAGE 82

66 -2.E-02 0.E+00 2.E-02 4.E-02 6.E-02 8.E-02 1.E-01-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs -2.E-02 0.E+00 2.E-02 4.E-02 6.E-02 8.E-02 1.E-01-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 640hrs Figure 6.24 Dark J-V at OC for 40 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02 -2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] initial after 640hrs -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02 -2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] initial after 640hrs Figure 6.25 Dark J-V at SC for 40 Thick Cu Sputtered Back Contact Annealed at 240oC Initial and After 640hrs From the dark J-V characteristics shown in Figure 6.24 and 6.25 we can see that for samples stressed at SC there is considerab le amount of increase in series resistance. This increase can be attribut ed to the increased amount of Cu accumulating in CdS and due to its photoconductive nature of CdS sh owing increased electr ical resistance for samples when they are in dark cycle.

PAGE 83

67 Figure 6.26 Dark and Light J-V at OC (Left) and SC (Right) Figure 6.27 Semi-Logarithmic Dark J-V Plot at OC (Left) and SC (Right) In all these cases were the Cu concentrati on varied in the back contact we can see there is degradation in Voc and FF from th e light J-V characteristics shown in Figures 6.10 through 6.27. In all the ca ses degradation is found to be more for samples stressed at OC. This result is comparable to the results di scussed in section 3.6.2[ 14] in the literature review where the degradation at OC was to a greater extent than the degradation at SC. When the degradation of devices is compar ed between 10, 20 and 40 we can see that degradation is more for the samples with mo re amount of Cu in the back contact. This result is in contradiction to the results discu ssed in section 3.2 in literature review where 1.E-08 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 0.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] initial LS for 640 hrs 1.E-08 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 0.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] initial LS for 640hrs

PAGE 84

68 samples with more copper showed improved pe rformance than the samples with less Cu in the back contacts. The cro ss-over is found to increase with amount of Cu in the backcontact this may be due to accumulation of Cu in CdS which might have increased with amount of Cu in the back-contact, in accordan ce to the literature review discussed in section 3.6.3 were accumulation of Cu in CdS increased with concentration of Cu in the back-contact. From Figure 6.25 we can say that there is maximum increase in dark seriesresistance for sample stressed with 40 wh en compared to 10 and 20 samples. Hence we can assume that more Cu might have accumulated in CdS for 40 sample resulting increasing acceptor doping in CdS in turn reducing the n-type nature of CdS and hence resulting in increase in th e electrical resistance. 6.2 Analysis for Samples with 10 Cu and Contact Annealing Temperatures of 175oC and 200oC In Table 6.3 initial performance of set 2 measured at room temperature is listed. Table 6.3 Initial Performance of Set 2 contact annealing temperature(oC) Cu( ) sample# Voc FF Area(cm2) Avg Voc Avg FF 175 10 3 22B 28 770 62.6 0.4806 765 62 770 61.3 0.3339 760 61 0.4762 760 63.1 0.3726 200 10 3 22A 22 800 59.6 0.3686 797.5 63.725 800 62.6 0.5075 790 66.2 0.424 800 66.5 0.4951 240 10 3 22A 22 830 68.2 0.4897 822.5 66.65 820 68.3 0.4369 820 65.3 0.3904 820 64.8 0.1547

PAGE 85

69 Initial Voc variation760 770 780 790 800 810 820 830 150175200225250 Contact annealing temperature(C)Voc(mV) Initial Voc Initial FF60 62 64 66 68 150175200225250 Contact annealing temperature(C)FF Initial FF Figure 6.28 Initial Voc (Left) and FF (Right) for 10 Thick Cu Sputtered Back Contact with Varying Contact Annealing Temperature From Figure 6.28 we can see that sample annealed at 240oC gave good results implying that 240oC is the optimum temperature for c ontact annealing when compared to 175 and 200oC. An optimum temperature for contact annealing is necessary for activation of dopant (Cu) and to enable diffusion of Cu from the back-contact in order to form Cu2Te layer near the back-contact. 6.2.1 Variation of Voc and FF with Time HT @ 175C-Cu-10 LS @ OC (3-22B-28)0.55 0.6 0.65 0.7 0.75 0.8 0.85 1101001000 Time [Hrs]VOC [mV] HT @ 175C-Cu-10 LS @ SC (3-22B-28)0.55 0.6 0.65 0.7 0.75 0.8 0.85 1101001000 Time [Hrs]VOC [mV] Figure 6.29 Degradation of Voc at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 175oC

PAGE 86

70 HT @ 175C-Cu-10 LS @ OC (3-22B-28)40 50 60 70 80 1101001000 Time [Hrs]FF HT @ 175C-Cu-10 LS @ SC (3-22B-28)40 50 60 70 80 1101001000 Time [Hrs]FF Figure 6.30 Degradation of FF at OC (left) & SC (right) for 10 Thick Cu Sputtered Back Contact Annealed at 175oC Figure 6.29 and 6.30 shows the degradation of Voc and FF for two cells (one cell per graph) held at OC and SC and light so aked for 700 hours. These cells were contact annealed at 175C.The Cu sputtered thickness is 10. The Voc for cells at OC and SC appeared to increase slightly in 700 hrs light soaking period. The FF degraded for both cells at OC and SC; the degradation at OC a ppeared to be more than that of SC. The improvement in the Voc can be considered due to the enhancement in contact anneal due to the stress conditions. HT @ 200C-Cu-10 LS @ OC (3-22A-22)0.5 0.55 0.6 0.65 0.7 0.75 0.8 1101001000 Time [Hrs]VOC [mV] HT @ 200C-Cu-10 LS @ SC (3-22A-22)0.5 0.55 0.6 0.65 0.7 0.75 0.8 1101001000 Time [Hrs]VOC [mV] Figure 6.31 Degradation of Voc at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 200oC

PAGE 87

71 HT @ 200C-Cu-10 LS @ OC (3-22A-22)40 50 60 70 80 1101001000 Time [Hrs]FF HT @ 200C-Cu-10 LS @ SC (3-22A-22)40 50 60 70 80 1101001000 Time [Hrs]FF Figure 6.32 Degradation of FF at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 200oC Figure 6.31 and 6.32 shows the degradation of Voc and FF for two cells (one cell per graph) held at OC and SC and light so aked for 700 hours. These cells were contact annealed at 200C and Cu thickness is 10.The Voc for cells at OC and SC appeared to increase by the end of 700 hour light soakin g period, whereas FF at both OC and SC decreased. HT @ 240C-Cu-10 LS @ OC (3-22A-22)0.5 0.55 0.6 0.65 0.7 0.75 0.8 1101001000 Time [Hrs]VOC [mV] HT @ 240C-Cu-10 LS @ SC (3-22A-22)0.6 0.65 0.7 0.75 0.8 0.85 0.9 1101001000 Time [Hrs]VOC [mV] Figure 6.33 Degradation of Voc at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 240oC

PAGE 88

72 HT @ 240C-Cu-10 LS @ OC (3-22A-22)40 50 60 70 80 1101001000 Time [Hrs]FF HT @ 240C-Cu-10 LS @ SC (3-22A-22)40 50 60 70 80 1101001000 Time [Hrs]FF Figure 6.34 Degradation of FF at OC (Left) & SC (Right) for 10 Thick Cu Sputtered Back Contact Annealed at 240oC Figure 6.33 and 6.34 shows the degradation of Voc and FF for two cells (one cell per graph) held at OC and SC and light so aked for 700 hours. These cells were contact annealed at 240C.The Cu s puttered thickness is 10. The Voc for cells at OC appeared exhibit a decreasing trend through out the 700 hour light soaking period. Whereas for the cells at SC degradation is not that predominant, The Voc was almost un-altered. The FF degraded for both cells at OC and SC; the degr adation at OC appeared to be more than that of SC condition. When degradation for samples with 175,200 and 240oC is considered degradation was found to be more for sample annealed at 240oC suggesting that is not efficient when long-term stability is considered.

PAGE 89

736.2.2 J-V Characteristics for Set 2 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 350 Hrs LS for 470 Hrs LS for 640Hrs 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 0.000.250.500.751.001.251.501.752.00 Voltage [Volts]Current Density [A/cm2] Initial LS for 350 Hrs LS for 470 Hrs LS for 640Hrs Figure 6.35 Dark J-V at OC for 10 Thick Cu Sputtered Back Contact Annealed at 175oC -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02 -2.0 0 -1.5 0 -1.0 0 -0.5 0 0.000.501.001.502.00 Voltage [Volts]Current Density [A/cm2] Initial LS for 350 Hrs LS for 470 Hrs LS for 640hrs 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 0.000.250.500.751.001.251.501.752.00 Voltage [Volts]Current Density [A/cm2] Initial LS for 350 Hrs LS for 470 Hrs LS for 640hrs Figure 6.36 Dark J-V at SC for 10 Thick Cu Sputtered Back Contact Annealed at 175oC Figure 6.35 and 6.36 shows the J-V character istics held at OC and SC and light soaked for 700 hours. These cells were contact annealed at 175C and Cu thickness is 10.From Figure 6.28 (right) we can see that cells stressed at OC had increase in the current at small voltages (0-0.4V), which is identified as dark shunting, may be considered as the reason of lowering the FF at OC with stress period.

PAGE 90

74 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 150 Hrs LS for 350 Hrs LS for 470 Hrs LS for 640Hrs 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 0.000.250.500.751.001.251.501.752.00 Voltage [Volts]Current Density [A/cm2] Initial LS for 150 Hrs LS for 350 Hrs LS for 470 Hrs LS for 640Hrs Figure 6.37 Dark J-V at OC for 10 Thick Cu Sputtered Back Contact Annealed at 200oC -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] Initial LS for 150 Hrs LS for 350 Hrs LS for 470 Hrs LS for 640 Hrs 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 0.000.250.500.751.001.251.501.752.00 Voltage [Volts]Current Density [A/cm2] Initial LS for 150 Hrs LS for 350 Hrs LS for 470 Hrs LS for 640 Hrs Figure 6.38 Dark J-V at SC for 10 Thick Cu Sputtered Back Contact Annealed at 200oC Figure 6.37 and 6.38 shows the J-V character istics held at OC and SC and light soaked for 700 hours. These cells were contact annealed at 200C and Cu thickness is 10. From Figure 6.37 (left) we can see that cells stressed at OC had increase in the series resistance as indicated by the slope of the J-V curve at high forward currents. Figures 6.10 through 6.13 shows the light and dark J-V characteristics held at OC and SC and light soaked for 700 hours. Thes e cells were contact annealed at 240C and Cu thickness is 10. From Figure 6.11 (left) an d 6.13 (left) we can see that cells stressed

PAGE 91

75 at SC had increase in the series resistance as indicated by the slope of the J-V curve at high forward currents and from Figure 6.10 and 6.12 there is a shift in the J-V curve after stress period. This may be due to change in potential barrier cause d by diffusion of Cu from the back contact towards the CdTe/CdS junction. 6.2.3 Variation in Voc and FF at OC and SC Conditions Each sample had four cells as shown in Figure 5.4, out of which a and c are stressed at OC and b and d are stressed at SC. Figure 6.33 shows the Initial Voc for 10 sample at different cont act annealing temperatures measured at initial stress conditions inside the stress chamber. 710 720 730 740 750 760 770 780 Voc(mV) Initial Voc at different contact annealing temperatures 175 740740739740 200 740740739740 240 780780760 acbd Figure 6.39 Initial Voc at Different Contact Annealing Temperatures Figure 6.39 shows initial Voc at OC and SC for 10 samples with varying contact annealing temperatures (175,200 and 240oC) after stress period. We can see that the initially Voc is more for the 240oC sample. From this we can say that 240oC may be an optimum temperature for contact annea ling to give high initial efficiency.

PAGE 92

76 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 Degradation in Voc 175 13.513.513.5 200 8.18.110.710.8 240 17.917.97.9 acbd Figure 6.40 %Degradation in Voc at Different Contact Annealing Temperatures Figure 6.40 shows the degradation in Vo c at OC and SC for 10 samples with varying contact annealing te mperatures (175,200 and 240oC) after stress period. We can see that the degradation is more for the 240oC sample, while it is minimum for 200oC sample. From this we can say that 200oC may be an optimum temperature for contact annealing when stability is concerned. Figure 6.41 shows the Initial FF for 10 sample at different contact annealing temperatures measured at initial stress conditions inside the stress chamber.

PAGE 93

77 0 10 20 30 40 50 60 FF Initial FF at different contact annealing temperatures 175 54.73551.880149.838 200 54.8158.0546.4448.41 240 52.6755.348.79 acbd Figure 6.41 Initial FF at Different Contact Annealing Temperatures Figure 6.42 shows the degradation in FF at OC and SC for 10 sample with varying contact annealing temp eratures (175,200 and 240oC) after stress period. 0 5 10 15 20 25 30 % degradation Degradation in FF for 10 sample 175 12.106881614.233592319.4081501 200 11.98644714.0337245827.149055821.7173351 240 22.085914110.875451319.9553758 a-OCc-OCb-SCd-SC Figure 6.42 %Degradation in FF at Different Contact Annealing Temperatures

PAGE 94

78 Comparing different contact annealing temperatures i.e. 175, 200 and 240oC with 10 of Cu in the back-contact we can say th at Voc improved at OC and SC for both 175 and 200 annealed samples Voc degraded for 240 annealed samples, whereas FF degraded in all the samples. The improvement in Voc for 175oC and 200oC might be because stress testing might worked as annealing for the samples which lead to improvement in cell properties. Contact annealing has impact on th e cell properties as disc ussed in section 3.4 of the literature review wher e effect of contact annealing temperature on efficiency is shown in Figure 3.10. Hence optimum annea ling temperature is required for obtaining good cell. 6.3 Effect of Light Soaking on Rse In the real cells, power is dissipated through resistances of contacts and through leakage currents through sides of the devices. These effects ar e equivalent to resistances in series (Rse) and parallel (R sh) to the cell. Series and s hunt resistances reduce the FF of the device, for maximum efficiency series resistance should be minimum and shunt resistance should as large as possible. In this section variation in series resistance has been depicted. In Figure 6.44 variation in Rse at OC and SC during dark cycle for 10, 20 and 40 is shown, we can see from the Figure 6.44 that increase in Rse at SC was more when compared to OC for all the cases. This mi ght be the reason for the degradation of FF being more when stressed at SC condition In Figure 6.45 variation in Rse during light cycle is shown, the increase in Rse at SC was more than OC condition

PAGE 95

79 Figure 6.43 Rse Variation in Dark Condition Figure 6.44 Rse Variation under Light Soaking From the Figure 6.43 and Figure 6.44 we can see that there was gradual increase in Rse in all the cases.

PAGE 96

806.4 Variation in Rsh with Light Soaking 6.4.1 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Dark and Light at OC Table 6.4 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Dark at OC For all the cells stressed at OC there was decrease in Rsh Table 6.5 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Light at OC From Table 6.7 we can see that there is im provement in Rsh for 20 and 40 whereas the Rsh value decreased for 10 sample. 10 20 40 Initial 4X10E4 5X10E4 5X10E4 After 350 hrs 3X10E4 9X10E6 18X10E3 After 450 hrs 4X10E4 3X10E5 12X10E3 After 650hrs 900 8X10E6 18X10E3 10 20 40 Initial 750 400 1450 After 350 hrs 470 600 50 After 450 hrs 440 500 400 After 650hrs 40 740 2500

PAGE 97

816.4.2 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Dark and Light at SC Table 6.6 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Dark at SC 10 20 40 Initial 4X10E4 1400 5X10E4 After 350 hrs 6X10E4 2000 4X10E4 After 450 hrs 1X10E4 2000 6X10E4 After 650hrs 3X10E4 2400 6X10E4 There was small improvement (Rsh increa sed) for the cells stressed at short circuit condition. Table 6.7 Shunt Resistance Variation for 10, 20, 40 up to 640hrs in Light at SC 10 20 40 Initial 600 420 540 After 350 hrs 580 370 290 After 450 hrs 495 200 800 After 650hrs 180 400 300 From Table 6.9 we can see that the dete rioration in FF is more for 10 sample

PAGE 98

826.5 Transient Behavior The light cycle is for 4 hours and dark cy cle follows for next 4 hours. During light cycle measurements are taken in 7 intervals. The seven intervals chosen for measurement are, after 5, 40, 80, 120, 150, 180 and 210m.each measuremen t cycle would take 15m. Similarly dark measurements are taken in three intervals, af ter 15, 120 and 195m. During the light cycle and dark cycle of the stress period a specific trend was observed in Voc and FF. As shown in the Fi gure 6.45 in the hour light soaking Voc dropped about 40-60mV.This drop in the Voc can be considered due to the heating of the sample due to the light soaking for four hours and we can also see from the figure that the Voc has been recovered in the next 4 hours of dark period. (The X-axis represent the total light soaked hours). Figure 6.45 Variation of Voc with Time (4 Hour Dark Cycle Transpired in Between is not Shown in the Figure)

PAGE 99

83 The degradation and recovery of Voc is compared at both OC and SC in the Figure 6.46. We can see that the amount of degradation and recovery at both OC and SC followed a similar trend. Figure 6.46 Variation of Voc at OC and SC (Not Shown in the Figure 4 Hour Dark Cycle) In Figure 6.47 change in FF in the hour light soaking is shown, we can see that the degradation was not similar to what we have observed in Voc. Initially in the 4 hour period FF was found to improve and dro pped consistently for next 3 hours and improved again during last 30 min. (This trend followed though out the light soaking period but only 16 hours has been shown in the figure for clarity).

PAGE 100

84 Figure 6.47 Variation of FF with Time As evident in four hour cycle in Voc, FF also should go down because sample would heated up as result of continuous soaki ng for 4 hours, but in our case as shown in Figure 6.47 the FF seemed to improve so therefor e it cannot be considered as the affect of heat, it might me initial heating up of the samp le is acting as annealing of the sample and improving the performance. The FF variation at OC and SC is compared in the Figure 6.48 we can see that the recovery and degradation of FF is exactly in opposite way for both OC and SC. When the FF decreases at OC at the same pe riod FF was found to improve at SC. Recovery due to the change in bias can be due to the conversion of defects either by capturing the carrier or by emitting the captured carrier, which changes recombination in the junction region. Thermal or field enhanc ed drift may create defects which trap the charge in the presence of light [23], when generation takes place and during dark cycle

PAGE 101

85 i.e. when there is no gene ration these defects may capture or emit the trapped charges causing the recovery in cell performance. Figure 6.48 Variation in FF at OC and SC (4 Hour Dark Cycle Transpired in Between is Not Shown in the Figure) The Figure 6.46 and Figure 6.48 are for the sample having 40 of sputtered Cu in the back contact. The 10 sample has been ob served to see if it behaved in the same manner. Figure 6.49 and Figure 6.50 are the Voc and FF variation for the sample with 10 sputtered Cu in the back cont act. We can see from Figure 6. 49 that the Voc at OC and SC degraded and recovered in similar pattern, th e degradation in the Voc is 40mV 60mV. This degradation and recovery can be cons idered due to heating of samples due to continuous light soaking for 4 hours.

PAGE 102

86 Figure 6.49 Variation of Voc at OC and SC for 10 Sample Figure 6.50 shows the varia tion in the FF at OC and SC, here the pattern of degradation and recovery is not similar to th e pattern of 40 sample and the extent of transience also varied. It is clearly seen that the transients is more for 40 sample when compared to 10 sample. This variation in ex tent of transients can be attributed to amount of Cu in the back contact. According to the study performed in Univer sity of Delaware [ 23], transients in CdS/CdTe solar cells was not affected by Cu in the back contact, but in our case we find the extent of transients varied with amount of Cu in the back contact. This variation may be due to the drift in Cu and also change in band bending changes the trap occupancy [23]. The reason for variation in transients ex tent might be due to change in the band bending which might be varied due to the amou nt of Cu in the back contact. The cells with more amount of might have enabled diffusion of Cu towards the junction to a greater extent hence reducing the ohmic prop erty of the back contact and creating a barrier at the back contact. This in turn ch anges the band bending. From the section 3.3 in

PAGE 103

87 literature review it is stated that the carrier concentrat ion in the back contact is responsible for barrier height. Figure 6.50 Temperature Dependence of FF at OC and SC for 10 Sample Figure 6.51 Variation of Voc and FF with Time for 20 Sample Figure 6.51 shows the variation in Voc and FF at different interv als of time i.e. 816 hrs, 408-424 hrs and 664-674 hrs, this gives us a clear picture of the variation in extent of transients as we approach to 700 hrs, we can see that the extent of transients seemed to die of as we approach 700hrs.

PAGE 104

88 Figure 6.52 Temperature Dependence of FF at OC and SC for 20 Sample Figure 6.53 Variation of Voc and FF with Time for 20 Sample Figure 6.52 shows the transient phenomenon in FF for the sample with 20 of Cu in the back contact. The pattern seemed to reverse at 720 hrs. Figure 6.53 shows the variation in Voc and FF at diffe rent intervals of time i.e. 8-16 hrs, 408-424 hrs and 664674 hrs, this gives us a clear picture of the vari ation in extent of tran sients as we approach to 700 hrs. We can see that the transients persisted in Voc as we approach 700 hrs.

PAGE 105

89 Figure 6.54 Temperature Dependence of FF at OC and SC for 40 Sample Figure 6.55 Variation of Voc and FF with Time for 40 Sample Figure 6.54 shows the transient phenomenon in FF for the sample with 40 of Cu in the back contact. The pattern seemed to reverse at 720 hrs. Figure 6.55 shows the variation in Voc and FF at diffe rent intervals of time i.e. 8-16 hrs, 408-424 hrs and 664674 hrs, this gives us a clear picture of the vari ation in extent of tran sients as we approach to 700 hrs. We can see that the transients fo r 40 persisted and were to a greater extent than 20 sample as we approach to 700 hrs.

PAGE 106

906.6 Conclusions Different Cu thickness and contact anneal temperatures where tested, due to the presence of Cu in back contact initial perf ormance of the device improved. Initially 10 of Cu in the back contact gave better Voc and FF. After 700 hrs of light soaking all the devices tend to degrade except for some of the devices where there was an improvement in Voc and FF. When samples with 10, 20 and 40 samples annealed at 240oC are compared, degradation in 40 was more When comparing different annealing temperatures (175oC, 200oC, 240oC) with 10 of Cu in the back contact, we can see that the degradation is minimum for 200oC annealed sample, so we can conclude that 200oC is optimum temperature for contact annealing. Transient phenomenon was observed for the sa mples with Cu in the back contact. The extent of transients increased with the am ount of Cu in the back contact, sample with 40 showed maximum extent of transients. The transient phenomenon seemed to die off approaching 700hrs for 10 and 20 samples whereas transient phe nomenon persisted in the sample with 40 of Cu in the back contact.

PAGE 107

91 References [1] Charles E. Backus, Solar Cells, IEEE Press, NY, 1976. [2] A. Wang, J. Zhao, and M.A. Green, 4% Efficient Silicon Solar Cells, Appl. Physics Letters, Vol. 57, No. 6, August 1990. [3] Xuanzhi Wu, High-efficiency polycryst alline CdTe thin-film solar cells, Solar Energy, Vol. 77, 2004, pp. 803 814. [4] K. Ramanathan, M. A. Contreras,C. L. Perk ins, S. Asher, F. S. Hasoon, J. Keane, D. Young, M. Romero, W. Metzger, R. Nouf i, J. Ward, A. Duda, Properties of 19.2% Efficiency ZnO/CdS/CuInGaSe2 Thin-Film Solar Cells, Progress in Photovoltaics: Research and Applic ations. Vol. 11, 2003; pp. 225-230. [5] Meena Dadu, A. Kapoor, K.N. Tripathi, E ffect of operating current dependent series resistance on the fill factor of a solar cell, Solar Energy Materials a Solar Cells, Vol. 71, 2002, pp.213. [6] Dieter Bonnet. Proceedings of 14th European Photovolta ic Solar Energy Conference, Barcelona, Spain, 1997, pp. 2688. [7] H.R. Moutinho, D. Albin, Y. Yan, R.G. Dhere, X. Li, C. Perkins, C.-S. Jiang, B. To, M.M. Al-Jassim, Deposition and propertie s of CBD and CSS CdS thin films for solar cell application, Thin Solid Films, Vol. 436, 2003, 175. [8] A.D. Campaan, J. R. Sites, R. W. Birkmi re, C.S. Ferekides, and A.L Fahrenbruch, Critical Issues and Research Needs for Cd Te Based Solar Cells, Photovoltaics for the 21st Century, Electrochemical So ciety, Incorporated, 1999, pp 241-251. [9] G.Streetman, Solid State Electroni c Devices. Prentice-Hall of INDIA,1995. [10] Kevin D. Dobson, Iris Visoly-Fisher, Garry Hodes, David Cahen, Stability of CdTe/CdS thin-film solar cells, Solar Energy Materials & Solar Cells, 62, pp 295325, 2000. [11] D.L. Batzner, A. Romeo, H. Zogg, A.N. Tiwari and R. Wendt, Development of Efficient and Stable Back Contacts on CdTe/CdS Solar Cells, Thin Solid Films, Vol.387, 2001, pp 151-154.

PAGE 108

92 [12] A.O. Pudov, M. Gloeckler, S.H. Demtsu and J.R. Sites, Effect of Back-Contact Copper Concentration on CdTe Cell Operation, Proceeding of 29th IEEE Photovoltaic Specialists Conference, 2002, pp 760-763. [13] Ashok Rangaswamy, Masters Thesis, Electrical Engineer ing, University of South Florida, 2003. [14] D.L. Batzner, A. Romeo, H. Zogg, A. N. Tiwari and R. Wendt, Effect of back contact metallization on CdTe/CdS solar cells Presented at the 16th European Photovoltaic Solar Energy Conference 2000 in Glasgow. [15] D.L. Batzner, A. Romeo, H. Zogg, A.N. Tiwari and R. Wendt, A Study of The Back Contacts of CdTe/CdS Solar Cells, Thin Solid Films, Vol. 61-362, Issue(12), pp 463-467, 2000. [16] Steven S. Hegedus, Brain E. McCandles s, and Robert W. Birkmire, Analysis of Stress-Induced Degradation in CdS/CdTe Solar Cells, Proceeding of 28th IEEE Photovoltaic Specialists Conference, 2000, pp 535-538. [17] S.E. Asher, F.S. Hasson, T.A. Gessert M.R. Young, P. Sheldon, J. Hiltner and J. Sites, Determination of Cu in CdTe/CdS Devices Before and After Accelerated Stress Testing, Proceeding of Photovolta ics Specialists Conference, pp 479-482, 2000. [18] T.A. Gessert, D. Duda, S.E. Asher, C. Narayanswamy, and D. Rose, Effects of Cu from ZnTe:Cu Contacts in CdS/CdTe Solar Cells, Proceeding of 28th IEEE Photovoltaic Specialists C onference, 2000, pp. 17-22. [19] K. Kuribayashi, H. Matsumoto, H. Uda, Y. Komatsu, A. Nakano, and S. Ikegami, Preparation of Low Resistance Contact El ectrode in Screen Printed CdS/CdTe Solar Cell, Japan Journal Applied Physics, Vol. 22 ,1983, pp. 178182. [20] Steven S. Hegedus and Antonio Luque, Status, Trends, Challenges and The Bright Future of Solar Electricity from Phot ovoltaics, Handbook of Photovoltaic Science and Engineering, John Wiley & Sons, 2003. [21] B. Tetalli, V. Vishwanath, D.L. More l, C. S. Ferekides, Effects of Thermal Stressing on CdTe/CdS Solar Cells, Proceeding of Photovoltaics Specialists Conference, pp 600 603, 2002. [22] M. Bashahu and A. Habyarimana, Rev iew and Test of Methods for Determination of the Solar Cell Series Resistance. Re newable Energy, 1995, 6, 2, pp 129 138.

PAGE 109

93 [23] Steven Hegedus, Darshini Desai, Dan Ryan, Drain McCandless, Transient Degradation and Recovery of CdS/CdTe Solar Cells, 31st IEEE PVSC, Orlando, 2005.

PAGE 110

94 Appendices

PAGE 111

95Appendix A: Measurement Automation Automation of the Stability measurement system is done in order to maintain consistency and accuracy of data, reducin g measurement time, create easy stress conditions, collection of large amount of data monitoring stress temperature, humidity and stress time. The hardware part consists of solid st ate relays which ar e controlled by PCI digital IO card via opto couplers. The relay is in NC position when not energized. The required stress conditions can be configured in this position via terminals provided from NC and GND (which is connected to n side of diode to be tested) outputs. The possible configurable stress conditions are open circuit, short circui t, maximum Load and, forward and reverse bias. When the relay is energized the device under test (DUT) is connected to measurement device. Care must be ta ken while programming relays to avoid simultaneous activation of more than one rela y. The dip switch is provided to isolate a relay from experiment in case of circuit failu re .Totally 48 relays are controlled by a single 48 DIO card which consists of 2-8255 PPI chips. DUT is thus continuously maintained in the desired stress condition ex cept for the duration in few seconds during measurement. Keithley 2400 source meter is us ed to sweep the voltage from -2V to +2V in steps of 0.02V and measure the current during I-V measurement. Omega 44-M data logger is used to monitor temperature and humidity. It is connected from the stress oven via RS 232 interface. Software is provided by Omega to collect the data. It contains internal temperature and humidity sensors and two external thermistors. Sampling interval can be varied from 0.5 sec. The soft ware is developed usi ng LabView. The front

PAGE 112

96 Appendix A (Continued) panel of main vi program and its block diagra m of data flow during dark cycle are shown. Some of the salient features of the software are Day and night cycle is monitored and logged, Logic is entirely based on real time test condition, The measurements are avoided in case of lamp failure, Figure A.1 Front Panel of VI Program for Stability Testing

PAGE 113

97 Appendix A (Continued) Figure A.2 Snap Shot of Block Diagram Figure A.3 Circuit Configuration of Hardware Setup

PAGE 114

98 Appendix A (Continued) The total light soaked hours are calculated with milli second accuracy, Day transition will not cause problems, Measurement timings can be custom defined for any number of times needed The code is optimized to avoid memory leaks The device parameters are stored in well defined manner for easy retrieval for data analysis. The measurement timings must be given in ascending order. Figure A.4 Input Panel for Measurement Interval Specification

PAGE 115

99Appendix B: Degradation in Solar Cell Parameters B.1 Variation of Rse B.1.1 During Dark Cycle Figure B.15 Variation of Series Resistance for Both Open and Short Circuit Conditions with Time for 10, 20 and 40 of Cu Concentration Respectively During Dark Cycle Table B.1 Variation of Series Resistance for both Open and Short Circuit Conditions with Time for 10, 20 and 40 of Cu Concentration Respectively During Dark Cycle Open Circuit 10 20 40 Short Circuit 10 20 40 Initial 1.02 2.24 Initial 1.07 1.5 2.02 After 350hrs 1.91 4.22 3.16 After 350hrs 1.86 4.1 2.64 After 450hrs 2.12 4.29 3.51 After 450hrs 2.06 4.39 2.88 After 650hrs 2.25 4.51 4.03 After 650hrs 2.2 4.86 3.42

PAGE 116

100 Appendix B (Continued) B.1.2 During Light Soaking Figure B.2 Variation of Series Resistance for Both Open and Short Circuit Conditions with Time for 10, 20 and 40 of Cu Concentration Respectively During Light Cycle Table B.2 Variation of Series Resistance for Both Open and Short Circuit Conditions With Time for 10, 20 and 40 of Cu Concentration Respectively During Light Cycle Open Circuit 10A 20A 40A Short Circuit 10A 20A 40A Initial 1.07 1.55 2.24 Initial 1.18 1.74 1.89 After 350hrs 2.00 3.63 2.21 After 350hrs 2.46 3.46 2.35 After 450hrs 2.32 3.76 2.43 After 450hrs 2.86 3.7 2.62 After 650hrs 2.15 3.68 2.46 After 650hrs 2.85 3.64 2.56

PAGE 117

101 Appendix B (Continued) B.2 Variation of Rsh B.2.1 During Dark Cycle at OC 3-22-a22-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-a22-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-a22-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b22-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 1.E+09 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b22-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 1.E+09 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b22-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 1.E+08 1.E+09 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b27-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b27-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b27-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ Figure B. 3 Variation of Shunt Resistance at Dark OC for Samples with 10A, 20A, 40A Cu Sputtered Contact

PAGE 118

102 Appendix B (Continued) B.2.2 During Dark Cycle at SC 3-22-a22-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-a22-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-a22-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b22-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b22-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b22-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b27-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b27-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ 3-22-b27-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Dark dV/dJ Figure B.4 Variation of Shunt Resistance at Dark SC for Samples with 10A, 20A, 40A Cu Sputtered Contact

PAGE 119

103 Appendix B (Continued) Degradation in Rsh in dark condition 1.00E+00 1.00E+01 1.00E+02 1.00E+03 1.00E+04 1.00E+05 1.00E+06 1.00E+07 0100200300400500600700 Time(hrs)Rsh(ohm-cm^2) 10A-OC 20A-OC 40A-OC 10A-SC 20A-SC 40A-SC Figure B.5 Degradation in Rsh in Dark Condition

PAGE 120

104 Appendix B (Continued) B.2.3 During Light Cycle at OC 3-22-a22-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-a22-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-a22-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b22-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b22-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b22-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b27-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b27-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b27-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ Figure B.6 Variation of Shunt Resistance at Light SC for Samples with 10A, 20A, 40A Cu Sputtered Contact

PAGE 121

105 Appendix B (Continued) B.2.4 During Light Cycle at SC 3-22-a22-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-a22-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-a22-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b22-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b22-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b22-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 1.E+07 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b27-3501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b27-4501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ 3-22-b27-6501.E-01 1.E+00 1.E+01 1.E+02 1.E+03 1.E+04 1.E+05 1.E+06 -2.00-1.50-1.00-0.500.000.501.001.502.00 Voltage [Volts]dV/dJ [ -cm2] Light dV/dJ Figure B.7 Variation of Shunt Resistance at Light OC for Samples with 10A, 20A, 40A Cu Sputtered Contact

PAGE 122

106 Appendix B (Continued) Degradation in Rsh under light soaking 1 10 100 1000 10000 0100200300400500600700 Time(Hrs)Rsh(ohm-cm^2) 10A-OC 20A-OC 40A-OC 10A-SC 20A-SC 40A-SC Figure B.8 Variation of Shunt Resistance Under Light Soaking B.3 Comparison of Cu Sputtered Thickness with Dark J-V 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 0.000.250.500.751.001.251.501.752.00 Voltage [Volts]Current Density [A/cm2] 10 20 40 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02 -2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] 10 20 40 Figure B.9 Dark Analysis of 10, 20 and 40 of Cu Sputtered Back Contact Annealed at 240oC at OC

PAGE 123

107 Appendix B (Continued) 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 0.000.250.500.751.001.251.501.752.00 Voltage [Volts]Current Density [A/cm2] 10 20 40 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02 -2.00-1.50-1.00-0.500.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] 10 20 40 Figure B.10 Dark Analysis of 10, 20 and 40 of Cu Sputtered Back Contact Annealed at 240oC at SC -4.E-02 -2.E-02 5.E-03 3.E-02 5.E-02 7.E-02 9.E-02 -2.00-1.50-1.00-0.50 0.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] 10 20 40 -4.E-02 -2.E-02 5.E-03 3.E-02 5.E-02 7.E-02 9.E-02-2.00-1.50-1.00-0.50 0.000.501.001.502.00Voltage [Volts]Current Density [A/cm2] 10 20 40 Figure B.11 Light Analysis of 10, 20 and 40 of Cu Sputtered Back Contact Annealed at 240oC at OC (Left) and SC (Right)

PAGE 124

108 Appendix B (Continued) HT @ 240C-Cu-40 LS @ OC (3-22A-23)0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 1101001000 Time [Hrs]VOC [mV] HT @ 240C-Cu-40 LS @ SC (3-22A-23)0.5 0.55 0.6 0.65 0.7 0.75 0.8 1101001000 Time [Hrs]VOC [mV] Figure B.12 Degradation of Voc at OC & SC for 40 Thick Cu Sputtered Back Contact Annealed at 240oC B.4 FF Variation in Monochromatic J-V Table B.3 Initial FF Comparison of 10, 20 and 40 of Cu Sputtered Back Contact Annealed at 240oC FF at wavelength Sample# Contact Annealing Temp in (oC) Cu thickness in () Cell 460 520 640 800 3 22A 23 240 40 A 61.8 56.2 49.2 40.4 B 70.5 65.8 57.8 49.3 C 43.1 44.7 41 36.8 D 69.1 66 60.7 51.6 3 22B 22 240 20 A 77.7 77.9 77.5 70.6 B 50.6 53 49.2 40.1 C 72.3 71 71.1 66.1 D 68.3 66.5 64 56.7 3 22A 22 240 10 A 75.1 74.9 72.6 63.9 B 75.9 74.2 71.7 64.4 C 74.2 73.6 72 64.4 D 63.2 62.9 62.4 53.1

PAGE 125

109 Appendix B (Continued) 3 22A 23 240/40A 0 10 20 30 40 50 60 70 80 90 460520640800 Wavelength in nmFF a b c d 3 22B 22 240/20A0 10 20 30 40 50 60 70 80 90 460520640800 Wavelength in nmFF a b c d 3 22A 22 240/10A0 10 20 30 40 50 60 70 80 90 460520640800 Wavelength in nmFF a b c d Figure B.13 Initial FF Comparison of 10, 20 and 40 of Cu Sputtered Back Contact Annealed at 240oC B.5 Effect of Contact anne aling Temperature on JV Figure B.14 Light J-V at OC & SC for Different Contact Annealing Temperatures -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02 0.000.250.500.751.001.251.50 Voltage [Volts]Current Density [A/cm2] 175 200 225 240 275 -3.E-02 -1.E-02 1.E-02 3.E-02 5.E-02 7.E-02 9.E-02 0.000.250.500.751.001.251.50 Voltage [Volts]Current Density [A/cm2] 175 200 225 240 275


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001681068
003 fts
005 20060215071208.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 051227s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001050
035
(OCoLC)62744298
SFE0001050
040
FHM
c FHM
049
FHMM
090
TK145 (Online)
1 100
Erra, Swetha.
0 245
Stability studies of cdte solar cells with varying amounts of cu in the back contact
h [electronic resource] /
by Swetha Erra.
260
[Tampa, Fla.] :
b University of South Florida,
2005.
502
Thesis (M.S.E.E.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 125 pages.
3 520
ABSTRACT: Solar energy is one of the abundant, non-polluting renewable energy options in our planet. During the last three decades considerable progress has been achieved in developing technologies to produce electricity from solar radiation, but producing electricity with low cost and low pollution is of concern. The CdTe solar cells are the leading source for the production of cost effective solar cells. The main issue of concern in these CdTe solar cells is degradation observed when stressed at elevated temperatures. The degradation in CdTe solar cells can be attributed to the back contact, which often contains Cu to improve the electronic properties of CdTe (absorber layer) and to enable a quasi ohmic back contact.The main objective of this thesis was to study effect of amount of Cu in the back contact and contact annealing temperature on device stability.
590
Adviser: Dr.Christos Ferekides.
653
Copper.
Transients.
Degradation.
Stress testing.
Light soaking.
690
Dissertations, Academic
z USF
x Electrical Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1050