University of Southern California Dissertations and Theses

Fabrication and study of organic and inorganic optoelectronics using a vapor phase deposition (VPD)

(USC Thesis Other)

Fabrication and study of organic and inorganic optoelectronics using a vapor phase deposition (VPD)

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

FABRICATION AND STUDY OF ORGANIC AND INORGANIC OPTOELECTRONICS
USING A VAPOR PHASE DEPOSITION (VPD)

by
Francisco Fabian Navarro
______________________________________________________________________________

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMICAL ENGINEERING)

May 2016

Copyright 2016 Francisco Fabian Navarro
I

Dedication

To my mother and father for all their
support and love. To my grandfather
(apa Cirio) who’s watching over me
from heaven.

A mi madre y padre por todo su
apoyo y amor. A mi apa Cirio quien
me cuida desde el cielo.

II

Acknowledgements

I want to thank my principal investigator professor Mark E. Thompson for believing in me and
for giving me the opportunity to complete my degree in his laboratory/group. Mark is a remarkable
individual with a great deal of self-effacement and I consider him to be among the finest graduate
advisors I have met. Special thanks to research assistant professor Peter I. Djurovich, whom I
consider my co-advisor, for his countless ideas, advice and unrestricted share of knowledge. I also
want to acknowledge my committee members, associate professor Andrea M. Armani, associate
professor Noah Malmstadt, associate professor C. Ted Lee, Jr. and professor G. K. Surya Prakash,
for making time out of their busy schedule to listen and examine my research findings. Special
thanks to Professor Stephen R. Forrest, Dr. Cedric Rolin and Garen Vartanian for helpful
discussions of the organic vapor phase deposition system. Thanks to assistant professor Matthew
T. Whited for the synthesis of Pt(
tbu
TPBP). Thanks to assistant professor Cody W. Schlenker for
helpful discussions and for teaching me fabrication and characterization of organic photovoltaics.
Thanks to Dr. Chao Wu for helpful discussions and for teaching me fabrication and
characterization of organic light emitting diodes. Professor William H. Steier and Dr. Hari
Mahalingam for help with thin film measurements with the broadband spectroscopic ellipsometer
and profilometer profiles. Professor Chongwu Zhou and Dr. Yang Song for help with AFM
images. The center for electron microscopy and microanalysis (CEMMA) and Dr. Yang Song for
help with SEM images. Special thanks to Mr. Phillip Sliwoski for glasswork and Mr. Donald
Wiggins, Mr. Michael Cowan and Mr. Ramon Delgadillo for machinery work. Their excellent
craftsmanship work enable me to modify the vapor phase deposition and successfully complete
my research. Patrick Saris and John Facendola for providing organic compounds helpful for the
III

metal deposition project. Special thanks to assistant professor Brent C. Melot for helpful
discussions on fabrication and characterization of perovskite solar cells. Kelsey Bass for
providing/synthesizing organic and inorganic materials for the perovskite solar cell project.
Thanks to Andrew J. Clough for his help with the x-ray photoelectron spectroscopy (XPS)
measurements. Thanks to Shiliang Zhou for help with the inductively coupled plasma optical
emission spectrometry (ICP-OES) measurements. Thanks to professor Andrea Hodge and Mikhail
Polyakov for their help with the Ambios XP-2 profilometer measurements. Special thanks to
Professor Craig A. Ogle and Banita W. Brown for their advice and support during my
undergraduate studies at UNC Charlotte. Their passion and guidance placed me on the doctorate
degree path. I want to thank Rebecca L. Pizzitola for her support and advice during my studies and
while writing articles and also for her assistance with editing. I want to show appreciation to the
institutions that funded my studies, the University Of Southern California, University Display
Corporation and the Department of Energy. Special thanks to administrative assistants Judy Hom
and Michele Dea, their dedication and hard work was integral to accomplish my research, without
them the laboratory and the department of chemistry would not have functioned efficiently.
Finally, I want to thank my family. My mother and father for all their love and support, I
wouldn’t be the person I am if it wasn’t for them. Thanks to my siblings, Veronica, Dianette la
flaca and Ricardo el peludo. I wasn’t the most sociable person during my studies and having their
tolerance and support meant everything for me. I also want to acknowledge my grandmother “ama
chabela” and my grandfather “apa cirio”, for their unconditional love, it kept me going and gave
me strength when I needed it the most.

IV

Table of Contents

DEDICATION ................................................................................................................................. I
ACKNOWLEDGEMENTS ............................................................................................................ II
LIST OF FIGURES ...................................................................................................................... VI
LIST OF TABLES ....................................................................................................................... XV
ABSTRACT ............................................................................................................................. XVIII
CHAPTER 1. ORGANIC VAPOR PHASE DEPOSITION (OVPD) ....................................... 1
1.1 INTRODUCTION ................................................................................................................ 1
1.2 THEORY ........................................................................................................................... 4
1.3 APPLICATIONS ................................................................................................................. 7
1.3.1 Organic light emitting diodes (OLEDs) ...................................................................... 7
1.3.2 Organic photovoltaics (OPVs) .................................................................................. 10
1.3.3 Organic thin film transistors (OTFTs) ...................................................................... 17
1.4 ORGANIC VAPOR PHASE MODIFICATIONS ....................................................................... 18
1.5 CONCLUSIONS ................................................................................................................ 20
1.6 CHAPTER 1 REFERENCES................................................................................................ 21
CHAPTER 2. EXCITON DIFFUSION LENGTH MEASUREMENT OF PLATINUM 1, 3-
DI-TERT-BUTYLPHENYL TETRAPHENYLBENZOPORPHYRIN PT(
TBU
TPBP) FOR
DESIGN AND OPTIMIZATION OF ORGANIC PHOTOVOLTAICS ..................................... 27
2.1 INTRODUCTION .............................................................................................................. 27
2.2 RESULTS AND DISCUSSION ............................................................................................. 30
2.2.1 Stern-Volmer relationship ......................................................................................... 30
2.2.2 Film morphology ...................................................................................................... 32
2.2.3 Spectrally resolved photoluminescence quenching (SR-PLQ) method .................... 34
2.2.4 Organic photovoltaics fabricated with the OVPD .................................................... 49
2.2.5 Organic photovoltaics fabricated with the VTE ....................................................... 56
2.2.6 Substrate holder heat conduction study .................................................................... 57
2.2.7 Optical electrical field distribution technique ........................................................... 64
2.3 CONCLUSIONS ................................................................................................................ 80
2.4 EXPERIMENTAL .............................................................................................................. 81
2.4.1 Substrate preparation ................................................................................................ 81
2.4.2 Device architecture ................................................................................................... 82
2.4.3 Fabrication of organic solar cells .............................................................................. 82
2.4.4 OPV testing ............................................................................................................... 83
V

2.4.5 Synthesis of platinum 1, 3-di-tert-butylphenyl Tetraphenylbenzoporphyrin ........... 83
2.5 CHAPTER 2 REFERENCES................................................................................................ 85
CHAPTER 3. METAL DEPOSITION FOR OPTOELECTRONIC DEVICES USING A
LOW VACUUM VAPOR PHASE DEPOSITION (VPD) .......................................................... 88
3.1. INTRODUCTION .............................................................................................................. 88
3.2. RESULTS AND DISCUSSION ............................................................................................. 90
3.2.1. Deposition parameters .......................................................................................... 90
3.2.2. Magnesium and zinc metal films .......................................................................... 94
3.2.3. Organic light-emitting devices with magnesium cathodes ................................... 99
3.2.4. Organic photovoltaic devices with magnesium cathodes ................................... 102
3.3. CONCLUSIONS .............................................................................................................. 106
3.4. EXPERIMENTAL ............................................................................................................ 107
3.4.1. Characterization of metal films ........................................................................... 107
3.4.2. OLED and OPV substrate preparation ................................................................ 108
3.4.3. OLED and OPV device structure ........................................................................ 109
3.4.4. VPD device fabrication ....................................................................................... 110
3.4.5. VTE device fabrication ....................................................................................... 112
3.4.6. OLED and OPV testing ...................................................................................... 113
3.5. CHAPTER 3 REFERENCES.............................................................................................. 113
CHAPTER 4. FABRICATION OF HYBRID ORGANIC-INORGANIC LEAD BASED
SOLAR CELLS WITH A VAPOR PHASE DEPOSITION (VPD) .......................................... 119
4.1. INTRODUCTION ............................................................................................................ 119
4.2. RESULTS AND DISCUSSION ........................................................................................... 121
4.2.1. Hybrid organic-inorganic perovskite films characterization .............................. 121
4.2.2. Hybrid organic-inorganic perovskite solar cells fabricated in the VPD ............. 132
4.2.3. Hybrid organic-inorganic perovskite solar cells fabricated in the VTE ............. 140
4.3. CONCLUSIONS .............................................................................................................. 145
4.4. EXPERIMENTAL ............................................................................................................ 147
4.4.1. Characterization of perovskite films ................................................................... 147
4.4.2. OPV Substrate preparation ................................................................................. 148
4.4.3. VPD device fabrication ....................................................................................... 148
4.4.4. VTE device fabrication ....................................................................................... 150
4.4.5. OPV testing ......................................................................................................... 151
4.5. CHAPTER 4 REFERENCES.............................................................................................. 151

VI

List of Figures

Figure 1.1.- Schematic of the OVPD taken from Burrows, P. E.; Forrest, S. R.; Sapochak,
L. S.; Schwartz, J.; Fenter, P.; Buma, T.; Ban, V. S.; Forrest, J. L. Journal of Crystal
Growth 1995, 156, 91-98 .................................................................................................... 2
Figure 1.2.- Schematic of the LP-OVPD taken from Baldo, M. A.; Kozlov, V. G.; Burrows,
P. E.; Forrest, S. R.; Ban, V. S.; Koene, B.; Thompson, M. E. Appl. Phys. Lett.
1997, 71, (21), 3033-3035................................................................................................... 3
Figure 1.3.- Schematic of the organic vapor phase deposition ....................................................... 4
Figure 1.4.- Sublimation of organics as a function of temperature at constant carrier gas,
left, and as a function of carrier gas at constant temperature, right. Taken from
Shtein, M.; Gossenberger, H. F.; Benziger, J. B.; Forrest, S. R. J. Appl. Phys. 2001,
89, (2), 1470-1476............................................................................................................... 6
Figure 1.5.- Fluorescent organic light emitting diode architecture ................................................. 7
Figure 1.6.- Organic photovoltaic architecture ............................................................................. 11
Figure 1.7.- Photo-conversion mechanism of organic photovoltaics ........................................... 12
Figure 1.8.- Photovoltaic device characterization diagram .......................................................... 14
Figure 1.9.- Ideal diode circuit diagram, including series and shunt resistance ........................... 14
Figure 1.10.- Organic thin film transistor architecture ................................................................. 18
Figure 1.11.- Schematic of the modified vapor phase deposition (VPD) ..................................... 20

VII

Figure 2.1.- Best recorded research cell efficiency diagram from the national renewable
energy laboratory (NREL), obtained from http://www.nrel.gov/ncpv/ on October 9,
2015................................................................................................................................... 28
Figure 2.2.- Platinum tetra 1, 3-di-tert-butylphenyl tetrabenzoporphyrin Pt(
tbu
TPBP),
platinum tetraphenyltetrabenzoporphyrin Pt(TPBP) and buckminsterfullerene (C60)
........................................................................................................................................... 29
Figure 2.3.- Emission and excitation spectra of Pt(
tbu
TPBP) with different C60
concentrations in toluene .................................................................................................. 31
Figure 2.4.- Stern-Volmer quenching rate plot for Pt(
tbu
TPBP) for different C60
concentrations in toluene .................................................................................................. 32
Figure 2.5.- Flattened and 3D surface AFM images for a 20nm C60 film on silicon ................... 33
Figure 2.6.- Flattened and 3D surface AFM plot for a 40nm Pt(
tbu
TPBP) film on silicon ........... 33
Figure 2.7.- X-ray diffraction patterns of bare quartz substrate and 40nm Pt(
tbu
TPBP) and
20nm C60 films on quartz substrates ................................................................................. 34
Figure 2.8.- Film architecture and coordinates used for the analysis of Pt(
tbu
TPBP) using
the SR-PLQ method .......................................................................................................... 35
Figure 2.9.- Emission and absorption spectra of Pt(
tbu
TPBP), C60 and BCP ................................ 42
Figure 2.10.- Raw and corrected absorption spectra of a 400nm Pt(
tbu
TPBP) film on quartz
with a blocking and quenching layer ................................................................................ 43
Figure 2.11.- Raw and corrected absorption and transmittance spectra of a 10nm BCP and
10nm C60 film on quartz substrates ................................................................................... 44
Figure 2.12.- Excitation emission spectra of two Pt(
tbu
TPBP) films, with and without a
quencher layer, e.g., 10nm of C60, on quartz substrates.................................................... 45
VIII

Figure 2.13.- Excitation emission spectra of Pt(
tbu
TPBP) with 10nm of BCP and 10nm of
C60 under N2 and 0.08torr vacuum .................................................................................... 45
Figure 2.14.- Corrected wavelength offset photoluminescence quenching ratio versus alpha
prime under N2 and 0.08torr vacuum ................................................................................ 46
Figure 2.15.- Raw and corrected emission spectra of two 400nm Pt(
tbu
TPBP) films with
10nm of BCP and 10nm of C60 ......................................................................................... 47
Figure 2.16.- Emission spectra of bare films of BCP and C60 ...................................................... 47
Figure 2.17.- Raw and wavelength offset corrected photoluminescence quenching ratio vs.
alpha prime........................................................................................................................ 48
Figure 2.18.- Raw and wavelength offset corrected photoluminescence quenching ratio
versus alpha prime ............................................................................................................ 48
Figure 2.19.- Organic photovoltaic device architecture................................................................ 50
Figure 2.20.- Organic photovoltaic HOMO-LUMO energy diagram........................................... 50
Figure 2.21.- Current-Voltage curves and EQE spectra for 10nm Pt(
tbu
TPBP) devices made
in the OVPD ...................................................................................................................... 51
Figure 2.22.- HOMO LUMO energy diagram of CuPc vs Pt(
tbu
TPBP) ....................................... 53
Figure 2.23.- Current-Voltage curves and EQE spectra for 20nm Pt(
tbu
TPBP) devices made
in the OVPD ...................................................................................................................... 54
Figure 2.24.- Current-Voltage curves and EQE spectra for 45nm Pt(
tbu
TPBP) devices made
in the OVPD ...................................................................................................................... 55
Figure 2.25.- Current-Voltage curves and EQE spectra for 7, 13 and 20nm Pt(
tbu
TPBP)
devices made in the VTE .................................................................................................. 56
IX

Figure 2.26.- Heat conduction through a stainless steel substrate holder and a quartz
substrate diagram .............................................................................................................. 58
Figure 2.27.- Optical electric field schematic inside an OPV device ........................................... 65
Figure 2.28.- Refractive indeces and extinction coefficients of ITO, Pt(
tbu
TPBP), C60, BCP
and aluminum.................................................................................................................... 73
Figure 2.29.- Optical electric field for different thicknesses of a Pt(
tbu
TPBP) device at
350nm and 430nm wavelengths........................................................................................ 74
Figure 2.30.- Power absorbed for a Pt(
tbu
TPBP) device for 350nm and 430nm wavelengths
........................................................................................................................................... 75
Figure 2.31.- Optical electric field for a device with 20nm Pt(
tbu
TPBP) film thickness .............. 76
Figure 2.32.- Power absorbed for a device with 20nm Pt(
tbu
TPBP) film thickness ..................... 76
Figure 2.33.- Measured absorption and reflection inside a device with 20nm Pt(
tbu
TPBP)
film thickness .................................................................................................................... 77
Figure 2.34.- Modeled current density of a 10nm and 20nm Pt(
tbu
TPBP) film thickness
device ................................................................................................................................ 78
Figure 2.35.- Modeled incident photon to current efficiency of a 10nm and 20nm
Pt(
tbu
TPBP) film thickness device ..................................................................................... 78
Figure 2.36.- Modeled exciton density profile for a 20nm Pt(
tbu
TPBP) film thickness device
........................................................................................................................................... 79
Figure 2.37.- Modeled exciton density profile for a 40nm C 60 film thickness device .................. 79
Figure 2.38.- Synthesis steps of Platinum tetra 1, 3-di-tert-butylphenyl tetrabenzoporphyrin
Pt(
tbu
TPBP) ........................................................................................................................ 84
Figure 2.39.- Chemical structures of Pt(
tbu
TPBP), BCP and C60 .................................................. 84
X

Figure 3.1.- Schematic of the modified vapor phase deposition ................................................... 92
Figure 3.2.- Cross-sectional SEM images of a 200 nm Mg film on a silicon substrate made
using either VPD (top x 8.5K) or VTE (bottom x 55K). The VPD film was
deposited at a substrate holder temperature of 20-25°C ................................................... 95
Figure 3.3.- 3D AFM image of 200 nm Mg films deposited on a silicon substrate using
either VPD (top) or VTE (bottom) methods ..................................................................... 95
Figure 3.4.- XRD patterns of 200 nm Mg films deposited on a silicon substrate using either
VPD (top) or VTE (bottom) methods. Blue lines are reference crystal planes of the
hexagonal phase of Mg ..................................................................................................... 96
Figure 3.5.- VPD (x 15K) Cross-sectional SEM images of 400nm Zn film on a silicon
substrate. Film was deposited at a substrate holder temperature of 20-25°C ................... 98
Figure 3.6.- VPD XRD patterns of 400nm Zn films deposited on a silicon substrate. Blue
lines are reference crystal planes for the Hexagonal phase of zinc from (MDI Jade
9 PDF# 03-065-3358) ....................................................................................................... 98
Figure 3.7.- External quantum efficiency/brightness vs voltage characteristics of (a)
fluorescent Alq3 and (b) phosphorescent Ir(ppy)3 OLEDs. Current density vs
voltage characteristics can be seen in the inset ............................................................... 100
Figure 3.8.- Normalized EL intensity of green Alq3 fluorescent OLEDs ................................... 101
Figure 3.9.- Normalized EL intensity of green Ir(ppy)3 phosphorescent OLEDs ...................... 101
Figure 3.10.- Current-Voltage (IV) curves of a CuPc/C 60/Mg OPV fully fabricated in the
VPD................................................................................................................................. 102
XI

Figure 3.11.- External quantum efficiency (%) of a CuPc/C60/Mg OPV fully fabricated in
the VPD ........................................................................................................................... 103
Figure 3.12.- Current vs voltage (IV) curves of CuPc/C 60/PTCBI/Mg OPVs. External
quantum efficiency (%) can be seen in the inset ............................................................. 104
Figure 3.13.- Current-Voltage (IV) curves of CuPc/C60/PTCBI/Mg OPVs fully fabricated
in the VTE. Organic films of device S2 were exposed to air for ~ 5 minutes while
organic films of device S1 were not exposed to air ........................................................ 105
Figure 3.14.- External quantum efficiency (%) of CuPc/C 60/PTCBI/Mg OPVs fully
fabricated in the VTE. Organic films of device S2 were exposed to air for ~ 5
minutes while organic films of device S1 were not exposed to air ................................ 106
Figure 3.15.- Phosphorescent and fluorescent OLED and OPV device configuration ............... 109

Figure 4.1.- (a) crystal structure of perovskites and (b) absorbance spectra of
CH3NH3PbBr3 and CH3NH3PbI3 .................................................................................... 120
Figure 4.2.- Non-annealed one-step 330nm perovskite films with different CH3NH3I
deposition rates on glass made in the VPD, (a) absorbance spectra and (b) XRD
patterns ............................................................................................................................ 123
Figure 4.3.- Annealed and non-annealed films one-step 160nm perovskite films on ITO
made in the VTE, (a) absorbance spectra and (b) XRD patterns .................................... 123
Figure 4.4.- SEM images of a 330nm MAPbI3 film on glass made in the VPD, (a) magnified
x 400 and (b) magnified x 2k .......................................................................................... 125
Figure 4.5.- High resolution XPS core level spectra of a 330nm MAPbI3 film on ITO made
in the VPD, (a) Pb 4f spectra from PbI2 and (b) I 3d spectra from the PbI2 & MAI ...... 125
XII

Figure 4.6.- 3D AFM images of a 330nm MAPbI3 film on glass, (a) VPD and (b) VTE .......... 126
Figure 4.7.- Inductively coupled plasma optical emission spectrometry (ICP-OES) lead
metal traces concentration study of MAPbI3 films made in the VPD and VTE. (a)
Standards target wavelength of 220.335nm and (b) Standards target wavelength of
216.999nm ...................................................................................................................... 126
Figure 4.8.- 330nm PbI2 films made in the VPD and VTE. Substrate #1 of both VPD and
VTE were annealed for 1h at 90°C before dipping into a CH3NH3I solution
(10mg/mL) in 2-propanol ............................................................................................... 129
Figure 4.9.- 330nm perovskite films made in the VPD and VTE. PbI2 films were dipped
into a CH3NH3I solution (10mg/mL) in 2-propanol for 3min, then annealed for 1h
at 90°C ............................................................................................................................ 129
Figure 4.10.- Two-step 330nm perovskite films made in the VPD and VTE, (a) absorbance
spectra and (b) XRD patterns .......................................................................................... 129
Figure 4.11.- Two-step 330nm perovskite films on ITO fully made in the VPD, (a)
absorbance spectra and (b) XRD patterns ....................................................................... 131
Figure 4.12.- Profilometer profiles of two-step 330nm perovskite films on ITO made in,
(a) VPD and (b) VTE ...................................................................................................... 131
Figure 4.13.- 330nm MAPbI3 devices made in the VPD, (a) current vs voltage (IV) curves
and (b) external quantum efficiency, S1 circle – ITO/MAPbI3/C60/BCP/Al and S2
square – ITO/MAPbI3/NPD-MoO3/Ag ........................................................................... 134
Figure 4.14.- HOMO-LUMO gap diagram for MAPbI3 with C60 (acceptor) and NPD
(donor)............................................................................................................................. 134
XIII

Figure 4.15.- 330nm MAPbI3 device made in the VPD, ITO/MAPbI3/C60/BCP/Al, (a)
current vs voltage (IV) curves and (b) external quantum efficiency,
S3 circle – small excess of PbI2, S4 square – excess of MAI, S5 triangle – ideal
film, and S6 diamond – excess of PbI2 ........................................................................... 136
Figure 4.16.- XRD patterns of 330nm MAPbI3 films made in the VPD,
ITO/MAPbI3/C60/BCP/Al, S3 – small excess of PbI2, S4 – excess of MAI,
S5 – ideal film, and S6 – excess of PbI2 ......................................................................... 137
Figure 4.17.- 160nm MAPbI3 device made in the VPD, S7 – S8 ITO/MAPbI3/C60/BCP/Al
and S9 – S10 FTO/TiO2/MAPbI3/NPD-MoO3/Al (a) current vs voltage (IV) curves
and (b) external quantum efficiency, S7 and S9 are annealed and S8 and S10 are
not annealed .................................................................................................................... 139
Figure 4.18.- 330nm MAPbI3 device made in the VPD, circle ITO/MAPbI3/C60/BCP/Ag
and square ITO/MAPbI3/Spiro-OMeTAD/Ag (a) current vs voltage (IV) curves and
(b) XRD patterns ............................................................................................................. 140
Figure 4.19.- 160nm MAPbI3 devices made in the VTE, S11 – S14
ITO/MAPbI3/C60/BCP/Al, (a) current vs voltage (IV) curves and (b) external
quantum efficiency.......................................................................................................... 142
Figure 4.20.- 330nm MAPbI3 device made in the VTE, S15 – S18
ITO/MAPbI3/C60/BCP/Al, (a) current vs voltage (IV) curves and (b) external
quantum efficiency.......................................................................................................... 143
Figure 4.21.- 330nm MAPbI3 device made in the VTE, (a) current vs voltage (IV) curves
and (b) XRD patterns, S19 square – NPD/Au, S20 circle – C60/BCP/Au,
S21 triangle – Au, and S22 diamond – NPD-MoO3/Au ................................................. 144
XIV

Figure 4.22.- Schematic of the vapor phase deposition ............................................................. 150

XV

List of Tables

Table 2.1.- Photovoltaic performance parameters for 10nm Pt(
tbu
TPBP) devices made in
the OVPD .......................................................................................................................... 52
Table 2.2.- Photovoltaic performance parameters for 20nm Pt(
tbu
TPBP) devices made in
the OVPD .......................................................................................................................... 54
Table 2.3.- Photovoltaic performance parameters for 45nm Pt(
tbu
TPBP) devices made in
the OVPD .......................................................................................................................... 55
Table 2.4.- Photovoltaic performance parameters for 7, 13 and 20nm Pt(
tbu
TPBP) devices
made in the VTE ............................................................................................................... 57
Table 2.5.- Water and nitrogen thermal conductivity, density, viscosity, heat capacity,
among others ..................................................................................................................... 58
Table 2.6.- Measured and calculated substrate holder and substrate temperature chart at
several temperature conditions ......................................................................................... 64

Table 3.1.- Chapman–Enskog diffusion coefficients and sublimation enthalpies for Zn, Mg,
CuPc, C60, Alq3 and NPD ................................................................................................. 92
Table 3.2.- Resistivity measurements of a VPD 400nm Zn film deposited on a silicon
substrate ............................................................................................................................ 97
Table 3.3.- Fluorescent and phosphorescent OLED device performance .................................. 100
Table 3.4.- Organic photovoltaics performance, CuPc/C 60/PTCBI/Mg, fully made in the
VPD. Substrate (S), Device (D) ...................................................................................... 104
XVI

Table 3.5.- Organic photovoltaics performance, CuPc/C 60/PTCBI/Mg, fully made in the
VTE. Organic films of device S2 were exposed to air for ~ 5 minutes while organic
films of device S1 were not exposed to air. Substrate (S), Device (D) .......................... 105
Table 3.6.- Temperature gradient conditions for low/high sublimation materials in the VPD
......................................................................................................................................... 111
Table 3.7.- Sublimation temperature conditions of organics/metals in the VPD ....................... 112

Table 4.1.- ICP-OES intensity emission of various wavelengths targeting traces of lead
metal. 240nm MAPbI3 films made in the VPD and VTE were digested in a 5% nitric
acid solution in DI water ................................................................................................. 127
Table 4.2.- Relative density calculations of 240nm perovskite films on glass made in the
VPD and VTE ................................................................................................................. 131
Table 4.3.- 330nm MAPbI3 device performance made in the VPD, (S1)
ITO/MAPbI3/C60/BCP/Al and (S2) ITO/MAPbI3/NPD-MoO3/Ag ................................ 133
Table 4.4.- Device performance parameters of 330nm MAPbI3 devices made in the VPD,
ITO/MAPbI3/C60/BCP/Al, S3 – small excess of PbI2, S4 – excess of MAI,
S5 – ideal film, and S6 – excess of PbI2 ......................................................................... 135
Table 4.5.- 160nm MAPbI3 device performance made in the VPD, S7 – S8
ITO/MAPbI3/C60/BCP/Al and S9 – S10 FTO/TiO2/MAPbI3/NPD-MoO3/Al,
S7 and S9 are annealed and S8 and S10 are not annealed .............................................. 138
Table 4.6.- 160nm MAPbI3 device performance made in the VTE, S11 – S14
ITO/MAPbI3/C60/BCP/Al ............................................................................................... 141
XVII

Table 4.7.- 330nm MAPbI3 device performance made in the VTE, S15 – S18
ITO/MAPbI3/C60/BCP/Al ............................................................................................... 143
Table 4.8.-Temperature gradient conditions for sublimation of MAI and PbI2 in the VPD....... 150

XVIII

Abstract

The majority of my research focuses on fabrication and characterization of organic and
inorganic optoelectronics using a vapor phase deposition (VPD). The VPD, formerly called
organic vapor phase deposition (OVPD), was developed to allow sublimation and deposition of
organic, inorganic and metal molecules that exhibit high sublimation temperatures, ≤1000°C.
Conversely, the OVPD was specifically developed to allow sublimation and deposition of
organic-inorganic compounds that display sublimation temperatures below 400°C. Chapter 1
provides an introduction to the original OVPD method and its accomplishments fabricating
organic light emitting diodes (OLEDs), organic photovoltaics (OPVs) and organic thin film
transistors (OTFTs). Further, chapter 1 also discusses how the OVPD evolved into the VPD,
specifically to enable deposition of thin metal films for optoelectronic devices (discussed in
chapter 3).
In chapter 2 we present analysis and characterization of a novel molecule to fabricate OPVs
using the OVPD. Specifically, we measured the exciton diffusion length (LD) of platinum tetra 1,
3-di-tert-butylphenyl tetrabenzoporphyrin Pt(
tbu
TPBP), by use of the spectrally resolved
photoluminescence quenching (SR-PLQ) method and the optical electric field distribution
technique for optimization of OPVs. SR-PLQ measurements were performed under vacuum
(0.08torr), where an average LD of 12.1nm ± 2.4nm was found for Pt(
tbu
TPBP). Concurrently,
analysis of the films under nitrogen was also performed. However, the LD varied significantly with
values ranging from 10nm to 120nm. Significant emission spectra intensity fluctuations were
observed during the measurements under nitrogen suggesting oxygen quenching of the porphyrin.
Clearly, oxygen quenching altered the LD thus revealing a weakness of the SR-PLQ method.
XIX

Likewise, using a program written in MATLAB language for the optical electric field distribution
technique, an LD of 16.9nm ± 4nm and 14.8nm ± 3.5nm was obtained for Pt(
tbu
TPBP) and C60,
respectively. Further, modeling of the optical electric field, photocurrent generation, exciton
diffusion profile density, absorption, reflection, among other was also completed using the optical
electric field distribution method. Both techniques proved to be great tools for the design and
optimization of OPVs. Photovoltaic cells were fabricated in the OVPD and in a vacuum thermal
evaporation (VTE) (reference). Fabricated OPVs had the following architecture:
ITO/Pt(
tbu
TPBP)
10-60nm
/C60
10-40nm
/BCP
10nm
/Al
100nm
. Open circuit voltages (Voc) in the order of 0.5V
and 0.65V were obtained for devices made in the OVPD and VTE, respectively. The results
suggest intermixing of the donor/acceptor layers during deposition of C60 in the OVPD devices,
thus leading to strong intermolecular interactions.
Further, intermixing is common when films are annealed and strong intermolecular
interactions are known to increase dark current, consequently, decreasing the Voc. The efficiencies
obtained for the Pt(
tbu
TPBP) devices were found to be 0.6% and 1.2% for the OVPD and VTE,
respectively. The efficiency discrepancy is due to poor current contribution from the C60 layer
(10nm for OVPD vs 40nm for VTE). The external quantum efficiencies (EQE) from the OVPD
devices clearly show weak contribution from the acceptor (C60), whereas a stronger current
contribution from C60 can be observed on the VTE devices. The results suggest efficiency can be
improved for devices made in the OVPD if a thicker C60 film is employed. Similarly, the Voc can
also be improved with proper substrate cooling to prevent intermixing of the donor/acceptor layers.
Further, a similar Jsc trend was observed on the Pt(
tbu
TPBP) versus the Pt(TPBP) devices,
confirming a short LD in both molecules is accountable for the decrease in current when a thicker
donor film is employed.
XX

In chapter 3 we introduce fully fabricated OPVs and OLEDs with the VPD, an improved
version of the OVPD. The objective of the improved/modified VPD was to enable deposition of
compounds with high sublimation temperatures e.g., metals and inorganics such as calcium, zinc,
cadmium, magnesium, antimony, bismuth, indium, silver, aluminum, zinc sulfide, and manganese,
among others. Of the above metals, magnesium (Mg) and zinc (Zn) were selected to fabricate
metal films and devices, as they have sublimation enthalpies that are comparable to organic
materials commonly used for OLEDs and OPVs. Magnesium and zinc films were deposited in the
VPD to prepare optoelectronic devices under low vacuum conditions, i.e. 1 torr. The thin metal
films, analyzed via scanning electron microscope (SEM), atomic force microscopy (AFM), x-ray
diffraction (XRD) and four-point probe resistivity measurements, revealed comparable
characteristics to metal films deposited in a VTE. Magnesium cathodes were fabricated for OLEDs
and OPVs. OLEDs were fully made in either the VPD or VTE employing
aluminum tris(8 hydroxyquinoline) [Alq3] as the green fluorescent emitter or
factris(2phenylpyridine)iridium [Ir(ppy)3] as the green emitting phosphor. Analysis of the OLED
devices made in the VPD showed external quantum efficiencies (EQE = 0.9 ±0.1%) and
(EQE = 7.6 ±0.6%) at a luminance of 100 cd/m
2
for the fluorescent and phosphorescent devices,
respectively. In addition, OPVs were fully fabricated by both methods employing copper
phthalocyanine (CuPc) and C60 as the donor/acceptor materials. Analysis of the OPV devices made
in the VPD showed a power efficiency of 0.5 ±0.1%, an open circuit voltage of 0.45 ±0.05% and
a fill factor of 0.50 ±0.05%.
The final chapter introduces what we thought was the next logical subject, after studying
organics (chapter 2) and metals (chapter 3), inorganic compounds. Chapter 4 presents hybrid
organic-inorganic lead based solar cells grown with the VPD, specifically, perovskite films i.e.,
XXI

methyl ammonium lead iodide (MAPbI3) as the active layer. Perovskite films were fabricated with
both the VPD and VTE and characterized via scanning electron microscopy (SEM), atomic force
microscopy (AFM), X-ray diffraction (XRD), ultraviolet-visible spectroscopy (Uv-Vis), X-ray
photoelectron spectroscopy (XPS), profilometry, spectroscopic ellipsometry, and inductively
coupled plasma optical emission spectrometry (ICP-OES). The films were found to be analogous
to perovskite films previously reported in literature. One-step and two-step deposition methods
were used to fabricate solar cells in the VPD. For the two-step method, efficiencies of 3.3 ± 0.3%
and 4.1 ± 0.4% and Voc of 0.6 ± 0.1V and 0.8 ± 0.1V were obtained for
ITO/MAPbI3(330nm)/C60(40nm)/BCP(10nm)/Ag and ITO/MAPbI3(330nm)/NPD-MoO3(140nm
@ 50% doping concentration)/Ag, respectively. Further, we found that having excess of PbI2
and/or CH3NH3I in the perovskite films considerably decreased device performance, i.e., current
(Jsc) and fill factor (FF) were impacted greatly. Conversely, fully converted perovskite films
proved to be critical to achieve diode like rectifying IV curves. Additionally, annealing also had a
great impact on device performance were a decreased on Jsc was observed on films deposited on
ITO electrodes, i.e., currents of 9.2 ± 0.6mA/cm
2
and 2.1 ± 0.4mA/cm
2
were observed

for annealed
and non-annealed devices, respectively. Contrariwise, annealing marginally impacted films
deposited on FTO/TiO2 thus suggesting that a much rougher TiO2 surface promotes crystallization
of the perovskite film while a flatter electrode, i.e., ITO, is not conducive for film crystallization.
However, regardless of the architecture used, we found that the devices fabricated in the VPD
were unreliable and highly unreproducible, displaying poor-rectifying IV curves and noticeably
weaker Jsc and/or EQE response from the devices. Several possible explanations as to why this
strange behavior happened were discussed. First, perovskite films were exposed to ambient air for
its completion so it is conceivable that ambient air adversely affected the perovskite films, i.e., air
XXII

and water diffused through the film creating problems at the ITO/perovskite interface. This
possibility is validated with devices made in the VTE, which exhibited acceptable rectifying IV
curves and comparable EQE and Jsc values. Second, unreacted (excess) PbI2 and/or MAI
influenced solar cells performance, delivering poor-rectifying IV curves and noticeable
discrepancies between Jsc and EQE values, whereas having a fully converted perovskite film lead
to acceptable rectifying IV curves and comparable EQE and Jsc values. Third, it is also possible
the perovskites films made in the VPD are not entirely crystalline, i.e., a blend of amorphous and
crystalline phase exists in the film. The latter is less probable since amorphous materials tend to
scatter X-rays in many directions, thus leading to bumps on the collected spectra. Whereas, the
periodicity of crystalline materials scatters X-rays only in certain directions, consequently, leading
to sharp peaks that are similar to those obtained in the collected XRD spectra. Regrettably, while
it is most likely that perovskite film damage at the interfaces is what caused the previously
mentioned problems, currently there are no methods/systems that can help us verify our theory.

1

Chapter 1. Organic vapor phase deposition (OVPD)
1.1 Introduction

Thin organic films have been used widely to fabricate optoelectronic devices such as
organic light emitting diodes (OLEDs), organic photovoltaics (OPVs) and organic thin film
transistors (OTFTs). The most commonly used techniques to deposit organics for optoelectronic
devices are spin cast and vacuum thermal evaporation (VTE). The latter, a high vacuum technique
and most used method to fabricate thin organic films, requires long pump down times to achieve
high vacuum (i.e., ≤10
-5
torr), has inefficient utilization of materials, poor film conformality and a
high expense incurred when deposition is applied on large substrates.
1
High vacuum is required to
achieve long mean free paths and thereby allow the atoms to reach the substrate without colliding,
thus realizing smooth, uniform films. However, deposition using VTE occurs along a ballistic
(directional line-of-sight) trajectory preventing the formation of conformal films on nonuniform
and large substrates. Therefore, interest has developed towards non-line-of-sight deposition
techniques such as the organic vapor phase deposition (OVPD). The OVPD was first introduced
in 1995 to fabricate organic thin films with large optical non-linearities.
2, 3
Specifically, it was used
to make films of 4’-dimethylamino-N-methyl-4-stilbazolium tosylate (DAST) by thermally
reacting, inside the OVPD chamber, 4’-dimethylamino-N-methyl-4-stilbazolium iodide (DASI)
with methyltysolate (MT). Film growth of DAST did not required cooling of the substrate, the film
was grown on a hot substrate and the temperature of the substrate was regulated with the
temperature of the chamber, Figure 1.1. The method relies on a carrier gas to move sublimed
gaseous molecules onto a substrate by forced convection. The carrier gas can be any inert gas and
2

in the above experiment, nitrogen was the carrier gas. Control of the carrier gas and pressure inside
the reactor/chamber is achieved by using mass flow controllers and a pump, respectively, while
temperature of the reactor/chamber is controlled with a furnace.

A second modified version of the OVPD was introduced in 1996, called LP-OVPD, to
deposit small molecular weight organics to make organic light emitting diodes (OLEDs), Figure
1.2.
4
However, this second version had a cooled substrate holder which was used to deposit
N’-diphenyl-N,N’-bis(3-methylphenyl)1-1’biphenyl-4-4’diamine (TPD) and aluminum
tris-(8 hydroxyquinoline) (Alq3) to fabricate a green fluorescent OLED with an external quantum
efficiency of 0.4 ± 0.05%. The above device was finalized by depositing lithium fluoride
(LiF – 1nm), a magnesium-silver alloy cathode (Mg-Ag – 50nm) caped with a 100nm protective
Ag layer using a vacuum thermal evaporation (VTE).
Two regimes, mass transport and kinetic, control film growth in vapor phase methods.
5

The first one, mass transport, happens when the growth rate is determined by the amount of

Figure 1.1.- Schematic of the OVPD taken from Burrows, P. E.; Forrest, S. R.; Sapochak, L.
S.; Schwartz, J.; Fenter, P.; Buma, T.; Ban, V. S.; Forrest, J. L. Journal of Crystal Growth 1995,
156, 91-98
3

material arriving onto the substrate. Further, slow carrier gas flow can form a boundary layer right
on the substrate, forcing the gaseous molecules to diffuse through this layer. Therefore, film
growth rate can be governed by two regimes in mass transport limit, transport and diffusion of
mass. The second, kinetic limit, is dependent on how fast reactions occur in the system. Low
pressures in the chamber increase the diffusion rate of the source materials thus creating a kinetic
limited process instead.

The design of the OVPD used in this research, Figure 1.3, is similar to the OVPD
previously described. The method relies on a carrier gas under laminar flow (fluid flow without
lateral mixing) to transport organic-inorganic vapor onto a cooled substrate by forced convection.
Laminar and turbulent flows are predicted using the dimensionless Reynolds number (Re), defined
as the ratio of inertia forces with respect to viscous forces.
6
Sublimation of the compound of
interest occurs when the source boat is introduced horizontally into the furnace. The carrier gas is

Figure 1.2.- Schematic of the LP-OVPD taken from Baldo, M. A.; Kozlov, V. G.; Burrows, P.
E.; Forrest, S. R.; Ban, V. S.; Koene, B.; Thompson, M. E. Appl. Phys. Lett. 1997, 71, (21),
3033-3035
4

fed through each source boat and a bypass, the latter to prevent vapor condensation in the rear
flange of the chamber. The OVPD can sublime four compounds during the same deposition,
however, only one crystal monitor sensor is used to measure film thickness. It’s possible to control
the rates of individual source boats when two or more compounds are sublimed by finding the
control parameters for a desired rate of each individual source boat, e.g., have a fixed temperature,
flow and pressure which will provide the desired deposition rate. Cooling of the substrate holder
is accomplished with a chiller and by maintaining a low pressure in the chamber, gas diffusivity is
improved thus promoting thickness and uniformity of the grown films.
6

1.2 Theory

The supply of sublimed material can be limited by the rate of evaporation or the flow of
the carrier gas. Organics-inorganics sublime from the source boat at a molar rate, revap, proportional
to the equilibrium vapor pressure of the material, 𝑃 𝑜𝑟 𝑔 𝑒𝑞
, as shown in Eq. (1.1).
7

Crystal monitor
Substrate holder
Vacuum
pump
Cooling supply
Zone 1 Zone 2 Zone 3 Zone 4
Source boat
Carrier gas
Shutters
Zone 1 Zone 2 Zone 3 Zone 4
Rear flange
Front flange
Carrier gas bypass

Figure 1.3.- Schematic of the organic vapor phase deposition
5

𝑟 𝑒 𝑣𝑎 𝑝 = 𝑘 𝑒 𝑣𝑎 𝑝 𝑃 𝑜𝑟 𝑔 𝑒𝑞

(1.1)
Where kevap is the kinetic evaporation factor. Conversely, the organic-inorganic vapor condenses
onto the substrate at a rate, rcond, proportional to the pressure of the material in the source boat,
Porg, as shown in Eq. (1.2).
𝑟 𝑒 𝑣𝑎 𝑝 = 𝑘 𝑒 𝑣𝑎 𝑝 𝑃 𝑜𝑟 𝑔 (1.2)
Where Kcond is a kinetic condensation factor. Contrariwise, organic-inorganic vapor is carried out
of the source boat at a rate, r, proportional to the volumetric flow rate of the carrier gas, V, and the
concentration of gaseous material in the source boat, Porg/RTcell, shown in Eq. (1.3).
𝑟 = 𝑉 ̇ 𝑃 𝑜𝑟 𝑔 𝑅 𝑇 𝑐𝑒 𝑙𝑙
(1.3)
Where Tcell is the temperature of the source boat and R is the universal gas constant. Combining
Eq. (1.1) and Eq. (1.3), a partial pressure of the gaseous vapor in the carrier gas stream can be
found, Eq. (1.4).
At very low carrier gas flow e.g., close to zero, the evaporation and condensation rate must
be equal thus kevap = kcond. Using Eq. (1.4) and Figure 1.4, it can be seen that at high evaporation
temperatures the gaseous material and the source material equilibrate i.e., Porg ~ 𝑃 𝑜𝑟 𝑔 𝑒𝑞
, thus the
concentration of material evaporating from the source is constant. Conversely, having high carrier
gas flows and low sublimation temperatures forces the gaseous material to be removed out of the
source boat as soon as it evaporates, driving the system towards the kinetic evaporation regime. In
the latter, the flux of evaporated material leaving the source boat is independent of the carrier gas
flow.
𝑃 𝑜𝑟 𝑔 𝑃 𝑜𝑟 𝑔 𝑒𝑞
=
𝑘 𝑒 𝑣𝑎 𝑝 𝑉 ̇ 𝑅𝑇
+ 𝑘 𝑐𝑜 𝑛𝑑
(1.4)
6

Equilibrium and kinetic partial pressures of organic-inorganic compounds leaving the
source boat can also be calculated by using Eq. (1.5) and Eq. (1.6), respectively. Clearly it can be
seen that the partial pressure depends exponentially on the temperature of the source boat.
Where ΔH
vap
and P0 are the enthalpy of evaporation and a pressure constant, respectively.

𝑃 𝑜𝑟 𝑔 𝑒𝑞
= 𝑃 0
exp ⁡ ( −
∆ 𝐻 𝑣𝑎 𝑝 𝑅 𝑇 𝑐𝑒 𝑙𝑙 )
(1.5)
𝑃 𝑜𝑟 𝑔 𝑘𝑖𝑛
= (
𝑅 𝑇 𝑐𝑒 𝑙𝑙 𝑉 ̇ ) 𝑘 𝑒 𝑣𝑎 𝑝 𝑃 0
exp ⁡ ( −
∆ 𝐻 𝑣 𝑎 𝑝 𝑅 𝑇 𝑐𝑒 𝑙𝑙 )
(1.6)

Figure 1.4.- Sublimation of organics as a function of temperature at constant carrier gas,
left, and as a function of carrier gas at constant temperature, right. Taken from Shtein, M.;
Gossenberger, H. F.; Benziger, J. B.; Forrest, S. R. J. Appl. Phys. 2001, 89, (2), 1470-1476
7

1.3 Applications
1.3.1 Organic light emitting diodes (OLEDs)

Organic light emitting diodes (OLEDs) are currently employed in a variety of applications
ranging from cell phones, displays, monitors, lighting bulbs, among others. The most simple
OLED architecture, shown in Figure 1.5, consists of a hole transport layer (HTL), e.g.,
N,N′Di[(1naphthyl)N,N′diphenyl]1,1′biphenyl)4,4′diamine (NPD) and an electron
transport layer (ETL), e.g., aluminum tris-(8 hydroxyquinoline) (Alq3). The latter, Alq3, is also the
emissive layer in the above fluorescent OLED device architecture. Further, it’s common to use an
electron injection layer (EIL), e.g., lithium fluoride (LiF), to aid in the injection of electrons from
the cathode into the organic layer.

The organic layers are typically positioned among two electrodes, the cathode and anode,
e.g., aluminum (Al) and indium tin oxide (ITO), respectively. Under an applied forward bias, the
cathode allows electrons to flow towards the anode while holes flow from the anode towards the
cathode. Delocalization of pi electrons caused by conjugation over the entirety or a portion of the
organic molecule is responsible for organic molecules to have electrically conductive properties.

Substrate/ITO
HTL
ETL
EIL
Cathode

Figure 1.5.- Fluorescent organic light emitting
diode architecture
8

In OLEDs, electrons are injected into the lowest unoccupied molecular orbital (LUMO) while
holes are in injected into the highest occupied molecular orbital (HOMO) of the organic
semiconductor. Further, excitons are formed when electrostatic forces bring the electrons and holes
together in the organic layers. An exciton is a bound neutral state, containing an electron and a
positive hole. Depending on how the electron and hole combined in the organic molecule, also
known as electronic transition processes, the exciton can either be in a singlet or triplet state.
Further, emission from singlet excited states is called fluorescence while emission from triplet
excited states is called phosphorescence. Because the formation of excitons is spin-independent,
the probability of forming triplets and singlets in an organic molecule is three to one in organic
molecules.
8
The latter ratio thus limits the maximum internal efficiency of fluorescence to 25%
while a maximum internal efficiency of 100% can be achieved in phosphorescence emission.
However, because triplet excited states are spin forbidden, introduction of a heavy metal ensues
strong spin-orbit coupling interactions thus allowing intersystem crossing (ISC) from singlet
excited states to triplet excited states.
9-11
In organic semiconductors, holes tend to have higher
mobility than electrons, thus excitons are generally formed closed to the emissive layer.
Additionally, excitons have two mechanisms of deactivation in OLEDs, namely radiative and
non-radiative processes. Among the most common non-radiative processes are vibrational
relaxation, internal conversion, triplet-triplet annihilation (TTA), triplet-polaron annihilation
(TPA), energy transfer and electron transfer.
12-14
Contrariwise, radiative processes result when
excitons relax to the ground state by emitting a photon with a given energy/frequency. Further, the
energy/frequency of the photon emitted is determined by the HOMO-LUMO gap of the organic
molecule.
9

The first OLED fabricated in the OVPD was a fluorescent device with the following
architecture, N’-diphenyl-N,N’-bis(3-methylphenyl) 1-1’biphenyl-4-4’diamine (TPD) and Alq3
and it achieved an external quantum efficiency (EQE) of 0.4 ± 0.05%.
4
The above device was
completed in a vacuum thermal evaporation (VTE) by depositing lithium fluoride (LiF – 1nm),
magnesium-silver (Mg-Ag – 50nm) and a protective layer of pure silver (Ag – 100nm). Several
device architectures followed, including a hybrid fluorescent OLED consisting of a polymer layer
of PEDT-PSS (spin cast), NPD – 40nm and Alq3 – 60nm which had a turn on voltage of 2.5V and
a power efficiency of 1.4 lm/W @ 100cd/cm
2
.
15
Phosphorescent devices were also fabricated in
the OVPD. The first phosphorescent device was green and its architecture was copper
phthalocyanine (CuPc – 10nm), NPD – 30nm, 4,4′Bis(Ncarbazolyl)1,1′biphenyl (CBP) doped
with factris(2phenylpyridine)iridium (Ir(ppy)3 @ 4.3% volume – 30nm),
aluminum(III)bis(2-methyl-8-quinolinato)-4phenalphenolate (BAlq2 – 10nm) and Alq3 – 40nm.
The green PHOLED achieved an EQE of 7% @ 1000 cd/cm
2
.
16
In the above device, CuPc,
LiF – 1nm and Al – 100nm were deposited with a VTE. Conversely, red phosphorescent devices
were also fabricated in the OVPD with the following architecture, NPD – 30nm,
4,4′Bis(Ncarbazolyl)1,1′biphenyl (CBP) doped with
bis(2-(2’-benzothienyl)pyridinato-N,C3’)acetyl-acetonate) (Btp2Ir(acac) @ 9% volume – 30nm),
bis(2-methyl-8-quinolinato)(para-phenylphenolato)aluminum(III) (BAlq – 10nm) and
Alq3 – 70nm. The red PHOLED achieved an EQE of 8.6% and 8.9 cd/A @ 100 cd/cm
2
.
17
To
complete the device, LiF – 0.6nm and Al – 100nm were deposited with a VTE. Further, OLEDs
made in a VTE were fabricated in parallel to compare efficiencies and it was found that both OVPD
and VTE techniques produced devices that displayed analogous performance. Another interesting
architecture used in the OVPD is layer cross-fading. Layer cross-fading is basically reducing the
10

deposition rate of one organic, e.g., hole transport layer, while increasing the deposition rate of the
subsequent layer, e.g., emissive layer. This technique was used to fabricate a red phosphorescent
OLED that exhibited an EQE of 18%, 36.5 cd/A and 33.7 lm/W @ 1000 cd/cm
2
.
18

White OLEDs for illumination applications have also attracted interest thus the next logical
step was to fabricate them in the OVPD. White OLEDs consisting of blue fluorescent compounds
combined with green and red phosphorescent materials, also known as hybrid OLEDs since a
fluorescent and phosphorescent material is used in the same device, have shown efficiencies of
18.9 cd/A to 28 cd/A and 13.3 lm/W to 20.6 lm/W @ 100cd/cm
2
.
19, 20
The above devices were
fabricated using mostly compounds that are intellectual property (IP) and the only materials that
were reported in the architecture are Ir(ppy)3, NPD, CBP, LiF, and Al. Yet again, LiF and Al were
deposited using a VTE for the above devices.

1.3.2 Organic photovoltaics (OPVs)

Thin film organic photovoltaics (OPVs) have attracted considerable interest as an
alternative source of energy in the past years. The key in OPVs success is the use of low cost
organic molecular and polymeric materials to enhance efficiency. Similarly to OLEDs, organic
semiconductors in OPVs are typically positioned among two electrodes (Figure 1.6), the cathode
and anode, e.g., aluminum (Al) and indium tin oxide (ITO), respectively.
11

However, while in OLEDs excitons are created under an applied forward bias to generate
light (electroluminescence), in OPVs excitons are produced when the active organic layers absorb
photons (light) to generate carriers. Consequently, the operation processes of OLEDs and OPVs
are contrasting to each other, i.e., injection of carries in OLEDs vs extraction of carries in OPVs.
Further, similar to OLEDs, OPVs also have hole and electron transport layers, typically called
donor (D) and acceptor (A) layers. Upon light absorption and thus generation of excitons, a
concentration gradient forces the excitons to diffuse towards the D/A interface, where the donor
material donates electrons and the acceptor material withdraws electrons, consequently, making
free carriers that are extracted at the electrodes. There are five steps for the photo-conversion of
OPVs, light absorption, exciton diffusion, charge separation, charge transport and charge
collection as illustrated in Figure 1.7.

Substrate/ITO
Donor
Acceptor
Buffer
Cathode

Figure 1.6.- Organic photovoltaic architecture
12

The first step, light absorption, occurs when a photon is absorbed by either the donor or
acceptor layer, which in turn produces an exciton. Excitons are bound neutral states, containing an
electron and a positive hole. The second step is for the created exciton(s) to diffuse to the D/A
interface. It’s worth mentioning that the majority of organic molecules used in OPVs have an
exciton diffusion length (LD) of ~5-20nm.
21
This short LD is one of the limiting factors for OPVs
to achieve higher efficiencies.
22
The third step, charge separation, occurs at the D/A interface and
is driven by a LUMO-LUMO and HOMO-HOMO energy offset of the donor and acceptor
materials. This HOMO and LUMO energy offset is commonly found to be 0.3eV or higher and
during this process, instead of truly free carrier charges being created, a charge transfer (CT) state
is formed. The fourth step is charge transport, where the CT state permits separation of the charges

Figure 1.7.- Photo-conversion mechanism of organic photovoltaics

13

and via a hoping mechanism the carriers reach the electrode interface where the fifth step occurs,
charge collection.
Some of the most important parameters when analyzing and characterizing an OPV are the
open circuit voltage (Voc),
23-25
short circuit current (Jsc),
26
fill factor (FF), device power efficiency
( η, also known as Pmax), optical density and exciton diffusion length (LD).
27, 28
Among the latter,
Voc, J sc, FF and η, are the four parameters utilized to characterized solar cells, Figure 1.8. These
parameters are extracted from current (J) and voltage (V) curves, commonly known as JV or IV
curves. These JV curves are described using the ideal diode equation and the ideal diode circuit
diagram, Eq. (1.7) and Figure 1.9, respectively. Current-voltage curves are obtained when a reverse
and forward bias, under dark and illumination conditions, are applied to both electrodes of the
device, Figure 1.8. The Jsc parameter is obtained where the curve intersects the vertical axis (J)
and it provides the amount of current drawn from the device when a load is applied to the circuit.
The Voc is located where the curve intersects the horizontal axis (V) and it provides the voltage
developed when the circuit has infinite resistance or open circuit. The FF is the ratio of the
maximum current and voltage of the device, Pmax = Jmax · Vmax, to the measured Jsc and Voc,
depicted in Figure 1.8.
14

0.9
Dark Current
Light Current
Voltage (V)
Current (a.u.)
-0.3 0.3 0.6 0.0
J
sc
V
oc
P
max
J
sc
x V
oc

Figure 1.8.- Photovoltaic device characterization diagram

Figure 1.9.- Ideal diode circuit diagram, including series and shunt
resistance
R
sa
V
D
A
J
dark
+
-
J
sc
R
sh
15

These three parameters are then used to calculate the maximum efficiency, defined as the
power density delivered by the device with respect to the incident light power density (P),
η = Jmax ·Vmax/P, which can be rewritten to obtain the maximum efficiency, Eq. (1.7).
For an ideal diode, the dark current density Jdark is described by Eq. (1.8), where Js is the
reverse saturation dark current, kB is Boltzmann’s constant, T is temperature in degrees Kelvin, q
is the elementary charge, and n is the ideality factor.
The overall current voltage response of the device can be approximated as the sum of the
Jsc and Jdark. Although the reverse current is not the equal to the current that flows in the dark, the
estimate is realistic for numerous OPVs and can be represented by Eq. (1.9), which becomes Eq.
(1.10) for an ideal diode.
The open circuit voltage provides the voltage developed when the circuit has infinite
resistance or open circuit and is equivalent to the situation where Jdark and Jsc cancel out. From
Eq. (1.10), we can derive a Voc, given by Eq. (1.11).
OPVs are commonly described using an ideal diode circuit diagram and if we include
parasitic resistances, series resistance (Rsa) and shunt resistance (Rsh), the electric circuit diagram
𝜂 𝑚 𝑎𝑥 = 𝐽 𝑠𝑐
∙ 𝐹𝐹 ∙ 𝑉 𝑜𝑐
(1.7)

(1.8)

(1.9)

(1.10)

(1.11)

  1  
T
B
nk
qV
e J J
s dark
sc dark
J J J  
 
sc s
J e J J
T
B
nk
qV
   1








  1 ln
s
sc
oc
J
J
q
nkT
V
16

can be described with Figure 1.9. The series resistance in the device occurs due to material
resistance to current flow and resistive contacts. The parallel or shunt resistance arises from current
leakage through the device, around the boundaries of the device and between contacts of different
polarity. Both parasitic resistances directly affect fill factor and in order to maximize device
efficiency, small Rsa and large Rsh are necessary. When parasitic resistances are included, the diode
equation can be converted into Eq. (1.12).
Lamellar (planar) and bulk heterojunction (BHJ) are the two most common types of organic
photovoltaic structures. The first, lamellar (Figure 1.6), has well-defined alternating interfaces or
layers (e.g., donor-acceptor) while the second, BHJ, is a blend of the donor acceptor. The latter,
increases the interfacial area that results in a higher probability for excitons to reach an interface
and dissociate into charge carriers. Because excitons have a short lifetime, having a BHJ structure
is beneficial, i.e., greater efficiencies are generally observed. Bulk heterojunction structures are
normally realized by spin casting a blend of the donor-acceptor material or by annealing a lamellar
device. Both, lamellar and BHJ structures, have been fabricated by OVPD.
29
However, whereas
BHJ films grown by VTE require annealing of the lamellar layers, BHJ structures prepared in the
OVPD are possible without requiring annealing. The devices had the following architecture,
CuPc – 45nm, 3, 4, 9, 10 perylenetetracarbonyl bisbenzimidazole (PTCBI – 50nm), BCP – 10nm,
and Ag – 100nm, where BCP and Ag were deposited using a VTE. Planar and BHJ devices
fabricated in the OVPD revealed efficiencies of 1.1 ± 0.1% and 2.7 ± 0.1%, respectively.
Analogous efficiencies were obtained for lamellar VTE devices. Conversely, VTE annealed BHJ
presented a significant lower efficiency of 1.4 ± 0.1%. In contrast, slightly modified BHJ devices

(1.12)
 
 
sh
s
sc s
R
JR V
J e J J
T
B
nk
s
JR V q

   

1
17

fabricated in the OVPD have been shown to have efficiencies ranging from 1.6% to 4.4 ± 0.2%.
30,
31
Specifically, the above efficiencies were reached by using a different acceptor, i.e., C60. Further,
planar CuPc – C60 devices have also been realized in the OVPD, displaying efficiencies analogous
to VTE devices, i.e., 1.0 ± 0.1%.
32

1.3.3 Organic thin film transistors (OTFTs)

Organic thin film transistors (OTFTs) have attracted attention in the past few years as a
low cost electronic component for flexible displays and monitors. Flexibility and fabrication
simplicity make OTFTs extremely appealing. Fabrication of OTFTs is far less complex than their
silicon based counterpart, which requires high temperature, high vacuum and photolithography
patterning techniques.
33
An organic thin film transistor is a three terminal device, where an applied
voltage to a gate electrode controls current flow between a source and drain electrode under an
imposed bias, Figure 1.10. Usage of a third terminal in the source-drain contacts enables OTFTs
to be used as switches. Their value as a switch is measured by the mobility, μ, which describes
charge carriers transport within the active layer under the influence of an electric field, thus
mobility is a measure of how fast a device can switch between on and off. Mobility is extracted
from current-voltage measurements with representative values ranging from 0.1 – 1 cm
2
/V·s for
amorphous-Si (a-Si) devices, with champion organic materials achieving mobilities of 1 – 10
cm
2
/V·s.
34, 35
Several methods have been used to fabricate OTFTs, among the most common are
solution precipitation, vacuum thermal evaporation, and organic molecular beam deposition.
36-38

Specifically, pentacene based OTFTs were fabricated using the above techniques.
18

The first OTFTs fabricated with the OVPD consisted of an n-doped silicon substrate (gate)
with a thermally grown silicon dioxide (SiO2) layer (dielectric layer) followed by a 20 – 5000nm
crystalline grain size pentacene film.
39
The above device was finalized by depositing 50nm of gold
(Au), source and drain contacts, with a VTE. Field effect hole mobilities of 0.05 ± 0.02 to 0.5 ± 0.1
cm
2
/V·s were obtained for the above device, respectively. Further, pentacene organic transistors
with mobilities of 0.9 – 1.5 cm
2
/V·s have been realized by varying the deposition conditions in the
OVPD.
40-42

1.4 Organic vapor phase modifications

Several modifications were made to the OVPD, however, among the most significant are
installation of retractable shutters, construction of an aluminum substrate holder and redesign of
the OVPD chamber to deposit metals. The shutters were installed to prevent organic condensation
while the desired deposition conditions were achieved. However, the original shutter design left
the shutters exposed to the hot zone by only allowing the shutters to flip down, thus causing cross-
contamination. A modification was made to enable the shutter to retract into the cold zone, behind
the substrate holder, to avert sublimation of condensed materials on the shutter and consequently

Gate
Dielectric Layer
Organic Layer
Source Drain

Figure 1.10.- Organic thin film transistor
architecture
19

prevent cross contamination during subsequent depositions. The substrate holder faced several
modifications, though the most important was to use aluminum as the core body to improve
cooling of the substrates. The third and last major modification was prompted due to high
diffusivities of metals. Calculation of diffusivities of organics and metals is discussed in more
detail in chapter 3. Certain metals exhibited high diffusivities that promoted metal condensation
in various regions of the sample chamber upstream from the source nozzles. Specifically, metals
and organics with high diffusivities, i.e., zinc, magnesium, C60, CuPc, etc., were condensing on
the rear flange of the chamber, shown on Figure 1.3. The above problem affected the quality of
the films rendering nonuniform films. Condensation on the rear flange can potentially be averted
by increasing the total flow i.e., carrier gas bypass flow. However, an increase in carrier gas flux
can raise the operating pressure if the pumping system is not suitable for large flows, thus affecting
film morphology and crystallinity. In our system, to realize an operating pressure of 1torr, the
maximum total flow could not exceed 40 standard cubic centimeters per minute (sccm). The
inability to pump at a higher velocity necessitated a modification in the deposition apparatus by
inserting a fused silica wall around the source tubes between zone 3 and zone 4 of our system, seen
in Figure 1.11. This fused silica wall is unnecessary when depositing materials with low
diffusivity, e.g., NPD and Alq3. The modified system was renamed organic vapor phase (VPD) for
its ability to now deposit metals (see chapter 3).
20

1.5 Conclusions

Since its inception from 1995 to date, the OVPD has demonstrated to be a versatile, reliable
and efficient method to make organic thin films.
2
Specifically, the OVPD was designed for organic
materials that have low sublimation temperatures. Upon sublimation, the material is transported
by an inert carrier gas through a hot-wall vessel to avoid unwanted deposition prior to condensation
on a cooled substrate, thereby achieving high material usage. The rate at which the material
sublimes and ultimately condenses onto the substrate is controlled by both the temperature of the
source boat and the flow rate of the inert gas, which enhances control over the deposition process
and the morphology of the films.
7, 39, 43
The OVPD has successfully been used to deposit molecular
organic materials, forming high efficiency OLEDs,
4, 16, 19, 20, 44, 45
OPVs,
31, 32, 46, 47
and TFTs.
39-42,
48-50
Further, an improved VPD method permitted deposition of thin metals films, i.e., magnesium
and zinc, that enabled the system to fully fabricate OLEDs and OPVs, validating its exceptional
value for the fabrication and study of organic optoelectronics.
51

Crystal Monitor
Substrate Holder
Vacuum
Pump
Cooling Supply
Zone 1 Zone 2 Zone 3 Zone 4
Source Boat
Fused
Silica wall
Carrier Gas
Shutters
Zone 1 Zone 2 Zone 3 Zone 4

Figure 1.11.- Schematic of the modified vapor phase deposition (VPD)
21

1.6 Chapter 1 References

1. Biswas, S.; Shalev, O.; Shtein, M. Annual review of chemical and biomolecular
engineering 2013, 4, 289-317.
2. Burrows, P. E.; Forrest, S. R.; Sapochak, L. S.; Schwartz, J.; Fenter, P.; Buma, T.; Ban,
V. S.; Forrest, J. L. Journal of Crystal Growth 1995, 156, 91-98.
3. Forrest, S. R.; Burrows, P. E.; Stroustrup, A.; Strickland, D.; Ban, V. S. Applied Physics
Letters 1996, 68, (10), 1326-1328.
4. Baldo, M. A.; Kozlov, V. G.; Burrows, P. E.; Forrest, S. R.; Ban, V. S.; Koene, B.;
Thompson, M. E. Appl. Phys. Lett. 1997, 71, (21), 3033-3035.
5. Stringfellow, G. B., Organometallic vapor-phase epitaxy : theory and practice. 2nd ed.;
Academic Press: San Diego, 1999; p xxiv, 572 p.
6. Bird, R. B.; Stewart, W. E.; Lightfoot, E. N., Transport Phenomena, Revised 2nd Edition.
John Wiley & Sons, Inc.: 2006; p 905.
7. Shtein, M.; Gossenberger, H. F.; Benziger, J. B.; Forrest, S. R. J. Appl. Phys. 2001, 89,
(2), 1470-1476.
8. Baldo, M. A.; O'Brien, D. F.; Thompson, M. E.; Forrest, S. R. Phys Rev B 1999, 60, (20),
14422-14428.
22

9. Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R. Applied
Physics Letters 1999, 75, (1), 4-6.
10. Adachi, C.; Baldo, M. A.; Forrest, S. R.; Thompson, M. E. Applied Physics Letters 2000,
77, (6), 904-906.
11. Baldo, M. A.; Forrest, S. R. Phys Rev B 2000, 62, (16), 10958-10966.
12. Buckley, A. Woodh Pub Ser Elect 2013, (36), 1-647.
13. Baldo, M. A.; Adachi, C.; Forrest, S. R. Phys Rev B 2000, 62, (16), 10967-10977.
14. Kalinowski, J.; Stampor, W.; Mezyk, J.; Cocchi, M.; Virgili, D.; Fattori, V.; Di Marco, P.
Phys Rev B 2002, 66, (23).
15. Hartmann, S.; Janssen, A.; Gersdorff, M.; Meyer, N.; Reinhold, M.; Schwambera, M.;
Strauch, G.; Marheineke, B.; Heuken, M.; Johannes, H. H.; Kowalsky, W.; Himcinschi, C.;
Zahn, D. SID Symposium Digest of Technical Papers 2004, 35, (1), 1388-1391.
16. Zhou, T. X.; Ngo, T.; Brown, J. J.; Shtein, M.; Forrest, S. R. Appl. Phys. Lett. 2005, 86,
021107-1-021107-3.
17. Hartmann, S.; Janssen, A.; Johannes, H. H.; Caspary, R.; Kowalsky, W.; Schwab, H.;
Bertram, D.; Blei, H.; Meyer, N.; Marheineke, B.; Heuken, M. SID Symposium Digest of
Technical Papers 2006, 37, (1), 1563-1566.
18. F. Lindla, M. B., C. Zimmermann, F. Jessen, P. V. Gemmerna, D. Bertram, D. Keiper, N.
Meyer, M. Heuken, H. Kalisch and R. H. Jansen. MRS Proceedings 2009, 1154, B05-07.
23

19. Boesing, M.; Lindla, F.; Zimmermann, C.; Gemmern, P. v.; Bertram, D.; Keiper, D.;
Meyer, N.; Heuken, M.; Kalisch, H.; Jansen, R. H. Mater. Res. Soc. Symp. Proc. 2010, 1212,
1212-S03-08.
20. Lindla, F.; Gemmern, P. v.; Keiper, D.; Heuken, M.; Kalisch, H.; Jansen, R. H. Journal
of Photonics for Energy 2011, 1, (1), 011013.
21. Terao, Y.; Sasabe, H.; Adachi, C. Applied Physics Letters 2007, 90, (10).
22. Ray, B.; Alam, M. A. Sol Energ Mat Sol C 2012, 99, 204-212.
23. Gong, X.; Tong, M.; Brunetti, F. G.; Seo, J.; Sun, Y.; Moses, D.; Wudl, F.; Heeger, A. J.
Advanced Materials 2011, 23, (20), 2272-2277.
24. Perez, M. D.; Borek, C.; Forrest, S. R.; Thompson, M. E. J. Am. Chem. Soc. 2009, 131,
(26), 9281–9286.
25. Lee, J.-Y.; Kim, S.-H.; Song, I.-S.; Moon, D.-K. J. Mater. Chem. 2011, 21, 16480-16487.
26. Perez, M. D.; Borek, C.; Djurovich, P. I.; Mayo, E. I.; Lunt, R. R.; Forrest, S. R.;
Thompson, M. E. Advanced Materials 2009, 21, (14 ‐15), 1517-1520.
27. Lunt, R. R.; Giebink, N. C.; Belak, A. A.; Benziger, J. B.; Forrest, S. R. Journal of
Applied Physics 2009, 105, (5), 053711.
28. Pettersson, L. A. A.; Roman, L. S.; Inganäs, O. Journal of Applied Physics 1999, 86, (1),
487-496.
29. Fan, Y., Max, Shtein and Stephen R. Forrest. Nature Materials 2004, 4, (1), 37-41.
24

30. Yang, F., Sun, K. and Forrest, S. R. Advanced Materials 2007, 19, (23), 4166-4171.
31. M. Rusu, J. G., S. Wiesner, N. Meyer, M. Heuken, K. Fostiropoulos, M.Ch. Lux-Steiner.
Thin Solid Films 2008, 516, (20), 7160–7166.
32. Rusu, M.; S. Wiesner, T. M., H. Blei, N. Meyer, M. Heuken, M.Ch. Lux-Steiner, K.
Fostiropoulos. Renewable Energy 2008, 33, (2), 254–258.
33. Reese, C.; Roberts, M.; Ling, M. M.; Bao, Z. N. Mater Today 2004, 7, (9), 20-27.
34. Kelley, T. W.; Muyres, D. V.; Baude, P. F.; Smith, T. P.; Jones, T. D. Mater Res Soc
Symp P 2003, 771, 169-179.
35. Sundar, V. C.; Zaumseil, J.; Podzorov, V.; Menard, E.; Willett, R. L.; Someya, T.;
Gershenson, M. E.; Rogers, J. A. Science 2004, 303, (5664), 1644-1646.
36. Brown, A. R.; Jarrett, C. P.; deLeeuw, D. M.; Matters, M. Synthetic Met 1997, 88, (1),
37-55.
37. Dimitrakopoulos, C. D.; Brown, A. R.; Pomp, A. Journal of Applied Physics 1996, 80,
(4), 2501-2508.
38. Gundlach, D. J.; Lin, Y. Y.; Jackson, T. N.; Nelson, S. F.; Schlom, D. G. Ieee Electr
Device L 1997, 18, (3), 87-89.
39. Shtein, M.; Mapel, J.; Benziger, J. B.; Forrest, S. R. Appl. Phys. Lett. 2002, 81, (2), 268-
270.
25

40. Rolin, C.; Vasseur, K.; Niesen, B.; Willegems, M.; Robert, M.; Soren, S.; Jan, G.; Paul,
H. Advanced Functional Materials 2012, 22, (23), 5050-5059.
41. Rolin, C.; Steudel, S.; Myny, K.; Cheyns, D.; Verlaak, S.; Genoe, J.; Heremans, P. Appl.
Phys. Lett. 2006, 89, 203502.
42. Rolin, C.; Vasseur, K.; Steudel, S.; Vicca, P.; Genoe, J.; Heremans, P. Proc. of SPIE
2009, 7417, 741712.
43. Baldo, M.; Deutsch, M.; Burrows, P.; Gossenberger, H.; Gerstenberg, M.; Ban, V.;
Forrest, S. Advanced Materials 1998, 10, (18), 1505-1514.
44. Brast, M.; Lindla, F.; Boesing, M.; Bertram, D.; Keiper, D.; Heuken, M.; Kalisch, H.;
Vescan, A. Mater. Res. Soc. Symp. Proc. 2012, 1402.
45. Bösing, M.; Lindla, F.; Brast, M.; Bertram, D.; Keiper, D.; Heuken, M.; Kalisch, H.;
Vescan, A. Mater. Res. Soc. Symp. Proc. 2012, 1402.
46. Yang, F.; Shtein, M.; Forrest, S. R. Nature Materials 2004, 4, (1), 37-41.
47. Axmann, S.; Brast, M.; Wilck, N.; Windgassen, H.; Heuken, M.; Kalisch, H.; Vescan, A.
Mater. Res. Soc. Symp. Proc. 2012, 1390.
48. Sung Hwan Kim, H. Y. C., Seung Hoon Han, Jin Jang, Sang Mi Cho. Journal of the
Korean Physical Society 2004, 45, 760-762.
49. Rolin, C.; Vasseur, K.; Schols, S.; Jouk, M.; Duhoux, G.; Müller, R.; Genoe, J.;
Heremans, P. Appl. Phys. Lett. 2008, 93, 033305.
26

50. Vasseur, K.; Rolin, C.; Vandezande, S.; Temst, K.; Froyen, L.; Heremans, P. J. Phys.
Chem. C 2010, 114, (6), 2730–2737.
51. Francisco F. Navarro, P. I. D., and Mark E. Thompson. Organic Electronics 2014, 15,
(11), 3052–3060.

27

Chapter 2. Exciton diffusion length measurement of platinum 1,
3-di-tert-butylphenyl tetraphenylbenzoporphyrin Pt(
tbu
TPBP) for
design and optimization of organic photovoltaics

2.1 Introduction

Thin film organic photovoltaics (OPVs) have attracted considerable interest as an alternative
source of energy in the past years. The key in OPVs success is the use of low cost organic
molecular and polymeric materials to enhance efficiency. A typical lamellar OPV cell consists of
a thin film electron donor (D) and acceptor (A) multilayer. There are five steps for the
photo-conversion of OPVs, light absorption, exciton diffusion, charge separation, charge transport
and charge collection, previously introduced in chapter 1.
While significant improvement has been observed in organic thin films photovoltaics
efficiencies since its development, currently the best observed efficiencies for lamellar thin film
OPV in a laboratory are just above 10%, Figure 2.1.
1-5
However, calculated theoretical maximum
efficiencies for OPVs have been shown to range between 12%
6
to 27%.
7, 8
Notwithstanding that
significant differences exist between measured and theoretical efficiencies, the key point is that
there is further progress to be made.

28

The design of new donor and acceptor molecules has proven to be fundamental to better
understand OPV device parameters that directly affect efficiency.
1
While all of the above
parameters are important when designing and optimizing OPVs, in this chapter special interest is
placed on the Voc and LD of a novel molecule, platinum tetra 1, 3-di-tert-butylphenyl
tetrabenzoporphyrin Pt(
tbu
TPBP), Figure 2.2. Efficient electrophosphorescent OLEDs have been
accomplished using Pt(TPBP).
9-11
However, when it was used as a donor in a lamellar OPV, the
device exhibited a remarkably high Voc (0.65V).
12, 13
The high Voc challenged the well-studied
correlation dependence of Voc on the energy difference between the highest occupied molecular
orbital (HOMO) energy of the donor and the lowest occupied molecular orbital (LUMO) energy
of the acceptor, ΔEDA/2.
14
The higher Voc was attributed to weak intermolecular interactions at the

Figure 2.1.- Best recorded research cell efficiency diagram from the national renewable energy
laboratory (NREL), obtained from http://www.nrel.gov/ncpv/ on October 9, 2015
29

D/A interface which directly affects dark current (Js) in OPVs, and to a linear logarithmic
dependence on the ratio of Jsc to the pre-exponential dark current factor (Jso), given by Eq. (2.1).
12

The goal of adding tert-butyl groups to Pt(TPBP) was to further decrease intermolecular
interactions at the D/A interface, thus potentially leading to a larger Voc with Pt(
tbu
TPBP) than
those observed with Pt(TPBP).
12
Because both Pt(TPBP) and Pt(
tbu
TPBP) have strong optical
densities, measuring the LD was also a logical step for the optimization of the OPV. The optical
density is a measure of the material’s ability to absorb photons and LD is the average distance an
exciton travels between generation and recombination, Eq. (2.2), or in other words, the distance
an exciton propagates before the exciton population exponentially decays to 1/e, i.e., ~35% of the
initial excitons. Where D is diffusivity in m
2
/s and τ is the lifetime in seconds.

(2.1)
q
E
J
J
q
nkT
V
DA
SO
SC
OC
2
ln











N
N
N
N
Pt

N
N
N
N
Pt

Pt(
tbu
TPBP) Pt(TPBP) C60

Figure 2.2.- Platinum tetra 1, 3-di-tert-butylphenyl tetrabenzoporphyrin Pt(
tbu
TPBP), platinum
tetraphenyltetrabenzoporphyrin Pt(TPBP) and buckminsterfullerene (C60)
30

Longer LD enables deposition of thicker films which in response absorb more photons
generating more excitons. Hence the importance the LD parameter plays in the design and
optimization of OPVs. There are several methods that can be used to measure the exciton diffusion
length of organic molecules, however, only two techniques were used, the spectrally resolve
photoluminescence quenching method (SR-PLQ)
15
and the optical electric field distribution inside
an OPV device.
16
Both techniques are discussed in more detail in the following sections.

2.2 Results and discussion
2.2.1 Stern-Volmer relationship

The most common acceptors used in OPVs are fullerene based molecules, e.g., C60,
PC70BM, PC61BM, among others. In this study we used C60, Figure 2.2, which has proven to be a
great quencher, via electron and/or energy transfer, for a large range of small organic donor
molecules and has also previously been used with Pt(TPBP).
12, 17
While C60 is expected to work
with Pb(
tbu
TPBP), a simple experiment, the Stern-Volmer relationship, can easily confirm that
indeed it is a good acceptor (quencher) for Pb(
tbu
TPBP). The Stern-Volmer relationship measures
intermolecular deactivation processes. The equation relies on fluorescence or phosphorescence
decay when the concentration of a quencher (e.g., C60) increases and is given by Eq. (2.3), where
I and I0 are the luminescence intensity with and without a quencher, respectively, Kq is the Stern-
Volmer quenching constant and [Q] is the quencher concentration.
𝐿 𝐷 = √ 𝐷 ∙ 𝜏
(2.2)
31

Emission and excitation intensities of Pt(
tbu
TPBP) in toluene were measured for
concentrations of C60 ranging from 0.0mM to 1.79x10
-3
mM, Figure 2.3. The intensity ratios where
then plotted against the concentration of C60 and from a linear fit of the data, Figure 2.4, the Stern-
Volmer rate constant was extracted and found to be Kq(emi) = 5.9x10
9
M
-1
s
-1
for emission and
Kq(exc) = 6.1x10
9
M
-1
s
-1
for excitation. The Stern-Volmer quenching constants confirm that C60 can,
via electron and/or energy transfer, quench Pt(
tbu
TPBP).

(2.3)
] [ * 1
0
Q K
I
I
q
 

Figure 2.3.- Emission and excitation spectra of Pt(
tbu
TPBP) with different C60 concentrations in
toluene
720 760 800 840
0
10000
20000
30000
40000
N N
N
N
Pt
Quencher Conc.
[mM] for C
60
0
7.15x10
-2
3.58x10
-2
1.79x10
-2
8.49x10
-3
1.79x10
-3
In Toluene
Emi. @ 430nm
4nm Slits &
385nm Filter
Pt(
tbu
TPBP) [1.25x10
-4
mM] Spectra

Intensity (a.u.)
Wavelength (nm)
350 400 450 500
0
10000
20000
30000
40000
N N
N
N
Pt
Quencher Conc.
[mM] for C
60
0
7.15x10
-2
3.58x10
-2
1.79x10
-2
8.49x10
-3
1.79x10
-3
In Toluene
Exc. @ 760nm
4nm Slits &
385nm Filter
Pt(
tbu
TPBP) [1.25x10
-4
mM] Spectra

Intensity (a.u.)
Wavelength (nm)
32

2.2.2 Film morphology

An integral part of OPV device performance is morphology of the films, which directly
affects D/A interface contact and interaction. Morphology of Pt(
tbu
TPBP) and C60 was studied
using atomic force microscopy (AFM) and X-ray diffraction (XRD). Films of 40nm Pt(
tbu
TPBP)
and 20nm C60 were fabricated on silicon substrates with the OVPD. The AFM measurements were
acquired with an atomic force microscopy digital instruments dimension 3100. Flattened and three
dimensional AFM surface images for a 20nm C60 film and a 40nm Pt(
tbu
TPBP) can be seen on
Figure 2.5 and Figure 2.6, respectively. Both films, Pt(
tbu
TPBP) and C60, exhibited a room mean
square (RMS) roughness value of 1.71nm and 1.62nm, respectively. The AFM results are
promising because large RMS can potentially yield a poor D/A interface interaction which can
adversely affect device performance.

Figure 2.4.- Stern-Volmer quenching rate plot for Pt(
tbu
TPBP) for different C60
concentrations in toluene
Quenching Rate Constant Plot for C
60
For Emission
y = 248.18x + 1.1044
R
2
= 0.9988
For Excitation
y = 255.1x + 1.0272
R
2
= 0.9999
0
5
10
15
20
25
0 0.02 0.04 0.06 0.08
Concentration (mM)
Intensity Ratio (Io/I)
Excitation
Emission
Linear (Excitation)
Linear (Emission)
33

Films of 40nm Pt(
tbu
TPBP) and 40nm C60 were also fabricated on quartz and analyzed via
XRD with a Rigaku Ultima IV powder/thin film diffractometer (XRD), Figure 2.7. Crystallinity
of Pt(
tbu
TPBP), C60 and a blank quartz substrate were measured. The diffraction patterns confirmed
no indication of crystal growth in the films, suggestive of amorphous films. While crystalline films
have proven to exhibit longer exciton diffusion lengths when compared to amorphous films,
18

control of the molecular orientation, crystal growth and crystallinity of the films is difficult to

Figure 2.5.- Flattened and 3D surface AFM images for a 20nm C60 film on silicon

Figure 2.6.- Flattened and 3D surface AFM plot for a 40nm Pt(
tbu
TPBP) film on silicon
34

achieve. In most cases these crystals act more as traps which hinder charge carriers thus also the
efficiency of the devices.

2.2.3 Spectrally resolved photoluminescence quenching (SR-PLQ) method

To measure the exciton diffusion length of Pt(
tbu
TPBP) we make use of the spectrally
resolved photoluminescence quenching technique (SR-PLQ).
15
The method relies on a thick
(200nm to 600nm) emissive organic film with a blocking and/or quenching layer (i.e., 10nm), from
which an excitation emission ratio at an angle of 45° is extracted, Figure 2.8. Film thickness is

Figure 2.7.- X-ray diffraction patterns of bare quartz substrate and 40nm Pt(
tbu
TPBP) and 20nm
C60 films on quartz substrates

10 20 30 40 50 60 70 80
0
200
400
600
800
10 20 30 40 50 60 70 80
0
200
400
600
800
10 20 30 40 50 60 70 80
0
200
400
600
800
Angle (2 )
Pt(
tbu
TPBP)
Intensity (a.u.)
C
60
Quartz
35

determined by the wavelength range of the experiment and is larger than the largest optical path
length 1/αmin.

The SR-PLQ technique employs Fick’s second law of diffusion, Eq. (2.4), to describe the
exciton density profile.
Under steady state the exciton density or concentration at any given time is equal to zero,
, and Eq. (2.4) can be simplified to obtain Eq. (2.5).
Where D is
0
) (



t
x 

(2.4)

(2.5)
    t
x n x I x n
x
x n
D











  

 ) (
cos
exp
cos
) ( ) (
0
2
2
 





   
0
cos
exp
cos
) ( ) (
0
2
2









  


 





x I x n
x
x n
D

Figure 2.8.- Film architecture and coordinates used for the analysis of
Pt(
tbu
TPBP) using the SR-PLQ method
ITO Substrate
400nm Pt(
tbu
TPBP)
10nm BCP or C
60
Film
x=t
x=0

36

The first term describes the diffusive transport. The second term the natural decay of
excitation and the third term accounts for the exciton generation rate. Where D is the diffusivity,
LD is the exciton diffusion length, τ is the exciton lifetime, I0 is the incident photon flux, α is the
absorption coefficient, θ λ, Eq. (2.6), is the incidence angle @ x=0, which include the wavelength
dependent refraction through the blocking or quenching film and n(x) is the exciton density or
concentration profile.
Where θ0 is the angle of incidence, n2 the index of refraction for the quenching layer, and
n3 is the index of refraction of the Pt(
tbu
TPBP) film. Eq. (2.6) can be simplified to Eq. (2.7) for
mathematical purposes and has the general form .
To solve Eq. (2.5) we must find a homogeneous and a particular solution .
The general solution to the homogeneous part, Eq. (2.8), can easily be found to be Eq. (2.9).

2
D
L
D 
'
2 ' '  x
Ce y a y  
part o
y y y  
hom

(2.6)

(2.7)
Where

(2.8)
Where
 















 
2
0 1
3
2 1
sin
sin sin
n n
n 


  0 exp
) ( ) (
' '
0 2
2
   


 

x I
x n
x
x n
D




cos
'

0
2 ' '
  y a y
37

The particular solution for the non-homogeneous part is solved by suggesting a solution
based on the non-homogeneous part of Eq. (2.7). It can be observed a good suggestion would be
. Substituting the suggested solution into Eq. (2.10) leads to Eq. (2.11), at
which point we can solve for A. Once A is found, we substitute back into the suggested particular
solution, thus obtaining the general solution, Eq. (2.12), to the particular part
of the equation. Combining both general solutions then provides the final solution, Eq. (2.13)
To find constants C1 and C2 we utilize the boundary conditions (BC) of the problem. We
assume formation of excitons to be zero at the interface between the substrate and the material of
 
'
exp  x A y
part
 
 
'
exp  x A y
part
 

(2.9)

(2.10)
Where

(2.11)
Where

(2.12)

(2.13)


















D D
o
L
x
C
L
x
C y sinh cosh
2 1 hom
) exp(
' ' 2 ' '
  x y a y  
2
0
D
L
I 
 
  ) exp( ) exp( exp
' ' ' 2 '
2
'
     x x A a x A      
) ( cos
) cos(
2 2 2
0


 
 


D
L
I
A
 
'
2 2 2
0
exp
) ( cos
) cos(

 
 


x
L
I
y
D
part



 
'
2 2 2
0
2 1
exp
) ( cos
) cos(
sinh cosh ) ( 
 
 


x
L
I
L
x
C
L
x
C x n y
D D D




















 
38

study for sufficiently thick films, hence n(x=L) = 0 where L is the film thickness. The other two
boundary conditions are obtained from the blocking and quenching layers. Assuming the blocking
layer prevents interface quenching, the derivative of the exciton density with respect to x should
be zero . The last BC requires every exciton that reaches the interface to be quenched,
therefore . We are now in a position to solve for the constants for both blocking and
quenching layer. The constants for nB(x) are,
and
And therefore

The constants for nQ(x) are
and
And therefore

0
) (
0



 x
x
x n
0 ) (
0

 x
x n
 


































D
D
D D
D
L
L
L
h L
L
L
L
L I
C

 
 



sec exp ) cos(
tanh
) ( cos
'
2 2 2
2
0
1
) ( cos
2 2 2
2
0
2

 



D
D
L
L I
C
 
 



































































D D D D
D
D D
D
B
L
x
L
x
L
x
L
L
L
h L
L
L
L
L I
x n

 

 
 




'
'
2 2 2
2
0
exp ) cos(
sinh cosh
sec exp ) cos(
tanh
) ( cos
) (
) ( cos
) cos(
2 2 2
0
1


 
 


D
L
I
C
 


































D
D
D
L
L
L
L
L
L
I
C coth
sinh
exp
) ( cos
) cos(
'
2 2 2
0
2

 
 


 
 














 



















































'
'
2 2 2
0
exp sinh coth
sinh
exp
cosh
) ( cos
) cos(
) ( 

 
 


x
L
x
L
L
L
L
L
L
x
L
I
x n
D D
D
D D
Q
39

We can also calculate the fraction (S) of photons entering the film that result in the
formation of excitons that reach L for both the blocking and quenching layer, given by

Thus, we can also find the fraction of photons for quenching and blocking, Eq. (2.14) and
Eq. (2.15) respectively, identical to ones derived by Simpson,

Comparing the photoluminescence of the blocking and quenching films, a normalized
quenching ratio η ( x ) can be obtain by integrating the exciton density profile, Eq. (2.16), over the
entire film thickness.
0
) (
I
x
L n
D
S


 
 
 
 
0
'
'
2 2 2
0
) cos(
exp
cosh coth coth exp sinh
) ( cos
) cos(
I
L
L
L
L
L
L
L
L
L
L
L L
I D
S
D D D D D D
Q











































 









 
 


 

 
 
 
 
0
'
'
2 2 2
2
0
exp cosh sinh
sec exp ) cos(
tanh
) ( cos
I
L
L
L
L
L
L
L
L
h L
L
L
L
DI
S
D D D
D
D D
B














 


















































 


 
 




(2.14)

(2.15)
 
 














 








 










'
'
2 2 2
2 2
exp
coth exp csc
) ( cos
) cos(



 
 


L
L
L
L
L
L
L
h
L
L
S
D
D D
D
D
Q
 




































  










D
D
D D
D
B
L
L
L
L
L
L
h
L
L
S



 



tanh ) cos(
1 exp sec
) ( cos
'
2 2 2
2 2
40

Considering the fact that L>>>LD, we are able to simplify nB(x) and nQ(x) from the above
equations to obtain Eq. (2.17) and Eq. (2.18) respectively, by using the following constants:
Replacing Eq. (2.17) and Eq. (2.18) into Eq. (2.16) and solving one can found that the
normalized quenching ratio, η ( x ), is described by Eq. (2.19), where the slope of the quenching
ratio vs. α’ yields the exciton diffusion length, LD.

 
'
2 2 2
0
2 1
exp
) ( cos
) cos(
exp exp ) ( 
 
 


x
L
I
L
x
C
L
x
C x n y
D D D











 








 
 
 


















































 
0
'
2 2 2
0
2 2 2
0
0
'
2 2 2
0
2 2 2
2
0
0
0
exp
) ( cos
) cos(
exp
) ( cos
) cos(
exp
) ( cos
) cos(
exp
) ( cos
) , (
) , (
) (
dx x
L
I
L
x
L
I
dx x
L
I
L
x
L
L I
dx x n
dx x n
x
D
D
D
D
D
D
D
Q
B

 
 
 
 

 
 
 












(2.16)
and

(2.17)
and

(2.18)




 
0
0
) , (
) , (
) (
) (
) (
dx x n
dx x n
PL
PL
x
Q
B
Q
B





0
1
 C
) ( cos
2 2 2
2
0
2

 



D
D
L
L I
C
 
'
2 2 2
0
2 2 2
2
0
exp
) ( cos
) cos(
exp
) ( cos
) ( 
 
 
 




x
L
I
L
x
L
L I
x n
D
D
D
D
B














0
1
 C
) ( cos
) cos(
2 2 2
0
2


 
 


D
L
I
C
 
'
2 2 2
0
2 2 2
0
exp
) ( cos
) cos(
exp
) ( cos
) cos(
) ( 
 
 
 
 




x
L
I
L
x
L
I
x n
D
D
D
Q














41

Equation (2.19) describes interaction where long distance energy transfer from the C60 film
to Pt(
tbu
TPBP) is absent. If there is an overlap in the absorption spectrum of the quenching film
with the emission spectrum of the Pt(
tbu
TPBP) film, non-diffusive Föster energy transfer,
dipole-dipole induced to the quenching film must be considered. Equation (2.20) illustrates the
modified equation accounting for long distance energy transfer.
The last term integrates the long distance energy transfer over the entire quenching
interface of area A. Where RQ is the Föster radius, R(x) is the distance from any molecule in the
quenching layer to any point in the Pt(
tbu
TPBP) film and ρA is the number of quenching molecules
per unit area. In order to account for Föster transfer we have to solve Eq. (2.20) and Eq. (2.7)
with the boundary conditions which are then fitted to the quenching ratio vs. α’ to extract LD. The
solution can be seen in Eq. (2.21).
 
 






























0
'
2 2 2
0
2
2 2 2
0
'
2 2 2
2
0
2 2 2
2 2
0
exp
) ( cos
) ( cos
exp
) ( cos
) cos(
exp
) ( cos
) ( cos
exp
) ( cos
) (

 
 
 
 

 
 
 









x
L
I
L
x
L
L I
x
L
I
L
x
L
L I
x
D D D
D
D D D
D
) cos(
) cos(
) ( cos
) ( cos
0
) ( cos
) cos(
0
) ( cos
) ( cos
0
) ( cos
0
) (
0
0
2 2 2
0
2
2 2 2
0
2 2 2
2
0
2 2 2
2 2
0









 
 

 
 
 
 
 
 
 





 



 



D
D D
D
D D
D
L
I
I
L
I
L
L I
L
I
L
L I
x

(2.19)

(2.20)
1
) cos(
) cos(
) cos(
) cos(
) (
) (
) (
'
0
0
 



 
D
D
D
Q
B
L
L
L
I
I
PL
PL
x 

 
 
 








   
0
) (
) (
cos
exp
cos
) ( ) (
6
6
0
2
2


























  



dA
x R
R
x n
x I x n
x
x n
D
Q
A






 
42

Derivation of the equations used in the SR-PLQ technique is useful in order to better
understand the method. Before we use the SR-PLQ technique, it’s important to know if Föster
energy transfer will occur between the Pt(
tbu
TPBP) and C60. As mentioned before, if energy
transfer occurs and we don’t account for it in the equation, it will affect the calculated LD. The
emission spectra of Pt(
tbu
TPBP) and absorption spectra of BCP and C 60 films deposited with the
OVPD can be seen in Figure 2.9. Emission spectra were taken using a photon technology
international QuantaMaster model C-60 fluorimeter, absorption spectra were taken using an
Agilent ultraviolet-visible spectrometer. It can be observed that the absorption overlap of C60 with
the emission from Pt(
tbu
TPBP) is small, hence negligible, allowing us to use Eq. (2.19) to measure
the LD of Pt(
tbu
TPBP).

(2.21)
   
0
5
2
) (
cos
exp
cos
) ( ) (
4
6
0
2
2


















  


x
R
x n
x I x n
x
x n
D
Q
A







 

Figure 2.9.- Emission and absorption spectra of Pt(
tbu
TPBP), C60 and
BCP

200 300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Absorbance (a.u.)
Bare Film Spectra
Intensity (a.u.)
Wavelength (nm)
10nm C
60
10nm BCP
20nm Pt(
tbu
TPBP)
Exc. @ 760nm
385nm Filter
4nm Slits
43

Now that we know that C60 is a good quencher for Pt(
tbu
TPBP) and which equation to use,
we can fabricate and analyze the films. Excitation emission measurements were performed at 45°
and to avoid wave guided corrections inside the quartz substrates, a correction is performed on the
actual absorbance and transmittance of the films. Correcting for the angle is done by using the
Pythagorean Theorem, were the hypotenuse is the thickness of the film at 45°. Excitation
measurements were performed at a fixed 760nm detection wavelength and a 385nm long pass filter
to avoid double harmonic emission. Raw and corrected absorption spectra of two 400nm
Pt(
tbu
TPBP) films, each with a 10nm blocking (BCP) and/or quenching layer (C60), can be seen in
Figure 2.10. Bare films of the blocking and quenching layer were also grown, Figure 2.11. These
bare films were grown in order to measure the absorbance and calculate transmittance losses owing
to the blocking and quenching layer. The absorbance and transmittance spectra were also corrected
to 45° and the data was used to account for any losses in the emission spectra.

Figure 2.10.- Raw and corrected absorption spectra of a 400nm Pt(
tbu
TPBP) film on quartz
with a blocking and quenching layer

200 300 400 500 600 700
0.0
0.5
1.0
1.5
2.0
2.5
PtCorrBCP
PtCorrC60
400nm Pt(
tbu
TPBP) Film Spectra

Absorbance (a.u.)
Wavelength (nm)
200 300 400 500 600 700
0.0
0.5
1.0
1.5
2.0
2.5
3.0
With 10nm BCP
With 10nm C
60
400nm Pt(
tbu
TPBP) Film Spectra

Absorbance (a.u.)
Wavelength (nm)
44

Initially, I was interested in knowing if the measurement could be done without a blocking
layer, therefore, the first films were made without a blocking layer. The emission measurements
were performed under N2, e.g., nitrogen gas blowing directly onto the films inside the
spectrophotometer. Unfortunately, quenching was observed on the emission spectra of Pt(
tbu
TPBP)
without the use of a blocking layer, Figure 2.12. Therefore, it was necessary to use a blocking layer
to stop the quenching mechanism at the interface. It’s worth noting that the emission measurements
were also performed under vacuum (0.08torr) with the use of a cylindrical Pyrex container. The
measurement was also performed under vacuum because a significant fluctuation of the LD (severe
quenching, specifically with C60) was observed when the measurement was completed using N2,
Figure 2.13. The result was surprising given the films were exposed to ambient air for
approximately one minute before a constant flow of N2 impinge the films in preparation for
emission measurement. The exciton diffusion length, for both N2 and vacuum, was extracted from
the spectra, Figure 2.14, resulting in an LD of 72nm ± 3.6nm and 13.4nm ± 0.7nm, respectively.
Clearly, blowing N2 gas to the films did not prevent O2 from quenching the excitation emission.

Figure 2.11.- Raw and corrected absorption and transmittance spectra of a 10nm BCP and
10nm C60 film on quartz substrates

200 300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
Transmittance (a.u.)
10nm BCP
10nm C60
Raw Data
Bare Film Spectra
Absorbance (a.u.)
Wavelength (nm)
200 300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
Transmittance (a.u.)
10nm BCP
10nm C60
Corrected to 45Deg
Bare Film Spectra
Absorbance (a.u.)
Wavelength (nm)
45

Considering the size of the molecules, C60 and BCP, it can be inferred that a thin film of such a
bulky molecule like C60 allows easier O2 diffusion as opposed to a denser film of a much smaller
molecule like BCP. Quenching of the films suggests O2 was physisorbed to Pt(
tbu
TPBP) and even
though N2 was blown directly onto the film when performing the emission measurements, it was
not sufficient to remove the physisorbed O2 from the film, thus requiring vacuum for its removal.

Figure 2.12.- Excitation emission spectra of two Pt(
tbu
TPBP) films,
with and without a quencher layer, e.g., 10nm of C60, on quartz
substrates

200 300 400 500 600 700
1000
2000
3000
4000
5000
6000
With 10nm C
60
Bare
Exc. @ 760nm
Raw Data
385nm Filter
4nm Slits
400nm Pt(
tbu
TPBP) Film Spectra

Intensity (a.u.)
Wavelength (nm)

Figure 2.13.- Excitation emission spectra of Pt(
tbu
TPBP) with 10nm of BCP and 10nm of C60
under N2 and 0.08torr vacuum
400 500 600 700
0
4000
8000
12000
16000
20000
24000
With 10nm C
60
With 10nm BCP
Exc. @ 760nm
Corrected Data
385nm Filter
4nm Slits
Under N
2
450nm Pt(
tbu
TPBP) Film Spectra

Intensity (a.u.)
Wavelength (nm)
400 500 600 700
0
4000
8000
12000
16000
20000
24000
28000
32000
With 10nm C
60
With 10nm BCP
Exc. @ 760nm
385nm Filter
4nm Slits
Under Vacuum
Corrected
450nm Pt(
tbu
TPBP) Film Spectra

Intensity (a.u.)
Wavelength (nm)
46

Vacuum measurements were completed with a cylindrical two piece flat flange O-ring seal
Pyrex container with a high vacuum inlet valve chem cap. A circular stainless steel substrate holder
with a diameter similar to the inside diameter of the Pyrex container was used to keep the substrate
standing vertically inside the container. Raw and wavelength offset corrected emission spectra of
Pt(
tbu
TPBP) under vacuum with a blocking and/or a quenching layer are shown on Figure 2.15
while bare films emission from BCP and C60 can be observed on Figure 2.16. A plot of the
photoluminescence ratio of the blocking layer with respect to the quenching layer versus
wavelength, Figure 2.17, helps corroborate quenching ratio and alpha prime relationship, a direct
measurement of the absorption coefficient of the organic of study with respect to the quenching
rate and a good correlation to a fine measurement. It can be observed a wavelength offset which
arises from a wavelength discrepancy between the ultraviolet-visible spectrometer emission and
the absorption spectra. Absorption measurements ranges are from 190nm to 1100nm, optical
constants from 345nm to 900nm and emission spectra from 190nm to 730nm. However, it can be

Figure 2.14.- Corrected wavelength offset photoluminescence quenching ratio versus alpha
prime under N2 and 0.08torr vacuum
0.00 0.01 0.02 0.03 0.04
0
5
10
15
20
25
30
Value Error
---------------------------------------------
A 3.10017 0.35139
B 721.57512 36.9851
---------------------------------------------
R SD
--------------------------------------------
0.9242 2.1098
--------------------------------------------
CorrQuenRatioAdj
Under N
2

Quenching Ratio (PL
B
/PL
Q
)
450nm Pt(
tbu
TPBP) Film Spectra
Alpha Prime (A
-1
)
0.00 0.01 0.02 0.03 0.04
0
2
4
6
Value Error
--------------------------------------------------
A 1.15084 0.07341
B 134.23257 7.38246
--------------------------------------------------
R SD
--------------------------------------------------
0.92355 0.41794
--------------------------------------------------
QRCorrVacuumAdj
Under Vacuum

Quenching Ratio (PL
B
/PL
Q
)
450nm Pt(
tbu
TPBP) Film Spectra
Alpha Prime (A
-1
)
47

observed that the best wavelength range, based on the emission data, is from 345nm to 730nm,
which is the range used in all the plots.

Figure 2.15.- Raw and corrected emission spectra of two 400nm Pt(
tbu
TPBP) films with 10nm
of BCP and 10nm of C60

400 500 600 700
0
2000
4000
6000
8000
10000
12000
With 10nm C
60
With 10nm BCP
Exc. @ 760nm
Raw Data
385nm Filter
4nm Slits
400nm Pt(
tbu
TPBP) Film Spectra

Intensity (a.u.)
Wavelength (nm)
400 500 600 700
0
2000
4000
6000
8000
10000
12000
With 10nm C
60
With 10nm BCP
Exc. @ 760nm
Corrected Data
385nm Filter
4nm Slits
400nm Pt(
tbu
TPBP) Film Spectra

Intensity (a.u.)
Wavelength (nm)

Figure 2.16.- Emission spectra of bare films of BCP and C60

400 500 600 700
200
400
600
800
1000
10nm C
60
10nm BCP
Exc. @ 760nm
Raw Data
385nm Filter
4nm Slits
Film Spectra

Intensity (AU)
Wavelength (nm)
48

Pt(
tbu
TPBP) has two distinct and strong bands, the Soret and Q band, which allows us to
observe equal quenching ratios at four different points in the data. Parallel quenching ratio values
can be clearly observed in the data obtained, Figure 2.15, confirming the measurements were

Figure 2.17.- Raw and wavelength offset corrected photoluminescence quenching ratio vs.
alpha prime
400 500 600 700
1
2
3
4
5
6
0.01
0.02
0.03
0.04
Alpha Prime(A
-1
)
Quenching Ratio (PL
B
/PL
Q
)
400nm Pt(
tbu
TPBP) Film Spectra
Wavelength (nm)
400 500 600 700
1
2
3
4
5
6
0.00
0.01
0.02
0.03
0.04
CorrQuenRatioAdj
& Spectrum Offset Adjustment
Alpha Prime (A
-1
)
Quenching Ratio (PL
B
/PL
Q
)
400nm Pt(
tbu
TPBP) Film Spectra
Wavelength (nm)

Figure 2.18.- Raw and wavelength offset corrected photoluminescence quenching ratio versus
alpha prime
0.00 0.01 0.02 0.03 0.04
1
2
3
4
5
Value Error
--------------------------------------------
A 0.99048 0.07714
B 97.31016 7.76411
--------------------------------------------
R SD
--------------------------------------------
0.82807 0.4959
--------------------------------------------
Quenching ratio
Quenching Ratio (PL
B
/PL
Q
)
400nm Pt(
tbu
TPBP) Film Spectra
Alpha Prime (Å
-1
)
0.00 0.01 0.02 0.03 0.04
0
1
2
3
4
5
6
7
Value Error
--------------------------------------------
A 0.92439 0.05304
B 108.41333 5.27546
--------------------------------------------
R SD
--------------------------------------------
0.92618 0.33626
--------------------------------------------

Corrected Quenching Ratio
& Offset Spectrum Adjustment
Quenching Ratio (PL
B
/PL
Q
)
400nm Pt(
tbu
TPBP) Film Spectra
Alpha Prime (Å
-1
)
49

completed correctly. The final step is to apply a linear fit, using Eq. (2.17), from which we are able
to extract the LD for Pt(
tbu
TPBP), Figure 2.18. The LD of Pt(
tbu
TPBP) was found to be 97.3Å ± 7.7Å
(9.7nm ± 0.8nm) and 108.4Å ± 5.2Å (10.8nm ± 0.5nm) for the raw and wavelength offset
corrected data respectively. Several LD measurements were performed and the average LD for
Pt(
tbu
TPBP) was found to be 12.1nm ± 2.4nm.

2.2.4 Organic photovoltaics fabricated with the OVPD

As discussed earlier, knowing the LD is essential to optimize OPVs. In the case of
Pt(
tbu
TPBP), it was found to be approximately 10nm ± 3nm. With this information we proceeded
to fabricate solar cells using the architecture depicted in Figure 2.19. Further, the HOMO-LUMO
energy diagram for the device can be seen in Figure 2.20, where the energetics align favorably for
charge carrier dissociation. Several sets of OPV devices were fabricated and evaluated.
Specifically, four sets of devices with 10, 20, 45, and 60nm of Pt(
tbu
TPBP)* were made in the
OVPD. The device architecture used was ITO/Pt(
tbu
TPBP)
*nm
/C60
10nm
/BCP
10nm
/Al
100nm
for the
devices fabricated in the OVPD. Reference devices were also made in the VTE, however, as
mentioned before, the VTE uses more material than the OVPD, therefore only one set of 7, 13 and
20nm of porphyrin were fabricated in order to optimize the material. The device structure used in
the VTE was similar to the OVPD except a thicker film of C60 was used (40nm). It’s common to
observe high sublimation temperatures in fullerenes and C60 is not the exception. The necessary
temperature for C60 to achieve sublimation is approximately 540°C. During deposition of C60 in
the OVPD, the chiller was not able to maintain the substrate holder temperature for a long period
50

of time, thus to prevent annealing the porphyrin film, a 10nm C60 film was used instead of the
frequently 40nm thick film used in OPVs.

Current-voltage curves and external quantum efficiencies (EQE) of the OVPD devices with
10nm of Pt(
tbu
TPBP) are shown in Figure 2.21. Performance parameters extracted from the J-V

Figure 2.19.- Organic photovoltaic device
architecture
Substrate
ITO
Pt(
tbu
TPBP)
40nm C
60
10nm BCP
100nm Al
N N
N
N
N
N
Pt

Figure 2.20.- Organic photovoltaic HOMO-LUMO energy diagram
-7
-6
-5
-4
-3
-2
-7
-6
-5
-4
-3
-2
E
DA
~1.4eV
Energy (eV)
Al
ITO C
60
BCP
Pt(
tbu
TPBP)
Energy (eV)
OPV HOMO-LUMO Energy Diagram

51

curves are shown in Table 2.1. For these devices the average efficiency and Voc was found to be
0.56% and 0.49V, respectively. While we were expecting a higher Voc (~0.65V), we think having
a thin C60 film (10nm) and potentially film intermixing (during deposition of C60) influenced the
open circuit voltage. When film intermixing ensues, strong intermolecular interactions at the D/A
interface are likely to take place, thus increasing dark current and consequently decreasing the Voc.
Further, a week C60 and a strong porphyrin current contribution can be observed on the EQE, which
can also contribute to a smaller Voc. However, 0.5V still exceeds the expected Voc if we use the
correlation. This correlation states the Voc of OPVs can be estimated by the following
relationship, ΔEDA/2.

DA
E 

Figure 2.21.- Current-Voltage curves and EQE spectra for 10nm Pt(
tbu
TPBP) devices made in
the OVPD
-0.8 -0.4 0.0 0.4 0.8
-4
-2
0
2
4
ITO/Pt(
tbu
TPBP)
10nm
/C
60
10nm
/BCP
10nm
/Al
S1D1
S2D1
S3D1
S4D1

Current Density (mA/cm
2
)
Voltage (V)
400 500 600 700 800
0
10
20
30
40
50
ITO/Pt(
tbu
TPBP)
10nm
/C
60
10nm
/BCP
10nm
/Al
S1D1
S2D1
S3D1
S4D1

Quantum Efficiency (%)
Wavelength (nm)
52

In the case of a CuPc/C 60 device architecture, a ΔEDA of 1.7V is obtained while a ΔEDA of
1.4V is obtained for a Pt(
tbu
TPBP)/C60 device structure, Figure 2.22. Based on this correlation, the
CuPc/C60 device should have a larger Voc (~0.6V) while the Pt(
tbu
TPBP) device should have a
smaller Voc (~ 0.4V). However, the CuPc/C60 device has a smaller Voc (~ 0.48V) than the
Pt(
tbu
TPBP) device.
12
As mentioned previously, it was found that weak intermolecular interactions
at the Pt(TPBP)/C60 interface leads to small dark currents and thus a greater Voc.
12
A similar event
is observed for Pt(
tbu
TPBP) devices, weak intermolecular interactions are achieved thanks to a
saddle shape and a highly distorted molecule. In contrast, CuPc is a flat molecule which creates
strong intermolecular interactions at the D/A interface, hence increasing dark current and reducing
the Voc.

OPV devices with 10nm Pt(
tbu
TPBP)
Pt(
tbu
TPBP) Jsc (mA/cm
2
) Voc (V) FF η (%) Js n Rsa
S1D1 2.85 0.506 0.48 0.693 0.0001 2.18605 0.00267
S2D1 2.23 0.469 0.46 0.485 0.00012 2.08771 0.00227
S3D1 3.05 0.485 0.4 0.599 0.00009 2.38676 0.0025
S4D1 2.2 0.492 0.42 0.454 0.00011 2.2579 0.00149
Average 2.58 0.49 0.44 0.56 0.0001 2.23 0.002
Table 2.1.- Photovoltaic performance parameters for 10nm Pt(
tbu
TPBP) devices made in the
OVPD
53

Similar performances were observed for the 20nm Pt(
tbu
TPBP) devices, Figure 2.23 and
Table 2.2, where the average efficiency and Voc was found to be 0.61% and 0.52V, respectively.
The external quantum efficiency of these devices behaved similarly to the 10nm OPVs, where
poor current contribution from C60 was observed. However, the Jsc of the 10nm OPVs exhibited
higher current than the 20nm devices. The same behavior was observed with the Pt(TPBP) devices
so it’s not a surprise Pt(
tbu
TPBP) behaved equally. The dissimilarity in current is attributed to a
short LD which reduces the probability of excitons reaching the D/A interface before they decay.
13

Figure 2.22.- HOMO LUMO energy diagram of CuPc vs Pt(
tbu
TPBP)
-7
-6
-5
-4
-3
-2
-7
-6
-5
-4
-3
-2
~300eV
CuPC
E
DA
~1.4eV
Energy (eV)
C
60
Pt(
tbu
TPBP)
Energy (eV)
HOMO LUMO Energy Diagram

54

Figure 2.23.- Current-Voltage curves and EQE spectra for 20nm Pt(
tbu
TPBP) devices made in
the OVPD
-0.8 -0.4 0.0 0.4 0.8
-4
-2
0
2
4
ITO/Pt(
tbu
TPBP)
20nm
/C
60
10nm
/BCP
10nm
/Al
S1D1
S2D1
S3D1
S4D1

Current Density (mA/cm
2
)
Voltage (V)
400 500 600 700 800
0
5
10
15
20
25
30
35
ITO/Pt(
tbu
TPBP)
20nm
/C
60
10nm
/BCP
10nm
/Al
S1D1
S2D1
S3D1
S4D1

Quantum Efficiency (%)
Wavelength (nm)
OPV devices with 20nm Pt(
tbu
TPBP)
Pt(
tbu
TPBP) Jsc (mA/cm
2
) Voc (V) FF η (%) Js n Rsa
S1D1 2.28 0.529 0.57 0.69 0.00005 2.13248 0.00053
S2D1 2.04 0.5 0.56 0.567 0.00011 2.17009 0.00049
S3D1 2.3 0.535 0.53 0.656 0.0001 2.38881 0.00243
S4D1 2.12 0.505 0.51 0.546 0.0001 2.19473 0.00196
Average 2.18 0.52 0.54 0.61 0.00009 2.22 0.001
Table 2.2.- Photovoltaic performance parameters for 20nm Pt(
tbu
TPBP) devices made in the
OVPD
55

Conversely, poor performance was observed on the 45nm Pt(
tbu
TPBP) devices, Figure 2.24
and Table 2.3, where an average efficiency of 0.27% and a 0.47V Voc were obtained. This result
was expected since we knew the LD of Pt(
tbu
TPBP) is 10nm ± 3nm, thus only a smaller fraction of
excitons can actually reach the D/A interface and dissociate with such a thick film. Yet again, poor
C60 and strong Pt(
tbu
TPBP) current contribution can be observed from the EQE spectra. Finally
and unsurprisingly, the set of devices with 60nm Pt(
tbu
TPBP) did not rectify. Similar behavior was
observed with Pt(TPBP) devices when a 40nm layer was used.
13

Figure 2.24.- Current-Voltage curves and EQE spectra for 45nm Pt(
tbu
TPBP) devices made
in the OVPD
-1.0 -0.5 0.0 0.5 1.0
-2
-1
0
1
2
ITO/Pt(
tbu
TPBP)
45nm
/C
60
10nm
/BCP
10nm
/Al
S3D1
S3D2

Voltage (V)
Current Density (mA/cm
2
)
400 500 600 700 800
0
4
8
12
16
20
ITO/Pt(
tbu
TPBP)
45nm
/C
60
10nm
/BCP
10nm
/Al
S3D1
S3D2
Wavelength (nm)
Quantum Efficiency (%)

OPV devices with 45nm Pt(
tbu
TPBP)
Pt(
tbu
TPBP) Jsc (mA/cm
2
) Voc (V) FF η (%) Js n Rsa
S1D1 1.17 0.494 0.53 0.309 0.0001 2.20557 0.00111
S2D1 1.18 0.455 0.43 0.228 0.00009 2.38097 0.00049
Average 1.18 0.47 0.48 0.27 0.000095 2.3 0.0008
Table 2.3.- Photovoltaic performance parameters for 45nm Pt(
tbu
TPBP) devices made in the
OVPD
56

2.2.5 Organic photovoltaics fabricated with the VTE

The set of reference devices fabricated in the VTE chamber to aid in the evaluation of the
devices made in the OVPD can be seen in Figure 2.25 and Table 2.4. All three devices made in
the VTE performed similarly, where an average efficiency of 1.1% and an open circuit voltage
greater than 0.6V was obtained. The Voc of Pt(
tbu
TPBP) is in agreement with its analogue molecule
Pt(TPBP), which exhibited a Voc of 0.69V.
12, 13
Furthermore, it’s clear that just like its analogue
molecule, Pt(
tbu
TPBP) also exhibits weak intermolecular interactions at the D/A interface and a
logarithmic dependence to JSC/JSO.

There is a decrease in Jsc with thickness and is analogous to what was observed with the
Pt(TPBP) devices.
13
Contrariwise, the VTE devices exhibited a higher Jsc as opposed to the OVPD
devices. This current improvement is undoubtedly due to the thicker C60 film used in the VTE
devices (40nm) and is clearly seen in the EQE spectra, where a stronger absorption band between

Figure 2.25.- Current-Voltage curves and EQE spectra for 7, 13 and 20nm Pt(
tbu
TPBP) devices
made in the VTE
-0.8 -0.4 0.0 0.4 0.8
-4
-2
0
2
4
ITO/Pt(
tbu
TPBP)/C
60
40nm
/BCP
10nm
/Al
S1D1 7nm
S2D1 13nm
S3D1 20nm

Current Density (mA/cm
2
)
Voltage (V)
400 500 600 700 800
0
5
10
15
20
25
30
35
ITO/Pt(
tbu
TPBP)/C
60
40nm
/BCP
10nm
/Al
S1D1 7nm
S2D1 13nm
S3D1 20nm

Quantum Efficiency (%)
Wavelength (nm)
57

350 and 450nm, not seen in devices with 10 – 40nm C60 OVPD devices, is now present in the EQE
spectra. Device efficiencies between both deposition methods are evidently not alike, however,
analogous performance parameters can be obtained in the OVPD if a 40nm C60 film is deposited.

2.2.6 Substrate holder heat conduction study

A temperature study can help us elucidate the impact deposition of C60 has on the substrate
and thus the organic film of Pt(
tbu
TPBP). For such study, it’s necessary to use Fourier’s Law of
heat conduction and Newton’s Law of cooling.
19
We first define the system of interest shown on
Figure 2.26, in which we have stainless steel and quartz in contact with each other, while there is
fluid contact at the ends of both materials. Stainless steel is in contact with flowing water at a
temperature of 25
o
C and the quartz substrate is exposed to flowing nitrogen at a given pressure
(torr) and temperature (°C), the latter dependent on the organic material to be deposit. To solve
our problem we make use of several equations, such as Sutherland’s formula, for which material
constants e.g., density, viscosity, thermal conductivity, among others will be needed, Table 2.5.
VTE devices with 7, 13 and 20nm Pt(
tbu
TPBP)
Pt(
tbu
TPBP) Jsc (mA/cm
2
) Voc (V) FF η (%) Js n Rsa
7nm 3.49 0.607 0.52 1.1 0.00002 2.09205 0.01666
13nm 3.35 0.649 0.59 1.28 0.00019 2.96768 0.00203
20nm 3.11 0.678 0.55 1.17 0.0002 3.32084 0.00311
Table 2.4.- Photovoltaic performance parameters for 7, 13 and 20nm Pt(
tbu
TPBP) devices
made in the VTE
58

Figure 2.26.- Heat conduction through a stainless steel substrate holder and a quartz substrate
diagram
Quartz Substrate
K
2
=1.4 W/m K
Stainless Steel
K
1
=16 W/m K
Water @ Ta=25
o
C
Flow=15L/min
P=760torr
Nitrogen @ Tb=520
o
C
Flow=40sccm
P=1torr
T
0
T
1
T
2
X
0
X
1
X
2
W
H
WATER DATA NITROGEN DATA
Ta = 25
o
C = 298.15 K Tb = 520
o
C = 793.15 K
Fa = 15L/min Fb = 40 sccm
Pa = 760 torr Pb = 1 torr
Ka = 0.58 W/ m K
22
Kb = 0.043 W/ m K
μa = 8.91 x 10
-4
Kg/m s
Ŧ
μb = 3.536 x 10
-5
Kg/m s
ρa = 997.13 Kg/m
3

§
ρb = 5.66 x 10
-4
Kg/m
3

Cpa = 4181.3 J/Kg K Cpb = 1122 J/Kg K
Tube Diameter = 0.4cm = 0.004m Tube Diameter = 10cm = 0.1m
Table 2.5.- Water and nitrogen thermal conductivity, density, viscosity,
heat capacity, among others
59

Sutherland’s formula,
Ŧ
Eq. (2.22), expresses the relationship between dynamic viscosity of
gases and absolute temperature
Where the subscripts a and b are defined by the following relationship, a = 0.555(T0) + C and
b = 0.555(T) + C. The values for the reference temperature, viscosity and constant for N2, are
T0 = 540.99R, μ0 = 0.01781 centipoise and C = 111, respectively. Sutherland’s formula is
applicable for temperatures between 273.15K ≤ T ≤ 800K providing fairly accurate values of
viscosity. The units of the viscosity obtained are in centipoises.
To calculate the thermal conductivity of N2 we use Eq. (2.23)
Where the value of the constants are -92.39, 1.647 and 5.255x10
-4
, for C1, C2 and C3 respectively,
and T is the temperature of the gas. The above thermal conductivity formula is valid for
temperatures ranging between 300K ≤ T ≤ 2200K.
20

The ideal gas law,
§
PV = nRT, can be manipulated to calculate the density of a gas at a
given temperature and pressure, by multiplying it with the molecular weight (MW) to obtain Eq.
(2.24), and since mass and density are given by m = n(MW)and ρ = m/V, respectively. Rearranging
the ideal gas law we can obtain a formula to calculate densities at any given pressure and
temperature using Eq. (2.25).
19

(2.22)

(2.23)

(2.24)
2
3
0
0















T
T
b
a
 
  





   T C C
T
C
T K *
3 2
1
) ( ) ( MW nRT MW PV 
60

Where R is the ideal gas constant, R = 0.08206 L · Atm/mol·K is the gas constant.
As we can observe in Figure 2.26, heat transfer @ X0 and X2 are given by Newton’s law of
cooling, providing a heat transfer coefficients h0 and h2 for each boundary as follows,
We are now able to write an energy balance equation for a slab of width (W) and height
(H). For the stainless steel slab we then have
We can now divide by WH and Δx and take the limit as Δx → 0 to get
Integrating the above equation, it can be observed that qx = q0, where q0 is the heat flux @
x = x0. The same result can be obtained by integrating for the quartz slab, indicating constant heat
flux in both slabs. Now Fourier’s law for heat transfer is introduce to each slab to get
And

(2.25)

(2.26)
And

(2.27)
(2.28)

(2.29)

(2.30)
RT
MW P ) (
 
0
0
0
h
q
T T
a
 
2
0
2
h
q
T T
b
 
0  
 
WH q WH q
x x x x x
0 
dx
dq
x






 
dx
dT
K q
1 0
61

If thermal conductivity is constant, true for our OVPD system, integrating Eq. (2.30) and
Eq. (2.31) will result in Eq. (2.32) and Eq. (2.33).
And
Adding Fourier’s and Newton’s equations, Eq. (2.26), Eq. (2.27), Eq. (2.32), and
Eq. (2.33), an overall heat flux can be deduced, Eq. (2.34).
The heat transfer coefficient (U), Eq. (2.37) can also be introduced to simplify the result of
Eq. (2.35), and if we multiply the cross sectional area, heat flow ( ) Eq. (2.36), can also be
evaluated.
Where
0
Q

(2.31)

(2.32)

(2.33)

(2.34)

Or
(2.35)

(2.36)






 
dx
dT
K q
2 0







 
 
1
0 1
0 1 0
K
x x
q T T







 
 
2
1 2
0 2 1
K
x x
q T T















2 2
1 2
1
0 1
0
0
1 1
h K
x x
K
x x
h
T T
q
b a
 
b a
T T U q  
0
   
b a
T T WH U Q  
0
62

It can be observed two parameters, the heat transfer coefficients h0 and h2, are yet to be
established from Eq. (2.37). Calculation of these two parameters is accomplished using the Nusselt
number. The Nusselt number is calculated as a function of free and forced convection,
Nufree = f(Ra,Pr) and Nuforced = f(Re,Pr), respectively. Where Ra, Pr and Re are the Rayleigh, Prandtl
and Reynolds numbers, Eq. (2.38), Eq. (2.39) and Eq. (2.40), respectively.
Where Grx, Cp, μ, ρ, ν, and D are the Grashof number at position x, heat capacity, dynamic
viscosity, density, velocity and characteristic linear dimension (hydraulic diameter of pipe),
respectively. The OVPD falls under the forced convection category, therefore Eq. (2.39) and
Eq. (2.40) can be used. Nusselt numbers vary depending on the type of flow we have and the
Prandtl number values. For the turbulent flow case we can use the Dittus-Boelter relationship,
19

Eq. (2.41).
Where n is 0.4 or 0.3, for hot and cold fluids, respectively. The Dittus-Boelter equation is
valid for 0.7 ≤ Pr ≤ 160 and Re < 10,000. If laminar flows are encountered, the Shah relationship,
21

Eq. (2.42), is the appropriate equation to use to calculate Nusselt numbers and it’s valid for
Reynolds numbers below Re < 2300.

(2.37)

(2.38)

(2.39)

(2.40)

(2.41)
2 2
1 2
1
0 1
0
1 1 1
h K
x x
K
x x
h U




 
 
r x a
P Gr R * 







K
Cp
P
r
 *










 v D
R
e
* *
n
r e
P R Nu * * 023 . 0 5
4

63

Calculation of both, Reynolds and Prandtl numbers for water and nitrogen in the OVPD
yields ReH2O = 89,056.2, ReN2 = 16.6, PrH2O = 6.4, and PrN2 = 0.8. Using the appropriate equations, the
Nusselt numbers for water and nitrogen are NuH2O = 441.1 and NuN2 = 4.5. The heat transfer
coefficients are then found to be h0 = 85284.5W/m
2
·K and h2 = 117.1W/m
2
·K and the heat flux
q0 = -27,697.5W/m
2
. The results obtained from the above equations can be seen in Table 2.6, as
well as temperatures measured inside the OVPD. The measurements were taken using an OMEGA
type-K surface thermocouple under vacuum at 1torr and N2 flows of 40sccm. The temperature
conditions surrounding the substrate holder were varied from 250
o
C to 450
o
C. As seen in Table
2.6, the calculated and measured temperatures are in agreement and thus having a higher
sublimation temperature will impact significantly the temperature of the substrate holder and the
substrate. At a sublimation temperature of 250°C, the substrate can reach a temperature of 71°C,
while the hot N2 carrier gas impacts the substrate. Conversely, a sublimation temperature of 450°C
can increase the substrate temperature up to 122°C.

(2.42)



































3 . 33 Pr* Re* Pr* Re* * 0722 . 0 364 . 4
3 . 33 Pr* Re* Pr* Re* * 953 . 1
3
1
L
D
for
L
D
L
D
for
L
D
Nu
64

2.2.7 Optical electrical field distribution technique

Another approach in evaluating and optimizing OPV devices is the analysis of surface
wave transmission, or electromagnetic waves in thin films channeled by a refractive index
gradient, also known as the optical electrical field distribution technique.
16
The microwave field
theory applies to structures where the wavelength is smaller or equal to the dimension of the
structure, wave guided and also resonances in structures. The theory is derived from Maxwell’s
Equations. A favored technique to evaluate the transmission and reflection coefficients in the
electromagnetic field is the use of matrices. OPV device films can be described as isotropic,
homogenous and with parallel plane interfaces. The governing equations for the propagation of
the electric field are linear and the tangential part of the electric field is continuous, thus a 2 x 2
matrix can be utilized to describe the optical electric field inside OPV devices.
Furnace
Temperature
Surrounding
Substrate
holder (
°
C)
Measured Interface
Temperature (
°
C)
Calculated Interface
Temperature (
°
C)
SS/Quartz Quartz/N2 SS/Quartz Quartz/N2
250 42.4 71.3 38.12 62.11
300 46.2 81.4 41.94 72.89
350 52 94.1 46.04 84.48
400 58.4 107.8 50.41 96.86
450 65.8 121.8 55.05 109.98
Table 2.6.- Measured and calculated substrate holder and substrate
temperature chart at several temperature conditions
65

If a plane wave taking place from left reaches the films with thicknesses dj of an OPV
device, the optical electric field at any point in the OPV device is represented by a propagating
electric field in the positive x direction and another propagating in the negative direction, resulting
in the total optical electric field in the OPV device, Figure 2.27. The optical properties of the films,
described by individual complex indices, are a function of incident light wavelength. At position
x and layer j (j = 1, 2….m), the positive and negative optical electric field is represented by Ej
+
(x)
and Ej
-
(x), respectively.

Each interface can then be described by an interface matrix, Eq. (2.43) where rjk and tjk are
the Fresnel complex reflection and transmission coefficients, respectively, at the interface jk.

(2.43)







1
1
1
jk
jk
jk
jk
r
t
t
I

Figure 2.27.- Optical electric field schematic inside an OPV
device
y
x
............
S
''
j
d
j
Substrate
ITO Pt(
tbu
TPBP) C
60
BCP Al
............ Layer m Layer j Layer 2 Layer 1 Ambient
E
+
2
E
+
j
E
+
0
E
+
m
E
-
0
E
-
2
E
-
j
E
-
m
E
-
1
E
+
1
E
+
m+1
E
-
m+1
S
'
j
66

If the electric field is perpendicular to the plane of incidence, the Fresnel complex
coefficients are given by Eq. (2.44) and Eq. (2.45).
And n0 is the refractive index of vacuum, ϕ0 is the angle of incidence, ϕj is the angle of refraction
in layer j. Wave propagation through each layer is described by Eq.(2.47) by a layer or phase
matrix
And ξjdj the layer phase thickness representing a phase change the wave experiences as it travels
layer j. Combining Eq. (2.43) and Eq.(2.47) a total system matrix S, also known as scattering
matrix, which relates the electric field at ambient side to the substrate side can be deduced, Eq.
(2.45).
Where the total system matrix is

(2.44)

(2.45)
Where

(2.46)

(2.47)
Where

(2.48)

(2.49)
k j
k j
jk
q q
q q
r



k j
j
jk
q q
q
t


2
 
2
1
0
2
0
2
sin cos   n n n q
j j j j
  








j j
j j
d i
d i
j
e
e
L


0
0
j j
q



2




















1
1
0
0
m
m
E
E
S
E
E
67

It is safe to assume wave propagation in the negative direction to be zero, E
-
m+1 = 0, when
light is incident from the ambient side in the positive direction, as the metal cathode prevents
incident light penetration, hence wave propagation inside the device. Therefore, a total layered
complex reflection and transmission coefficients, Eq.(2.51) and Eq. (2.52), can be found by
rearranging the total system transfer matrix, Eq.(2.49) and Eq. (2.50).
In order to simplify calculating the optical electric field inside the device, the system can
be divided into two systems separated by layer j, allowing us to rewrite the total system transfer
matrix as
And the partial system transfer matrices for layer j to the left of layer j

are defined by
Where E ’j
+
and E ’j
-
are the left boundaries (j – 1) · j of layer j and the partial system transfer
matrices for layer j to the right of layer j are defined by

(2.50)

(2.51)

(2.52)
(2.53)

(2.54)

(2.55)
    1
1
1
22 21
12 11



 





 







 m m v
m
v
v v
L
S S
S S
S
11
21
0
0
S
S
E
E
r  


11 0
1
1
S E
E
t
m
 



' ' '
j j j
S L S S 

















'
'
'
0
0
j
j
j
E
E
S
E
E
    j j v
j
v
v v j
L
S S
S S
S
1
1
1
1
'
22
'
21
'
12
'
11 '













 








68

Where E ”j
+
and E ”j
-
are the right boundaries (j + 1) · j of layer j. It is also possible to define
complex reflection and transmission coefficients for layer j using Eq.(2.54), Eq. (2.56) and Eq.
(2.57) and obtain
Derivation of an internal transfer coefficient in the positive direction can be found by
combining Eq. (2.51) through Eq. (2.61), which relates the positive electric field in layer j with
respect to the positive ambient electric field (incident plane wave)
Where

(2.56)

(2.57)

(2.58)

(2.59)

(2.60)

(2.61)

(2.62)

(2.63)



















1
1 ' '
' '
' '
m
m
j
j
j
E
E
S
E
E
    1
1
1
' '
22
' '
21
' '
12
' '
11 ' '

 










 







 m m v
m
j v
v v j
L
S S
S S
S
'
11
'
21 '
j
j
J
S
S
r 
'
11
'
21 '
j
j
J
S
S
r 
' '
11
' '
21 ' '
j
j
j
S
S
r 
' '
11
' '
1
j
j
S
t 
j j
d i
j j
j j
j
e r r
t
E
E
t
 2
' ' '
'
0
1





 
'
11
'
12 '
j
j
J
S
S
r  

69

An internal transfer coefficient in the negative direction can also be derived, which relates
the electric field in the negative direction of layer j with respect to the positive ambient electric
field
Combining Eq. (2.62) and Eq.(2.64), a total electric field for layer j at a distance x to the
right of the boundary (j – 1) · j

in terms of the incident plane wave, we can establish
The power absorbed by the device at position x per photon incident is directly dependent
on the energy absorbed and the energy dissipation of the electromagnetic field inside the device.
The power absorbed of the electromagnetic field of energy dissipated per second in layer j at
position x is given by Eq.(2.68)
Where c is the speed of light, ɛ0 is the electric constant, ηj is the refractive index (wavelength
dependent), |Ej(x)|
2
is the modulus squared of the electric field and αj is the absorption coefficient
of each film, also wavelength dependent,
And by expanding Eq. (2.68) and using Eq. (2.67), the power absorbed equation becomes

(2.64)
(2.65)
(2.66)
(2.67)

(2.68)

(2.69)
j j
j j
j j
d i
j j d i
j j
d i
j j j
j
e r t
e r r
e r t
E
E
t



2
' '
2
' ' '
2
' ' '
0
1







 
) ( ) ( ) ( x E x E x E
j j j
 
 
 
   
 
0
) ( E e t e t x E
x i
j
x i
j j
j j
 
 
  
 
0
) 2 ( ' '
) ( E e r e t x E
x d i
j
x i
j j
j j j
 
2
0
) (
2
1
) ( x E c x Q
j j j j
   



j
j
4

70

where Tj = ( ηj / ηj )|tj
+
|
2

is the internal transmittance, ρ
”
j is the absolute value of the complex
reflection coefficient for the second subsystem, δ
”
j

is the argument value of the complex reflection
for the second subsystem, both acquired using Eq. (2.60) and I0 is the intensity of the incident light.
Using Fick’s second law, similarly to the SR-PLQ method, an exciton diffusion profile for
light incident involving the optical electric field inside the device can be obtained, Eq. (2.71).
Where D, is the diffusion constant, θ1 is the quantum efficiency, hv

is the excitation incident light
energy (Planck’s constant and frequency respectively) and τ is the lifetime of the exciton
generation.
Again consider the system to be at steady state, Eq. (2.49) and solve the equation to find
its general solution, Eq. (2.50), where A and B, Eq. (2.77) and Eq. (2.78), are constants that can be
found by using boundary conditions, identical to the SR-PLQ technique and C1 and C2 are given
by Eq. (2.75) and Eq. (2.76).
Where the exciton diffusion length is given by the following term

(2.70)

(2.71)

(2.72)
 














       
   
" "
) 2 (
2 "
0
4
cos 2 ) (
j
j d
j
x d
j
x
j j j
x d e e e I T x Q
j j j j j



  
  
) (
) ( ) (
1
2
2
x Q
hv
x n
x
x n
D
t
n
j


 





0 ) (
) ( ) (
2
2
  


x Q
Dhc D
x n
x
x n
j


71

The short exciton current density at the interface x is given by Eq. (2.79) and is related to
the photo current by Jphoto = q · θ2 · JExc, where θ2 is the quantum efficiency dissociation of the
exciton at x = 0 interface and q is the electron charge. Solving for the photocurrent Jphoto at x = 0
Eq. (2.80) can be found.
The same can be done for short circuit photocurrent Jphoto

at the interface x = d, resulting
in Eq. (2.82).

(2.73)

(2.74)
(2.75)

(2.76)

(2.77)

(2.78)

(2.79)

(2.80)

(2.81)
 D L
D

2
 












         


 
"
4
cos
) (
) (
2 1 2 2
1



 
 
   
x d C e C e e B e A
D
TN
x n
x x x x
j j
d
j
e C



 
"
1
j j
d
j
e C


  
  




"
2 2
2 2
2
2
) ) / 4 ( (
) (
) (
) " cos( "
4
cos cos ) ( ) (
2 1
d d
d d d d d
e e
d e C e e C e e
A
 
    
 
















     



) (
) " cos( "
4
cos ) ( ) (
2 1
d d
d d d d d
e e
d e C e e C e e
B
 
    
 
















     


   
0 

x
Exc
dx
dn
D J












         



"
4
sin
4
) (
2 1 2 2 0





   
 

d C C B A
TN q
J
x
Photo
d x
Exc
dx
dn
D J

 
72

Since all necessary equations have been solved, the next step is to write a MATLAB
program in order to calculate the optical electric field, power absorbed, extract the exciton
diffusion length, model the short circuit photocurrent of a given device, calculate the absorbance
and reflection of the device, Eq. (2.83) and Eq. (2.84), and model the incident photon to current
efficiency (IPCE), Eq. (2.85).
Where r is given by Eq. (2.51).
Where Jphoto is the short circuit current in μA/c m
2
, λ

and I0 are the wavelength and intensity in nm
and W/m
2
, respectively.
The optical constants for each material used in the OPV device, Figure 2.28, were obtained
using a J.A. Woollam Co., Inc. VASE Ellipsometer. The organic material was deposited with the
OVPD on silicon substrates with film thicknesses varying from 40 to 50nm. The Al film was
deposited using the VTE and the Pyrex substrate was purchased coated with ITO with thickness
varying from 100 to 120nm.

(2.82)
(2.83)

(2.84)

(2.85)
 






    


 

" sin
4
) (
2 1 2 2 0



   
 

   
C e C e Be Ae
TN q
J
d d d d
x
Photo
R A   1
2
r R 
O
Photo
I
J
IPCE




1240
(%)
73

Figure 2.28.- Refractive indeces and extinction coefficients of ITO, Pt(
tbu
TPBP),
C60, BCP and aluminum
400 500 600 700 800 900
0.0
0.5
1.0
1.5
2.0
2.5
n
k
40nm Pt
tbu
TPBP Film

Optical Constants
Wavelength (nm)
400 500 600 700 800 900
0.0
0.4
0.8
1.2
1.6
2.0
2.4
n
k
100nm ITO Film

Optical Constants
Wavelength (nm)
400 500 600 700 800 900
0.0
0.4
0.8
1.2
1.6
2.0
2.4
2.8
n
k
40nm C
60
Film

Optical Constants
Wavelength (nm)
400 500 600 700 800 900
0.0
0.4
0.8
1.2
1.6
2.0
2.4
0.00
0.01
0.02
0.03
0.04
0.05
0.06
n
50nm BCP Film
Optical Constants
Wavelength (nm)
k
300 400 500 600 700 800 900
1
2
3
4
5
6
7
n
k
100nm Al Film

Optical Constants
Wavelength (nm)
74

Once we have the optical constants of every material used in the OPV device, they are
introduced into the MATLAB program to analyze the optical electric field and power absorbed by
the device. An analysis of different thicknesses of the donor layer at two wavelengths, 350nm and
430nm was performed and it was observed that at 350nm there is no significant impact on the
exciton generation for a given thickness of Pt(
tbu
TPBP), Figure 2.29. However, for a 430nm
wavelength, a clear thickness dependence can be observed as we increase Pt(
tbu
TPBP) thickness.
From the plot an optimum thickness seems to be located between 5nm to 20nm.

Analysis of the power absorbed, relative to the optical electric field at 350nm and 430nm
corroborate the optimum thickness of Pt(
tbu
TPBP) to be between 5nm to 20nm, as it can be seen
in Figure 2.30.

Figure 2.29.- Optical electric field for different thicknesses of a Pt(
tbu
TPBP)
device at 350nm and 430nm wavelengths
80 100 120 140 160 180 200
0.0
0.2
0.4
0.6
0.8
1.0
Optical Electric Field Distribution Plot
ITO/Pt(
tbu
TPBP)
*
/C
60
40nm
/BCP
10nm
/Al
100nm
ITO

/E/
2
Distance from Glass/ITO Interface (nm)
Thickness of
Pt(
tbu
TPBP)
*
5nm
10nm
15nm
20nm
30nm
40nm
60nm
80nm
100nm
Analysis @
350nm
80 100 120 140 160 180 200
0.0
0.2
0.4
0.6
0.8
1.0
ITO
Optical Electric Field Distribution Plot
ITO/Pt(
tbu
TPBP)
*
/C
60
40nm
/BCP
10nm
/Al
100nm

/E/
2
Distance from Glass/ITO Interface (nm)
Thickness of
Pt(
tbu
TPBP)
*
5nm
10nm
15nm
20nm
30nm
40nm
60nm
80nm
100nm
Analysis @
430nm
75

However, further analysis can be made using the program, such as plotting the entire
spectral range (345-900nm) to see the behavior of the optical electric field and power absorbed of
the device with a 20nm Pt(
tbu
TPBP) thickness, shown on Figure 2.31 and Figure 2.32. From the
optical electric field we observe the maximum occurring between 100nm to 160nm, which is what
we expect since absorption of Pt(
tbu
TPBP) and C60 occur in that region. From the power absorbed
plot we can see the contributions from the Pt(
tbu
TPBP) and C60 films. C60 exhibits a strong power
absorbed from 500nm to higher energy, while on the other hand, Pt(
tbu
TPBP) exhibits power
absorbed in two areas equal in energy to its Soret and Q band, again proving an anticipated result
based on the absorption coefficients of the organic materials.

Figure 2.30.- Power absorbed for a Pt(
tbu
TPBP) device for 350nm and 430nm
wavelengths
80 100 120 140 160 180 200
0.0
5.0x10
-6
1.0x10
-5
1.5x10
-5
2.0x10
-5
2.5x10
-5
ITO
Optical Electric Field Distribution Plot
ITO/Pt(
tbu
TPBP)
*
/C
60
40nm
/BCP
10nm
/Al
100nm

Power Absorbed (Q)
Distance from Glass/ITO interface (nm)
Thickness of
Pt(
tbu
TPBP)
*
5nm
10nm
15nm
20nm
30nm
40nm
60nm
80nm
100nm
@ 430nm
80 100 120 140 160 180 200
0.0
5.0x10
-6
1.0x10
-5
1.5x10
-5
2.0x10
-5
2.5x10
-5
3.0x10
-5
ITO
Optical Electric Field Distribution Plot
ITO/Pt(
tbu
TPBP)
*
/C
60
40nm
/BCP
10nm
/Al
100nm

Power Absorbed (Q)
Glass/ITO Interface Distance (nm)
Thickness of
Pt(
tbu
TPBP)
*
5nm
10nm
15nm
20nm
30nm
40nm
60nm
80nm
100nm
@ 350nm
76

Figure 2.31.- Optical electric field for a device with 20nm Pt(
tbu
TPBP) film
thickness

Figure 2.32.- Power absorbed for a device with 20nm Pt(
tbu
TPBP) film
thickness
77

Reflection and absorption of the films inside the device was also calculated using the
optical constants, Eqs. (59-60), and the results show similar absorption patterns of the Pt(
tbu
TPBP)
and C60 films, Figure 2.33.

From the data obtained from the OPV devices, we are able to model the current density,
exciton density profile, incident monochromatic photon to current efficiency (IPCE) and extract
the exciton diffusion length for both the donor and acceptor layer. Fitting the experimental data to
the current density equations was performed by varying photocurrent, diffusivity and lifetime
while film thickness was kept constant, Figure 2.34. It’s important to note that thickness can also
be varied during the fit; however, it was not done in the results shown below. The results for the
IPCE fit can be observed in Figure 2.35. From the fit the LD for the donor Pt(
tbu
TPBP), and acceptor
C60, was found to be 16.9nm ± 4nm and 14.8nm ± 3.5nm, respectively. This result is in close
agreement with the LD obtained for Pt(
tbu
TPBP) using the SR-PLQ technique,

Figure 2.33.- Measured absorption and reflection inside a device with
20nm Pt(
tbu
TPBP) film thickness
78

LD = 12.1nm ± 2.4nm. From the results obtained from the fit, namely diffusivity and life time, the
exciton density profile of the donor and acceptor can also be plotted, Figure 2.36 and Figure 2.37,
respectively. The density profile essentially illustrates the absorbance profile of both molecules
and is a good way to validate the processed data has no mistakes.

Figure 2.34.- Modeled current density of a 10nm and 20nm Pt(
tbu
TPBP) film
thickness device
400 500 600 700 800
0.000
0.005
0.010
0.015
0.020
20nm Experimental
20nm Fitted

ITO/Pt(
tbu
TPBP)
20nm
/C
60
40nm
/BCP
10nm
/Al
100nm
Current Density, mA/cm
2
Wavelength, nm
400 500 600 700 800
0.000
0.005
0.010
0.015
0.020
10nm Experimental
10nm Fitted

ITO/Pt(
tbu
TPBP)
10nm
/C
60
40nm
/BCP
10nm
/Al
100nm
Current Density, mA/cm
2
Wavelength, nm

Figure 2.35.- Modeled incident photon to current efficiency of a 10nm and 20nm
Pt(
tbu
TPBP) film thickness device
400 500 600 700 800
0
10
20
30
40
50
60
70
10nm Experimental
10nm Fitted

ITO/Pt(
tbu
TPBP)
10nm
/C
60
40nm
/BCP
10nm
/Al
100nm
IPCE %
Wavelength, nm
400 500 600 700 800
0
10
20
30
40
50
60
70
20nm Experimental
20nm Fitted

ITO/Pt(
tbu
TPBP)
20nm
/C
60
40nm
/BCP
10nm
/Al
100nm
IPCE %
Wavelength, nm
79

Figure 2.36.- Modeled exciton density profile for a 20nm Pt(
tbu
TPBP) film
thickness device

Figure 2.37.- Modeled exciton density profile for a 40nm C60 film thickness
device
80

2.3 Conclusions

The Exciton diffusion length (LD) of a novel molecule, platinum tetra
1, 3-di-tert-butylphenyl tetrabenzoporphyrin Pt(
tbu
TPBP) was measured using two methods, the
spectrally resolved photoluminescence quenching SR-PLQ method and the optical electric field
distribution technique. Both techniques gave insight into the optimal film thickness of Pt(
tbu
TPBP)
to maximize efficiency on a solar cell. Quenching of the Pt(
tbu
TPBP) films was observed when the
SR-PLQ method was performed under N2, affecting LD results significantly. Performing the
measurements under vacuum prevented O2 quenching and an average LD of 12.1nm ± 2.4nm was
acquired for Pt(
tbu
TPBP) using the SR-PLQ method. Conversely, an LD of 16.9nm ±4nm and
14.8nm ±3.5nm was obtained for Pt(
tbu
TPBP) and C60, respectively, using the optical electric field
distribution technique. Both techniques proved to be a great tool for the design and optimization
of OPVs. Further, modeling of the optical electric field, photocurrent generation, exciton diffusion
profile density, absorption, reflection, among other was also completed using the optical electric
field distribution method.
Photovoltaic cells were made and evaluated with the following structure:
Substrate/ITO/Pt(
tbu
TPBP)
10-60nm
/C60
10-40nm
/BCP
10nm
/Al
100nm
. Devices were fabricated using an
organic vapor phase deposition and a vacuum thermal evaporation (reference). Open circuit
voltages in the order of 0.5V and 0.65V were obtained for devices made in the OVPD and VTE,
respectively. The results suggest intermixing of the donor/acceptor layers during deposition of C60
in the OVPD devices, thus leading to strong intermolecular interactions. Further, intermixing is
common when films are annealed and strong intermolecular interactions are known to increase
dark current, consequently, decreasing the Voc. The efficiencies obtained for the Pt(
tbu
TPBP)
81

devices were found to be 0.6% and 1.2% for the OVPD and VTE, respectively. The efficiency
discrepancy is due to poor current contribution from the C 60 layer (10nm for OVPD vs 40nm for
VTE). The external quantum efficiencies (EQE) from the OVPD devices clearly show weak
contribution from the acceptor (C60), whereas a stronger current contribution from C60 can be
observed on the VTE devices. The results suggest efficiency can be improved for devices made in
the OVPD if a thicker C60 film is employed. Similarly, the Voc can also be improved with proper
substrate cooling to prevent intermixing of the donor/acceptor layers. Further, a similar Jsc trend
was observed on the Pt(
tbu
TPBP) versus the Pt(TPBP) devices, confirming a short LD in both
molecules is accountable for the decrease in current when a thicker donor film is employed.

2.4 Experimental
2.4.1 Substrate preparation

Patterned ITO substrates with a resistivity of 1:20 ± 5 ohms/sq and an ITO thickness of
2000 ± 50Å were used for the OPVs made in the OVPD. The substrates, patterned ITO, silicon
and quartz were prepared/cleaned by scrubbing them with Tergitol NP9 (SigmaAldrich Co.)/DI
solution followed by a thorough rinse with DI water. The substrates were then rinsed with acetone
(Sigma-Aldrich Co.) followed by 2-Propanol (SigmaAldrich Co.) and were then blow dried with
N2 gas. They were then washed with tetrachloroethylene (J.T. Baker), acetone (Macron Chemical)
and ethyl alcohol anhydrous reagent (J.T. Baker). The wash consists of placing the substrates, in
the order mentioned above, inside a beaker with each solvent for 10 min while heating the solvent
to its boiling point. After washing (boiling), the substrates were blow dried with N2 gas and placed
82

in a ultra-violet ozone cleaning system, model T10X10/OES for 10 min. Substrates were then
transferred to the OVPD for device fabrication.

2.4.2 Device architecture

The OPV device structure consisted of platinum tetra 1, 3-di-tert-butylphenyl
tetrabenzoporphyrin (Pt(
tbu
TPBP) – 10 to 40nm), Buckminsterfullerene (C60 – 10 to 40nm), 2,9-
Dimethyl-4,7-diphenyl-1,10-phenanthroline (BCP – 10nm) and aluminum (Al – 100nm). The
organic materials C60 and BCP were purchased from Sigma-Aldrich Co and MER Corporation,
respectively. Pt(
tbu
TPBP) was synthesized using the method described in section 2.3.5 [M. Whited,
et. al. 2011]. All organic materials were purified at least once via vacuum-train sublimation prior
to use in the OVPD.

2.4.3 Fabrication of organic solar cells

Organic films grown in the OVPD were performed under a pressure of 1 torr, gas flow
rates of 40 sccm and organic deposition source temperatures of 390°C – 410°C, 200°C – 220°C
and 520°C – 540°C for Pt(
tbu
TPBP), BCP and C60, respectively. Deposition rates of 1.0-2.0 Å/s
were used during the deposition of the organic material. The electrode (Al) was deposited using a
Kurt J. Lesker VTE system. Depositions or organics and electrodes in the VTE were performed
under a pressure of ≤ 4 x 10
-4
Pa, deposition rates of 2 – 4 Å/s and at room temperature. Material
thickness in both the OVPD and VTE was controlled using a 6MHz Inficon quartz monitor gold
coated crystal sensor. Proper calibration of the Inficon crystal sensor via spectroscopic
83

ellipsometry was performed using a J.A. Woollam Co., Inc. VASE variable-angle ellipsometer
with a VB-200 control module and a CVI instruments Digikrom 242 monochromator with a 75 W
xenon light source to ensure accurate thickness of films.

2.4.4 OPV testing

OPV current density (J) as a function of applied voltage (V) characteristics were measured
in air at room temperature, in the dark and under spectral mismatch corrected 100 mW/cm
2
white
light illumination from an AM-1.5G filtered 300 W Xenon arc lamp (Newport Inc.) and a Keithley
power source meter model 2635A. Routine spectral mismatch correction for ASTM G173-03 was
performed using a filtered silicon photodiode, calibrated by the National Renewable Energy
Laboratory (NREL) to reduce measurement errors. Frequency modulated monochromatic light
(250 Hz, 10 nm FWHM) and lock-in detection was used to perform all spectral responsivity and
spectral-mismatch correction measurements.

2.4.5 Synthesis of platinum 1, 3-di-tert-butylphenyl
Tetraphenylbenzoporphyrin

Platinum tetra 1, 3-di-tert-butylphenyl tetrabenzoporphyrin Pt(
tbu
TPBP), Figure 2.38, was
synthesized mixing 4,5,6,7-tetrahydroisoindole with 3,5-Di-tert-butyl benzaldehyde in the
presence of 1) BF3•Et2O, boron trifluoro ethyl ether, followed by 2) DDQ, 2,3-Dichloro-5,6-
dicyano-1,4-benzoquinone which is then metallated followed by an oxidation step to the
84

benzoporphyrin [M. Whited, et. al. 2011]. The chemical structures of Pt(
tbu
TPBP), BCP and C60
are shown in Figure 2.39.

Figure 2.38.- Synthesis steps of Platinum tetra 1, 3-di-tert-butylphenyl
tetrabenzoporphyrin Pt(
tbu
TPBP)
H
N
+
Ar H
O
1) BF
3
·Et
2
O
2) DDQ
N
NH N
HN
Ar
Ar
Ar
Ar
M
2+

N
N N
N
Ar
Ar
Ar
Ar
M
N
N N
N
Ar
Ar
Ar
Ar
M
DDQ

Ar=
M
2+
=PtCl
2
DDQ=
BF
3
Et
2
O=
O
O
CN
CN Cl
Cl
B
+
O
-
Pt(
tbu
TPBP)

Pt(
tbu
TPBP) BCP C60
Figure 2.39.- Chemical structures of Pt(
tbu
TPBP), BCP and C60
N
N
N
N
Pt
N N
85

2.5 Chapter 2 References

1. Lin, L.-Y.; Chen, Y.-H.; Huang, Z.-Y.; Lin, H.-W.; Chou, S.-H.; Lin, F.; Chen, C.-W.;
Liu, Y.-H.; Wong, K.-T. J. Am. Chem. Soc. 2011, 133, (40), 15822–15825.
2. Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.; Yu, L. Advanced
Materials 2010, 22, (20), E135–E138.
3. Zhou, H.; Yang, L.; Stuart, A. C.; Price, S. C.; Liu, S.; You, W. Angewandte Chemie
International Edition 2011, 50, (13), 2995-2998.
4. Laboratory, N. R. E. Research Cell Efficiency Records. (29),
5. GmbH, H., http://www.heliatek.com, 2010.
6. Scharber, M. C.; Mühlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A. J.;
Brabec, C. J. Advanced Materials 2006, 18, (6), 789-794.
7. Giebink, N. C.; Wiederrecht, G. P.; Wasielewski, M. R.; Forrest, S. R. Phys. Rev. B 2010,
82, (15), 155305.
8. Giebink, N. C.; Lassiter, B. E.; Wiederrecht, G. P.; Wasielewski, M. R.; Forrest, S. R.
Phys. Rev. B 2010, B2, (15), 155306.
9. Borek, C.; Hanson, K.; Djurovich, P. I.; Thompson, M. E.; Aznavour, K.; Bau, R.; Sun,
Y.; Forrest, S. R.; Brooks, J.; Michalski, L.; Dr., J. B. Angewandte Chemie International Edition
2007, 46, (7), 1109-1112.
86

10. Mark, T. Mater. Res. Soc. Bull. 2007, 32, (09), 694-701.
11. Currie, M. J.; Mapel, J. K.; Heidel, T. D.; Goffri, S.; Baldo, M. A. Science Magazine
2008, 321, (5886), 226-228.
12. Perez, M. D.; Borek, C.; Forrest, S. R.; Thompson, M. E. J. Am. Chem. Soc. 2009, 131,
(26), 9281–9286.
13. Perez, M. D.; Borek, C.; Djurovich, P. I.; Mayo, E. I.; Lunt, R. R.; Forrest, S. R.;
Thompson, M. E. Advanced Materials 2009, 21, (14 ‐15), 1517-1520.
14. Rand, B. P.; Burk, D. P.; Forrest, S. R. Phys. Rev. B 2007, 75, (11-15), 115327.
15. Lunt, R. R.; Giebink, N. C.; Belak, A. A.; Benziger, J. B.; Forrest, S. R. Journal of
Applied Physics 2009, 105, (5), 053711.
16. Pettersson, L. A. A.; Roman, L. S.; Inganäs, O. Journal of Applied Physics 1999, 86, (1),
487-496.
17. Lin, Y.; Li, Y.; Zhan, X. Chem Soc Rev 2012, 41, (11), 4245-72.
18. Kampas, F. J.; Gouterman, M. J Lumin 1976, 14, (2), 121-129.
19. R. Byron Bird, W. E. S., Edwin N. Lightfoot, Transport Phenomena. Wiley: New York,
2002; p 928.
20. Hoshino, T.; Mito, K.; Nagashima, A.; Miyata, M. International Journal of
Thermophysics 1986, 7, (3), 647-662.
87

21. Shah, R. K., Thermal entry length solutions for the circular tube and parallel plates. In
Proc. 3rd Natnl. Heat Mass Transfer Conference, Indian Inst. Technol. Bombay, 1975; Vol. 1.

88

Chapter 3. Metal deposition for optoelectronic devices using a
low vacuum vapor phase deposition (VPD)

3.1. Introduction

In this chapter we introduce a new technique involving a vapor phase deposition (VPD) as a
non-line-of-sight method to fabricate thin metal films. The VPD technique is an adaptation from
the organic vapor phase deposition (OVPD) method, previously introduced and explained in
chapter 1. Thin metal films have been used widely as cathodes and anodes for optoelectronic
devices such as organic light emitting diodes (OLEDs), organic photovoltaics (OPVs) and thin
film transistors (TFTs) to inject/extract charges from such devices.
1, 2
The most commonly used
techniques to deposit metals are electrodeposition and physical vapor deposition (PVD); among
the latter, vacuum thermal evaporation (VTE), a high vacuum method, is extensively utilized to
deposit thin metal films for OLEDs, OPVs and TFTs. The disadvantages of high vacuum methods
are long pump down times needed to achieve high vacuum (i.e., ≤10
-5
torr), inefficient utilization
of materials, poor film conformality and the high expense incurred when deposition is applied on
large substrates.
3
High vacuum is required to achieve long mean free paths and thereby allow the
metal atoms to reach the substrate without colliding, thus realizing smooth, uniform films.
However, deposition using VTE occurs along a ballistic (directional line-of-sight) trajectory
preventing the formation of conformal films on nonuniform substrates. Therefore, interest has
developed towards non-line-of-sight deposition techniques such as chemical vapor deposition
(CVD) and atomic layer deposition (ALD).
4
Moreover, in contrast to the VTE methods used to
deposit metal films for organic electronic applications, which involve vacuums ranging from
10
-4
– 10
-10
torr, the VPD employs a comparatively low vacuum ranging from 0.1–10 torr. Other
89

methods to deposit metals at low pressure (10
-3
–760 torr), such as sputtering, metal-organic
chemical vapor deposition (MOCVD) or atomic layer deposition (ALD), have limitations that
make them unsuitable for fabricating organic electronic devices. Sputtering relies on plasma to
produce charged ions and high speed electrons that can potentially damage organic thin films.
MOCVD requires a hot substrate ( 100°C) in order to decompose the metal-organic to produce a
thin metal film. The heated substrate will, in turn, disrupt morphology of the organic films,
particularly in multilayer heterostructures. Further, MOCVD can cause organic contamination,
also known as parasitic deposition, during the decomposition of the metal-organic compound.
ALD typically requires the successive application of precursor compounds in the gas phase,
necessitating lengthy growth times for thicker films such as those needed for OLEDs, OPVs and
TFTs (1–2 h for 100 nm films).
5
In addition, ALD requires high substrate temperatures ( 100°C)
to promote the successive reactions on the substrate surface to deposit metals, consequently
degrading organic films and/or causing parasitic deposition.
Since its inception from 1995 to date, OVPD has demonstrated to be a versatile, reliable and
efficient method to make organic thin films.
6
Specifically, the OVPD was designed for organic
materials that have low sublimation temperatures. Upon sublimation, the material is transported
by an inert carrier gas through a hot-wall vessel to avoid unwanted deposition prior to condensation
on a cooled substrate, thereby achieving high material usage. The rate at which the material
sublimes and ultimately condenses onto the substrate is controlled by both the temperature of the
source boat and the flow rate of the inert gas, which enhances control over the deposition process
and the morphology of the films.
7-9
However, while VPD processes have been used to deposit
molecular organic materials, forming high efficiency OLEDs,
10-15
OPVs
16-19
and TFTs,
8, 20-25
the
90

method has not been used for the deposition of metal films needed to fully complete fabrication of
the device.

3.2. Results and discussion
3.2.1. Deposition parameters

A series of metals were considered for use in the VPD, including calcium, zinc, cadmium,
magnesium, antimony, bismuth, indium and manganese, among others. Of these, magnesium (Mg)
and zinc (Zn) were selected to fabricate metal films and devices, as they have sublimation
enthalpies that are comparable to organic materials commonly used for OLEDs and OPVs (see
Table 1).
7, 26-28
Both metals exhibit low toxicity and have been used as electrodes to inject charge
into OLEDs.
2, 29, 30
The effective volatility of both metals was evaluated by calculation of molar
flow rates (r) using Eq. (3.1) below;
7
where V
̇ is the volumetric flow velocity of the carrier gas, R
is the ideal gas constant, Tcell is the temperature of the source boat and Porg/met is the vapor pressure
of the organic/metal.
The calculations show that Mg and Zn can have molar flow rates comparable to values
obtained for organics with high sublimation temperatures, such as CuPc and C60 (Table 3.1). In a
laminar flow regime, the diffusion of organics and metals within the carrier gas stream plays an
important role in determining the transport efficiency of the material. Diffusion coefficients for
Mg and Zn were calculated using the Chapman–Enskog theory, Eq. (3.2); where T is temperature,
M is the molar mass, p is pressure, Ω is the temperature dependence collision integral,
𝑟 = 𝑉 ̇ ∙
𝑃 𝑜𝑟 𝑔 / 𝑚 𝑒 𝑡 𝑅 𝑇 𝑐𝑒 𝑙𝑙
(3.1)
91

σ12 = ( σ1 + σ2)/2 is the average collision diameter, D is the diffusion coefficient, and 1 and 2 are
indices for the two molecules present in the gaseous mixture.
The calculated results are shown in Table 3.1. The calculations reveal that the diffusivity
of these metals are one order of magnitude greater than values determined for typical OLED and
OPV compounds. Such high diffusivities promote metal deposition in various regions of the
sample chamber upstream from the source nozzles. These problems can potentially be averted by
increasing the total flow of the carrier gas. However, an increase in carrier gas flux can raise the
operating pressure if the pumping system is not suitable for large flows, thus affecting film
morphology and crystallinity. In our system, condensation was observed upstream on the rear
flange of the VPD chamber when the pump did not maintain a pressure of 1 torr in flows exceeding
60 standard cubic centimeters per minute (sccm). The inability to pump at a higher velocity
necessitated a modification in the deposition apparatus by inserting a fused silica wall around the
source tubes between zone 3 and zone 4 of our system, seen in Figure 3.1. This fused silica wall
is unnecessary when depositing materials with low diffusivity, such as organic compounds.

𝐷 =
1 . 858 𝑥 10
− 3
𝑇 3 2 ⁄
√ 1 𝑀 1
⁄ + 1 𝑀 2
⁄
𝑝 𝜎 12
2
𝛺
(3.2)
92

Crystal Monitor
Substrate Holder
Vacuum
Pump
Cooling Supply
Zone 1 Zone 2 Zone 3 Zone 4
Source Boat
Fused
Silica wall
Carrier Gas
Shutters
Zone 1 Zone 2 Zone 3 Zone 4

Figure 3.1.- Schematic of the modified vapor phase deposition
Table 3.1.- Chapman–Enskog diffusion coefficients and sublimation enthalpies for Zn, Mg,
CuPc, C60, Alq3 and NPD
Compound
Enthalpy of
sublimation
[KJ/mol]
a, b, c

Vapor
pressure
[atm]
b, d, *

Molar flow
rate [mol/s]
Diffusion
coefficient
[cm
2
/s]
Molar
mass
[g/mol]
Average
collision
diameter
[Å]
¥

nitrogen - - - - 28.0 4.2
zinc 130.4 ± 0.4
a
7.4E-03
d
8.0E-01 614.7 65.4 2.8
magnesium 147.1 ± 0.8
a
3.9E-04
d
4.2E-02 497.6 24.3 4.4
CuPc 211.1 ± 0.1
b
1.7E-03
b
1.8E-01 51.9 576.1 18.1
C60 219.6 ± 0.1
b
3.9E-06
b
4.2E-04 118.5 720.6 10.5
Alq3 137.7 ± 0.1
b
1.1E-04
b
1.5E-02 45.5 459.4 15.2
NPD 139.0 ± 0.3
c
- - 23.7 588.7 22.6
*
Temperatures used to calculate vapor pressures of Zn, Mg, CuPc, and C60 was 823 K and
623 K for Alq3 and NPD.
¥
Average collision diameter, ⁡𝜎 1
: N2 and 𝜎 2
: Zn, Mg, CuPc, C60, Alq3 and NPD were measured
using Titan 1.0.7 software (diameters have Van der Waals radii integrated).
Calculations of molar flow rates and diffusion coefficients were performed at a pressure of 1
torr. The collision integral (Ω) was taken as unity given that it has values between 0.96 and
1.03 for temperatures between 300 K and 1000 K.
a
W. Plieth (2008). Electrochemistry for Materials Science. Oxford, UK: Elsevier
b
K. Yase, Y. Takahashi, N. Ara-Kato & A. Kawazu. Jpn. J. Appl. Phys. 34, 636 (1995)
c
Shtein, M. Gossenberger, H. F.; Benziger, J. B.; Forrest, S. R. J. Appl. Phys. 89, 1470 (2001)
d
CRC Handbook of Chemistry and Physics, 94th Edition, 2013-2014. www.hbcpnetbase.com
(accessed December 12, 2013)
93

During the sublimation of Mg and Zn, the VPD chamber was kept at 450ºC (zone 1) and
550ºC (zones 2-4) to prevent the metals from condensing on the chamber walls. However, metals
such as Mg and Zn exhibit high surface mobility and poor wettability, a problem that can be
correlated to the critical density and critical temperature of the metals.
31-36
The critical density is a
measure of the number of atoms striking 1 cm
2
/s needed to achieve condensation on a substrate
and is directly related to the substrate temperature,
32, 33
whereas the critical temperature is reached
when the probability of an atom condensing on the substrate is unity.
33
The high temperatures in
the VPD chamber, along with the high surface mobility and poor wettability of Mg and Zn on a
hot substrate, requires efficient cooling to produce high quality films. Therefore, a liquid-cooled
substrate holder was built using aluminum (thermal conductivity ≈ 220 W/mK) in order to
maintain adequate control of the substrate temperature. In addition, during initial attempts at
fabricating metal films we found that a water/ethylene glycol (50/50) mixture cooled at -20°C was
incapable of keeping the substrate below 120°C. While this temperature was sufficient to promote
metal deposition, the organic films were not stable at these temperatures. Thus, we switched our
cooling medium to N2 gas cooled to -185°C. This last modification allowed us to reliably maintain
substrate temperatures below 50°C during metal deposition, and thereby prepare good quality
OLEDs and OPVs (vide infra).
The above modifications to the VPD chamber (fused Pyrex wall) and cooling system gave
deposition rates of 0.1–35.0 Å/s and enabled us to make metal films with thicknesses ranging from
100 to 10,000 nm. However, the elevated temperature of the chamber during metal deposition
caused organics previously condensed on the shutters to sublime, leading to cross contamination
during fabrication of the cathode and subsequent device failure. To overcome this obstacle and
94

avoid cross-contamination with organics during the fabrication of the cathode, the shutters were
cleaned prior to deposition of the metal.

3.2.2. Magnesium and zinc metal films

For comparative purposes, thin films of Mg and Zn were fabricated in both VPD and VTE
on silicon and patterned ITO. Metal films were deposited at rates ranging from 4–8 Å/s and for
VPD films, using substrate holder temperatures of 20–25°C and total flows of 60 sccm. Metal
films made by both methods were reflective and glossy. Calibration of the quartz crystal monitor
sensor for Mg and Zn films, deposited in the VPD on patterned ITO, was accomplished using
profilometry. The Mg and Zn films were analyzed to determine their morphology, roughness,
crystallinity and resistivity. Cross-sectional SEM images of Mg films deposited on silicon
substrates using VPD and VTE (Figure 3.2) show complete substrate coverage and a crystalline,
rough surface morphology. Films made using VPD display a loose packing of blade-like
nanocrystals whereas films made using VTE are more densely packed.
Three dimensional AFM images (Figure 3.3), confirmed the results showing both films
exhibit full substrate coverage and a rough surface texture. However, while the SEM and AFM
images of the VTE film appeared to have higher peak-to-valley aspect ratios than the VPD film,
the average film roughness/uniformity of both films is similar (RMS values for VPD = 36 ±7 nm,
VTE = 36 ±8 nm and bare silicon = 0.4 ±0.1 nm).

95

Grazing incidence XRD diffraction patterns collected for the VPD and VTE films show
peaks that can be assigned to the hexagonal phase of Mg and the underlying Si substrate (Figure
3.4).
37
The pattern of peak intensities of the VPD film is similar to the expected powder pattern,

Figure 3.2.- Cross-sectional SEM images of a 200 nm Mg film on a
silicon substrate made using either VPD (top x 8.5K) or VTE
(bottom x 55K). The VPD film was deposited at a substrate holder
temperature of 20-25°C
200nm

200nm
200nm

200nm
Si Substrate

Si Substrate
Si Substrate

Si Substrate
VTE

VTE
VPD

VPD

Figure 3.3.- 3D AFM image of 200 nm Mg films deposited on a
silicon substrate using either VPD (top) or VTE (bottom) methods
VPD

VPD
VTE

.-
Resistivit
y
measure
ments of
a VPD
400nm
Zn film
deposited
on a
silicon
substrate
VTE
96

indicating that the Mg crystallites are randomly arranged on the substrate during film growth. In
contrast, the VTE film shows the 002 peak is markedly higher intensity than expected from the
powder pattern indicating a preferred alignment of the Mg crystallites with the 002 planes parallel
to the substrate. Further, the response from the silicon lattice plane is attenuated in the XRD data
from the VTE film, indicating denser packing than the VPD film. However, we believe that
through appropriate modification of the deposition conditions (substrate temperature, chamber
pressure and carrier gas flow rates),
8, 9
the VPD method should lead to films with identical crystal
patterns to those made using VTE.

Four-point probe resistivity measurements were carried out to examine the electrical
properties of the metal films. Resistivities of 1.8 x 10
-7
Ω-m and 8.4 x 10
-8
Ω-m were obtained for
VPD and VTE Mg films, respectively. The observed higher resistivity of films made using VPD
30 40 50 60 70 80
VTE
VPD

Intensity (a.u.)
2  (degree)
(100)
(002)
(101)
(102)
(Silicon)
(110)
(103)
(112)
(200)
(004)
(202)

Figure 3.4.- XRD patterns of 200 nm Mg films deposited on a
silicon substrate using either VPD (top) or VTE (bottom)
methods. Blue lines are reference crystal planes of the
hexagonal phase of Mg
97

can be attributed to grain boundaries, (more pronounced in VPD metal films), and a looser
packing/growth of the crystallites during deposition, which can enhance air/water diffusion and
thus rapid oxidation of the film. The latter, associated with crystal orientation of grains
(crystallographic orientation), is known to play a role in the oxidation properties of metals, e.g.,
Mg and copper, by directly impacting the atomic packing density and surface energy of the
metals.
38-40
Further, these metals were shown to exhibit lower surface energy (high atomic packing
density) and a greater corrosion resistance, in solution (Mg) and thin films (Cu), under certain
crystallographic orientations, e.g., 0001(Mg), 100(Cu) and 110(Cu). Regardless, while the
resistivity is higher for films made using VPD, the value is still low enough to make it suitable for
use as a cathode/anode in organic devices.
Similarly, Zn films fabricated in the VPD and deposited on silicon substrates show
complete substrate coverage, but a rougher surface morphology than Mg films (RMS = 77 ± 3
nm), Figure 3.5. Resistivity measurements were also performed, Table 3.2, obtaining an average
value of 4.6 x 10
-6
Ω-m, much lower than the Mg films. Further, XRD diffraction pattern peaks
that can be assigned to various crystal planes of the hexagonal phase of Zn were obtained from the
films, Figure 3.6.
Table 3.2.- Resistivity measurements of a VPD 400nm Zn film deposited on a silicon substrate

Measurement #1 Measurement #2 Measurement #3 Average
Film Resistivity (Ω) 2.08 2.37 3.23 2.56
Film Thickness (nm) 400 400 400 400
Ω/square 9.44 10.73 14.64 11.60
Ω-m 3.78 x 10
-6
4.29 x 10
-6
5.85 x 10
-6
4.64 x 10
-6

98

Figure 3.6.- VPD XRD patterns of 400nm Zn films deposited on a
silicon substrate. Blue lines are reference crystal planes for the
Hexagonal phase of zinc from (MDI Jade 9 PDF# 03-065-3358)
30 40 50 60 70 80
(110)
(103)
(102)
(101)
(100)
(002)

Intensity (a.u.)
2(degree)

Figure 3.5.- VPD (x 15K) Cross-sectional SEM images of 400nm
Zn film on a silicon substrate. Film was deposited at a substrate
holder temperature of 20-25°C
99

3.2.3. Organic light-emitting devices with magnesium cathodes

Fluorescent and phosphorescent OLEDs with Mg as the cathode were fully fabricated using
either the VPD or VTE (reference) methods. The deposition rates used in both methods for
organics and metals were 1–4 Å/s and 4–8 Å/s, respectively. The fabrication of devices using the
VPD was done at a substrate holder temperature of 40–45 °C for organics and 20–25ºC for metals
and total carrier gas flows of 60 sccm. Fluorescent devices EQE spectra, brightness, emission,
turned on voltage, among other performance parameters can be seen on Figure 3.7a, Figure 3.8
and Table 3.3, respectively. Fluorescent OLEDs, Figure 3.7a, (ITO/NPD (40 nm)/Alq3 (40 nm))
made in both the VPD and VTE with 200 nm Mg cathodes exhibited similar turn-on voltages
(Von = 2.7 V at 1 cd/m
2
) and external quantum efficiencies (EQE = 0.9 ±0.1% at 100 cd/m
2
).
Phosphorescent devices EQE spectra, brightness, emission, turned on voltage, among other
performance parameters can be seen on Figure 3.7b, Figure 3.9 and Table 3.3. Phosphorescent
devices, Figure 3.7b, (ITO/NPD (40 nm)/CBP-Ir(ppy)3 7 wt% (30 nm)/BCP (10 nm)/Alq3
(40 nm)) made in the VPD had a higher turn-on voltage than devices made in the VTE (Von at
1 cd/m
2
= 4.1 and 3.3 V, respectively). A lower efficiency was observed for devices made in the
VPD than for those made in the VTE (EQE = 7.6 ±0.6 and 8.3 ±0.5% at 100 cd/m
2
, respectively).
The efficiencies of fluorescent and phosphorescent devices are comparable to values previously
reported for devices with the same architecture used here (EQE = 0.6-1.3% and EQE = 7.5-8.5%
at 100 cd/m
2
, respectively).
41-45

The current density vs voltage plots of OLEDs made by VPD and VTE are similar,
suggesting that the small differences in turn-on voltage and EQE between PHOLED devices is due
to exposure to air for approximately 5 minutes during exchange of source boats, electrode mask
100

placement and cleaning of the shutters. Fluorescent and phosphorescent devices fabricated in the
VPD exhibited analogous electroluminescence spectra to devices made in the VTE, due to Alq3
and Ir(ppy)3 emission, respectively.

2 4 6 8 10 12
0.0
0.2
0.4
0.6
0.8
1.0
1.2
2 4 6 8 10 12
0
200
400
600
800
Current Density (mA/cm
2
)
Voltage (V)
EQE (%)
Voltage (V)
VPD
VTE
10
-2
10
-1
10
0
10
1
10
2
10
3
10
4
Brightness (cd/m
2
)

2 4 6 8 10 12
-2
0
2
4
6
8
10
EQE (%)
Voltage (V)
VPD
VTE
10
-2
10
-1
10
0
10
1
10
2
10
3
10
4
2 4 6 8 10 12
0
50
100
150
200
Current Density (mA/cm
2
)
Voltage (V)

Brightness (cd/m
2
)

Figure 3.7.- External quantum efficiency/brightness vs voltage characteristics of (a) fluorescent
Alq3 and (b) phosphorescent Ir(ppy)3 OLEDs. Current density vs voltage characteristics can be
seen in the inset
(a)

(a)

(b)

(b)

Table 3.3.- Fluorescent and phosphorescent OLED device performance
Device
Turn on
Voltage @
1cd/m
2

EQE (%)
@
100cd/m
2

EQE (%)
@
1000cd/m
2

Power
Efficiency
lm/W @
2000cd/m
2

Current
Efficiency cd/A
@ 2000cd/m
2

VTE (Alq3) 2.7 0.8 ± 0.1 0.9 ± 0.1 1.2 ± 0.1 2.4 ± 0.2
VPD (Alq3) 2.7 0.9 ± 0.1 0.9 ± 0.1 1.3 ± 0.1 2.5 ± 0.2
VTE [Ir(ppy)3] 3.3 8.3± 0.5 6.5 ± 0.6 8.1 ± 1.2 17.7 ± 2.2
VPD [Ir(ppy)3] 4.1 7.6 ± 0.6 5.9 ± 0.3 5.9 ± 0.7 16.1 ± 0.7

101

Figure 3.8.- Normalized EL intensity of green Alq3 fluorescent
OLEDs
400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
1.2

Normalized EL intensity (a.u.)
Wavelength (nm)
VPD
VTE

Figure 3.9.- Normalized EL intensity of green Ir(ppy)3
phosphorescent OLEDs
400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
1.2

Normalized EL intensity (a.u.)
Wavelength (nm)
VPD
VTE
102

3.2.4. Organic photovoltaic devices with magnesium cathodes

Organic photovoltaic devices using 200 nm thick Mg cathodes were fully made in either
the VPD or VTE. Organics and metals were deposited at a range of 1–4 Å/s and 4–8 Å/s,
respectively. Devices prepared with the VPD used substrate holder temperatures of 60–65ºC for
organics and 20–25ºC for metals and total carrier gas flows of 60 sccm for both organics and
metals. OPVs were initially fabricated on ITO using CuPc (40 nm)/C60 (40 nm) and the most
common material, bathocuproine (BCP, 10 nm) as a buffer layer. While it has been established
that BCP is an effective buffer when aluminum is used as a cathode in the same structure, non-
rectifying current-voltage (IV) curves were obtained for analogous devices with Mg cathodes.
Likewise, devices fabricated without a buffer layer exhibited poor charge extraction
(ηp = 0.16 ±0.05%). Poor contact of the metal cathode with the acceptor layer, Figure 3.10 and
Figure 3.11, was observed with this device structure.

Figure 3.10.- Current-Voltage (IV) curves of a CuPc/C 60/Mg OPV fully
fabricated in the VPD
-1.0 -0.5 0.0 0.5 1.0
-2
0
2

Current Density (mA/cm
2
)
Voltage (V)
J
sc
= 0.986
V
oc
= 0.442
FF= 0.38
= 0.166
103

Conversely, working OPVs were obtained when a 10 nm buffer layer of 3, 4, 9, 10
perylenetetracarbonyl bisbenzimidazole (PTCBI) was employed in the device. Device
performance can be seen on Table 3.4. Current-voltage curves and efficiency-wavelength
characteristics for the devices can be seen in Figure 3.12. OPVs with Mg cathodes fabricated in
the VTE have efficiencies comparable to values obtained for devices using the same CuPc-C60
architecture (ηp = 0.75-1.3%) and an aluminum cathode.
46-49
Devices made by either deposition
method exhibited similar open circuit voltages (Voc = 0.45 ±0.06%) and fill factors
(FF = 0.50 ±0.06%). However, devices made using VPD displayed a lower current
(Jsc = 2.2 ±0.2 mA/cm
2
) than those using the VTE (Jsc = 4.6 ±0.4 mA/cm
2
) thus rendering a lower
power efficiency (ηp = 0.5 ±0.1%).

Figure 3.11.- External quantum efficiency (%) of a CuPc/C60/Mg OPV
fully fabricated in the VPD
400 500 600 700 800
0
5
10
15
20

EQE (%)
Wavelength (nm)
104

The decrease in current density from devices made using VPD can be ascribed to exposure
to air (~ 5 min) during electrode mask placement and cleaning of the shutters. To corroborate this
hypothesis, a device was made in the VTE where the organic films were exposed to air (~ 5 min)
prior to deposition of the electrode. A reference device was concurrently prepared where organic
films were not subject to air during fabrication. Devices exposed to air displayed lower current
Table 3.4.- Organic photovoltaics performance, CuPc/C60/PTCBI/Mg, fully made in the VPD.
Substrate (S), Device (D)
Device
Jsc
(mA/cm
2
)
Voc (V) FF η (%) EQE n
Js (μA/cm
2
) Rsa (Ω/cm
2
)
S3D1 2.23 0.46 0.49 0.51 2.41 3.64 0.01 0.0040
S3D2 2.43 0.47 0.47 0.53 2.41 3.79 0.02 0.0027
S3D3 2.17 0.47 0.47 0.47 2.41 4.06 0.02 0.0018
S3D4 2.01 0.47 0.46 0.43 2.41 4.14 0.02 0.0017
Average 2.21 0.47 0.47 0.49 2.41 3.91 0.02 0.0025

-0.9 -0.6 -0.3 0.0 0.3 0.6
-4
0
4
8
12
400 500 600 700 800 900
0
5
10
15
20

EQE (%)
Wavelength (nm)
VPD
VTE

Current Density (mA/cm
2
)
Voltage (V)
VPD Dark
VPD Light
VTE Dark
VTE Light

Figure 3.12.- Current vs voltage (IV) curves of CuPc/C 60/PTCBI/Mg
OPVs. External quantum efficiency (%) can be seen in the inset
105

density (10%) and a ~ 25% decrease in efficiency, compared to the control device Table 3.5, Figure
3.13 and Figure 3.14.

Table 3.5.- Organic photovoltaics performance, CuPc/C60/PTCBI/Mg, fully made in the VTE.
Organic films of device S2 were exposed to air for ~ 5 minutes while organic films of device S1
were not exposed to air. Substrate (S), Device (D)
Device
Jsc
(mA/cm
2
)
Voc (V) FF η (%) EQE n
Js (μA/cm
2
) Rsa (Ω/cm
2
)
S1D1 4.13 0.44 0.56 1.03 3.05 2.99 0.01 0.00
S1D2 4.21 0.44 0.57 1.06 3.14 2.52 0.00 0.00
Average 4.17 0.44 0.57 1.05 3.10 2.76 0.01 0.00
S2D1 4.09 0.38 0.52 0.81 2.70 2.63 0.01 0.00
S2D2 3.52 0.39 0.54 0.74 2.70 2.52 0.00 0.00
Average 3.80 0.39 0.53 0.77 2.70 2.57 0.01 0.00

-0.9 -0.6 -0.3 0.0 0.3 0.6
-8
-4
0
4
8
S1 Dark
S2 Dark
S1 Light
S2 Light

Current Density (mA/cm
2
)
Voltage (V)

Figure 3.13.- Current-Voltage (IV) curves of CuPc/C 60/PTCBI/Mg OPVs
fully fabricated in the VTE. Organic films of device S2 were exposed to air
for ~ 5 minutes while organic films of device S1 were not exposed to air
106

3.3. Conclusions

In summary, we present a new low vacuum method to deposit thin metal films using a
vapor phase deposition (VPD). Characterization of Mg and Zn films via SEM, AFM, XRD and
four-point probe resistivity indicate thin metal films fabricated in the VPD and VTE show
complete substrate coverage and have comparable roughness/uniformity, crystallinity and
electrical conductivity. By adequately controlling deposition conditions and substrate
temperatures, a VPD was used to deposit Mg cathodes for optoelectronics devices. We
successfully fabricated fluorescent and phosphorescent OLEDs and OPVs with Mg cathodes,
which perform similarly to those prepared by VTE. This new method allows complete fabrication
400 500 600 700 800 900
0
5
10
15
20
S1
S2

EQE (%)
Wavelength (nm)

Figure 3.14.- External quantum efficiency (%) of CuPc/C60/PTCBI/Mg
OPVs fully fabricated in the VTE. Organic films of device S2 were exposed
to air for ~ 5 minutes while organic films of device S1 were not exposed to
air
107

of optoelectronic devices at chamber pressures of 1-10 torr. Further, low pumping times and high
material utilization in the VPD unlocks the opportunity to compete with high-volume high-vacuum
production methods for the manufacturing of optoelectronic devices. Deposition of metals with
higher sublimation temperatures than magnesium and zinc, (e.g., aluminum and silver) in the VPD
should also be feasible. However, to realize Al and Ag electrodes it’s important to consider that
both metals have significantly higher enthalpies of sublimation (330 and 284 KJ/mol, respectively)
than Mg and Zn.
27
Consequently, the need for higher deposition temperatures (  700°C) would
require use of a vessel in the VPD that could sustain higher temperatures (e.g., quartz or stainless
steel). Moreover, additional cooling would be necessary to maintain substrate temperatures at
values low enough (≤ 50°C) to prevent damage to the organic films and thus obtain working
devices.

3.4. Experimental
3.4.1. Characterization of metal films

Profilometer profiles were obtained with a Sloan Dektak IIA stylus profilometer. Analysis
of the Mg thin film profile showed that the 200 nm film fabricated in the VPD was soft, leaving a
scratch mark during measurement, and had a thickness between 160 nm and 180 nm. In addition,
profiles were acquired for 200 nm Mg thin films deposited in the VTE. Metal film morphology,
roughness, crystallinity and resistivity was characterized using a JEOL JSM-7001F scanning
electron microscope (SEM), Digital Instruments Dimension 3100 atomic force microscopy
(AFM), Rigaku Ultima IV powder/thin film diffractometer (XRD) and a Signatone 4-point probe
108

resistivity head connected to a Keithley power source meter model 2400. XRD measurements were
performed at 4 degrees per minute, 0.1 step, 0.3 degrees gamma and a 10mm slit.

3.4.2. OLED and OPV substrate preparation

Patterned ITO substrates with a resistivity of 1:20 ±5 ohms/sq and an ITO thickness of
2000 ± 50Å were used for OLEDs and OPVs made in both the VPD and VTE. OLED substrates
were prepared/cleaned by scrubbing them with Tergitol NP9 (Sigma-Aldrich Co.)/DI solution
followed by a thorough rinse with DI water. The substrates were then rinsed with acetone
(SigmaAldrich Co.) followed by a blow drying step with N2 gas. They were then placed for 10
min in a ultra-violet ozone cleaning system, model T10X10/OES. Substrates were then transferred
to either the VPD or VTE for device fabrication. OPV substrates were prepared/cleaned by
scrubbing them with Tergitol NP9 (SigmaAldrich Co.)/DI solution followed by a thorough rinse
with DI water. The substrates were then rinsed with acetone (Sigma-Aldrich Co.) followed by 2-
Propanol (SigmaAldrich Co.) and were then blow dried with N2 gas. They were then washed with
tetrachloroethylene (J.T. Baker), acetone (Macron Chemical) and ethyl alcohol anhydrous reagent
(J.T. Baker). The wash consists of placing the substrates, in the order mentioned above, inside a
beaker with each solvent for 10 min while heating the solvent to its boiling point. After washing,
the substrates are blow dried with N2 gas and placed in a ultra-violet ozone cleaning system, model
T10X10/OES for 10 min. Substrates were then transferred to either the VPD or VTE for device
fabrication.
109

3.4.3. OLED and OPV device structure

Fluorescent and phosphorescent OLED devices were made in both the VPD and VTE
(reference). The green fluorescent emitter device consists of
N,N′Di[(1naphthyl)N,N′diphenyl]1,1′biphenyl)4,4′diamine (NPD-40 nm), aluminum
tris(8 hydroxyquinoline) [Alq3-40 nm] and Mg (Mg-200 nm). The Green phosphor emitter device
consists of N,N′Di[(1naphthyl)N,N′diphenyl]1,1′biphenyl)4,4′diamine (NPD-40 nm),
factris(2phenylpyridine)iridium (Ir(ppy)3) doped into 4,4′Bis(Ncarbazolyl)1,1′biphenyl
(CBP-30 nm) @ 7% wt, Bathocuproine (BCP-10 nm), aluminum tris(8 hydroxyquinoline)
[Alq3-40nm] and Mg (Mg-200 nm). Organic photovoltaic devices were also made in both the VPD
and VTE. The OPV device structure consists of copper phthalocyanine (CuPC-40 nm), Fullerene
(C60-40 nm), 3,4,9,10 perylenetetracarboxylic bisbenzimidazole (PTCBI-10 nm) and magnesium
(Mg-200 nm). The organic materials NPD, CBP and Alq3 were obtained from Universal Display
Corporation, BCP was purchased from MER corporation, factris(2phenylpyridine)iridium
(Ir(ppy)3) was synthesized according to literature,
50
and fullerene (C60) and magnesium chips (Mg)
were purchased from Sigma-Aldrich Co. Both OLEDs and OPVs device architecture can be seen

Substrate/ITO
NPD (40nm)
CBP-Ir(ppy)
3
(30nm @ 7% wt.)
BCP (10nm)
Alq
3
(40nm)
Mg (200nm)

Substrate/ITO
NPD (40nm)
Alq
3
(40nm)
Mg/Zn (200nm)
Substrate/ITO
CuPc (40nm)
C
60
(40nm)
PTCBI (10nm)
Mg (200nm)

Figure 3.15.- Phosphorescent and fluorescent OLED and OPV device configuration
110

in Figure 3.15. All organic materials were purified at least once via vacuum-train sublimation prior
to use in the VPD or VTE.

3.4.4. VPD device fabrication

A general schematic of the VPD setup can be seen in Figure 3.1, in which the VPD 4”
Pyrex tube/reactor is housed in a Carbolite TVS 12/600/2416CG three zone (1–3) tube furnace.
Preheating of the 4” Pyrex tube (Zone 4) is performed with a 4” x 4” 120 volt, 1100 watts Watlow
mineral insulated band heater. Temperature of the source boats, mineral insulated band heater and
substrate holder was measured and controlled with Omega type K thermocouple probes attached
to Omega CN76000 temperature controllers. Flow of inert gas for each individual source boat,
four in total, was measured using individual MKS mass flow controllers (0–500 sccm) while
pressure was measured with a 10 torr Model 626A Baratron Pressure Transducer and adjusted to
desire with a MKS Model 153D Smart Downstream Throttle Valve. All MKS instruments were
monitored and controlled with a MKS Model 647C flow channel controller box while vacuum was
achieved with a Varian IDP3 Oil Free Dry Scroll Pump. The substrate holder, made out of
aluminum, its position approximately 1” away from zone 1 and it can hold two 1” x 1” patterned
substrates. The substrate holder has two shutters which are utilized to control total thickness
deposited onto each individual substrate. The substrate holder was cooled using N2 gas that was
passed through a copper coil submerged in liquid nitrogen. Substrate holder temperature was
controlled by adjusting the flow/volume of cold N2 gas by means of a needle valve. Material
thickness was measured using a 6 MHz Inficon quartz monitor gold coated crystal sensor attached
to an Inficon XTC/2 thin film deposition controller. Proper calibration of the Inficon crystal sensor
111

via spectroscopic ellipsometry was performed using a J.A. Woollam Co., Inc. VASE variable-
angle ellipsometer with a VB-200 control module and a CVI instruments Digikrom 242
monochromator with a 75 W xenon light source to ensure accurate thickness of films made. The
Inficon crystal sensor was kept at 15 °C at all times using a VWR 1140A chiller.
By introducing the source boat into the hot zone the organic/metal was evaporated into the
inert gas stream (N2) which carried the organic compound to the cooled substrate where the
organic/metal condensed onto the surface of the substrate. Organic and metal compounds were
deposited at a constant pressure of 1 torr, total gas flow rates of 60 sccm and deposition rates of
1–4 Å/s and 4–8 Å/s, respectively. High and low sublimation temperature gradients utilized in the
VPD can be seen in Table 3.6. Because the VPD has only four source boats, during the fabrication
of phosphorescent devices two source boats, Alq3 and Mg, were exchanged for NPD and BCP in
order to complete the device. Further, VPD organic films were exposed to air for approximately 5
minutes during exchange of source boats and metal mask placement. Substrate holder temperature
during deposition of NPD, Alq3 and metals was between 40–45 °C while CuPc, C60 and PTCBI
depositions occurred at a substrate holder temperature of 60–65 °C. Organic/metal sublimation
temperatures during deposition are shown in Table 3.7.

Low sublimation temperature materials High sublimation temperature materials
Zone 1 Pre-Substrate Zone 300°C Zone 1 Pre-Substrate Zone 450°C
Zone 2 Hot Zone 350°C Zone 2 Hot Zone 550°C
Zone 3 Hot Zone 350°C Zone 3 Hot Zone 550°C
Zone 4 Pre-Heating Zone 350°C Zone 4 Pre-Heating Zone 550°C
Table 3.6.- Temperature gradient conditions for low/high sublimation materials in the VPD
112

3.4.5. VTE device fabrication

Reference devices were made in an EvoVac 800 VTE system attached to a glove box and
an Inficon SQS–242 deposition software was used to control deposited material thicknesses using
a 6MHz Inficon quartz monitor gold coated crystal sensor. All films deposited in the VTE were
performed at pressures ≤4 x 10
-4
Pa and with deposition rates ranging between 1-5 Å/s. Because
the VTE is attached to a glove box, organic films were never exposed to ambient air. Proper
calibration of the Inficon crystal sensor via spectroscopic ellipsometry was performed using a J.A.
Woollam Co., Inc. VASE variable-angle ellipsometer with a VB-200 control module and a CVI
instruments Digikrom 242 monochromator with a 75 W xenon light source to ensure accurate
thickness of films made.

Compound Sublimation temperature (°C)
NPD 260 270
Alq3 290 300
CuPc 440 480
C60 500 530
PTCBI 500 530
Zn 430 520
Mg 510 550
Table 3.7.- Sublimation temperature conditions of organics/metals in
the VPD
113

3.4.6. OLED and OPV testing

OLED current-power and current-voltage curves, under applied forward bias of 0-12V,
were measured using a Keithley power source meter model 2400, a Newport multi-function optical
meter model 1835-C, a low power Newport silicon photodiode sensor model 818-UV and a fiber
bundle (used to direct the light into the photodiode). The silicon diode was set to measure
power/photons at an energy of 520 nm which was later corrected, during data processing, to each
individual device electroluminescence average wavelength. Electroluminescence of OLEDs was
collected with a photon technology international QuantaMaster model C-60 fluorimeter at several
voltages, between 3-11 V, to ensure emission characteristics remained constant.
OPV current density (J) as a function of applied voltage (V) characteristics were measured
in air at room temperature, in the dark and under spectral mismatch corrected 100 mW/cm
2
white
light illumination from an AM-1.5G filtered 300 W Xenon arc lamp (Newport Inc.) and a Keithley
power source meter model 2635A. Routine spectral mismatch correction for ASTM G173-03 was
performed using a filtered silicon photodiode, calibrated by the National Renewable Energy
Laboratory (NREL) to reduce measurement errors. Frequency modulated monochromatic light
(250 Hz, 10 nm FWHM) and lock-in detection was used to perform all spectral responsivity and
spectral-mismatch correction measurements.

3.5. Chapter 3 References

1. Scott, J. C. Journal of Vacuum Science Technology 2003, 21, (3), 521-531.
114

2. Matsumura, M.; Akai, T.; Saito, M.; Kimura, T. J. Appl. Phys. 1996, 79, (1), 264-268.
3. Biswas, S.; Shalev, O.; Shtein, M. Annual review of chemical and biomolecular
engineering 2013, 4, 289-317.
4. Bahlawane, N.; Kohse-Hoeinghaus, K.; Premkumar, P. A.; Lenoble, D. Chemical Science
2012, 3, (4), 929-941.
5. Lim, B. S.; Rahtu, A.; Gordon, R. G. Nature Materials 2003, 2, (11), 749-754.
6. Burrows, P. E.; Forrest, S. R.; Sapochak, L. S.; Schwartz, J.; Fenter, P.; Buma, T.; Ban,
V. S.; Forrest, J. L. Journal of Crystal Growth 1995, 156, 91-98.
7. Shtein, M.; Gossenberger, H. F.; Benziger, J. B.; Forrest, S. R. J. Appl. Phys. 2001, 89,
(2), 1470-1476.
8. Shtein, M.; Mapel, J.; Benziger, J. B.; Forrest, S. R. Appl. Phys. Lett. 2002, 81, (2), 268-
270.
9. Baldo, M.; Deutsch, M.; Burrows, P.; Gossenberger, H.; Gerstenberg, M.; Ban, V.;
Forrest, S. Advanced Materials 1998, 10, (18), 1505-1514.
10. Baldo, M. A.; Kozlov, V. G.; Burrows, P. E.; Forrest, S. R.; Ban, V. S.; Koene, B.;
Thompson, M. E. Appl. Phys. Lett. 1997, 71, (21), 3033-3035.
11. Zhou, T. X.; Ngo, T.; Brown, J. J.; Shtein, M.; Forrest, S. R. Appl. Phys. Lett. 2005, 86,
021107-1-021107-3.
115

12. Lindla, F.; Gemmern, P. v.; Keiper, D.; Heuken, M.; Kalisch, H.; Jansen, R. H. Journal
of Photonics for Energy 2011, 1, (1), 011013.
13. Boesing, M.; Lindla, F.; Zimmermann, C.; Gemmern, P. v.; Bertram, D.; Keiper, D.;
Meyer, N.; Heuken, M.; Kalisch, H.; Jansen, R. H. Mater. Res. Soc. Symp. Proc. 2010, 1212,
1212-S03-08.
14. Brast, M.; Lindla, F.; Boesing, M.; Bertram, D.; Keiper, D.; Heuken, M.; Kalisch, H.;
Vescan, A. Mater. Res. Soc. Symp. Proc. 2012, 1402.
15. Bösing, M.; Lindla, F.; Brast, M.; Bertram, D.; Keiper, D.; Heuken, M.; Kalisch, H.;
Vescan, A. Mater. Res. Soc. Symp. Proc. 2012, 1402.
16. Yang, F.; Shtein, M.; Forrest, S. R. Nature Materials 2004, 4, (1), 37-41.
17. Rusu, M.; S. Wiesner, T. M., H. Blei, N. Meyer, M. Heuken, M.Ch. Lux-Steiner, K.
Fostiropoulos. Renewable Energy 2008, 33, (2), 254–258.
18. M. Rusu, J. G., S. Wiesner, N. Meyer, M. Heuken, K. Fostiropoulos, M.Ch. Lux-Steiner.
Thin Solid Films 2008, 516, (20), 7160–7166.
19. Axmann, S.; Brast, M.; Wilck, N.; Windgassen, H.; Heuken, M.; Kalisch, H.; Vescan, A.
Mater. Res. Soc. Symp. Proc. 2012, 1390.
20. Sung Hwan Kim, H. Y. C., Seung Hoon Han, Jin Jang, Sang Mi Cho. Journal of the
Korean Physical Society 2004, 45, 760-762.
21. Rolin, C.; Steudel, S.; Myny, K.; Cheyns, D.; Verlaak, S.; Genoe, J.; Heremans, P. Appl.
Phys. Lett. 2006, 89, 203502.
116

22. Rolin, C.; Vasseur, K.; Schols, S.; Jouk, M.; Duhoux, G.; Müller, R.; Genoe, J.;
Heremans, P. Appl. Phys. Lett. 2008, 93, 033305.
23. Rolin, C.; Vasseur, K.; Steudel, S.; Vicca, P.; Genoe, J.; Heremans, P. Proc. of SPIE
2009, 7417, 741712.
24. Vasseur, K.; Rolin, C.; Vandezande, S.; Temst, K.; Froyen, L.; Heremans, P. J. Phys.
Chem. C 2010, 114, (6), 2730–2737.
25. Rolin, C.; Vasseur, K.; Niesen, B.; Willegems, M.; Robert, M.; Soren, S.; Jan, G.; Paul,
H. Advanced Functional Materials 2012, 22, (23), 5050-5059.
26. Yase, K.; Takahashi, Y.; Ara-kato, N.; Kawazu, A. Japanese Journal of Applied Physics
1995, 34, 636-637.
27. Plieth, W., Electrochemistry for Materials Science. 1 ed.; Elsevier Science: 2008; p 420.
28. CRC Handbook of Chemistry and Physics. 94 ed.; CRC Press: 2013; p 4-126.
29. Adachi, C.; Tokito, S.; Tsutsui, T.; Saito, S. Japanese Journal of Applied Physics 27
(1988) 1988, 27, L713-L715.
30. Stössel, M.; Staudigel, J.; Steuber, F.; Simmerer, J.; Winnacker, A. Applied Physics A
1999, 68, (4), 387-390.
31. Kaye, G. W. C.; Ewen, D. Proceedings of the Royal Society of London. Series A,
Containing Papers of a Mathematical and Physical Character 1913, 89, (607), 58-67.
32. Wood, R. W. Philosophical Magazine Series 6 1915, 30, (176), 300-304.
117

33. Knudsen, M. Annalen der Physik 1916, 355, (12), 472-488.
34. Langmuir, I. Proceedings of the National Academy of Sciences 1917, 3, (3), 141-147.
35. Cockcroft, J. D. Proceedings of the Royal Society of London. Series A, Containing
Papers of a Mathematical and Physical Character 1928, 119, (782), 293-312.
36. Benjamin, P.; Weaver, C. Royal Society of London Proceedings Series A, Mathematical
and Physical Sciences 1961, 261, (1307), 516-531.
37. MDI Jade 9 PDF#03-065-3365.
38. Ming Liu, D. Q., Ming-Chun Zhao, Guangling Song, Andrej Atrens. Scripta Materialia
2008, 58, (5), 421–424.
39. Kwang Seon Shinm, M. Z. B., Nguyen Dang Nam. JOM 2012, 64, (6), 664-670.
40. Jie Gao, A. H., Ming Li, Dali Mao. Applied Surface Science 2009, 255, (11), 5943–5947.
41. O'Brien, D. F.; Burrows, P. E.; Forrest, S. R.; Koene, B. E.; Loy, D. E.; Thompson, M. E.
Advanced Materials 2014, 10, (14), 1108-1112.
42. Yu, M.-X.; Duan, J.-P.; Lin, C.-H.; Cheng, C.-H.; Tao, Y.-T. Chemistry of Materials
2002, 14, (9), 3958–3963.
43. Cai, M.; Xiao, T.; Liu, R.; Chen, Y.; Shinar, R.; Shinar, J. Applied Physics Letters 2011,
99, 153303.
44. Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R. Applied
Physics Letters 1999, 75, (1), 4-6.
118

45. Adachi, C.; Baldo, M. A.; Forrest, S. R.; Thompson, M. E. Applied Physics Letters 2000,
77, (6), 904-906.
46. Sullivan, P.; Heutz, S.; Schultes, S. M.; Jones, T. S. Applied Physics Letters 2004, 84, (7),
1210.
47. Lin, H.; Yu, J.-s.; Huang, J.; Jiang, Y.-d. Optoelectronics Letters 2008, 4, (5), 321-323.
48. Kim, Y. W.; Monroe, M. L.; Seol, J.; Truong, N. T. N.; Cho, S. M.; Anderson, T. J.;
Park, C. Korean Journal of Chemical Engineering 2008, 25, (5), 1036-1039.
49. Trinh, C.; Bakke, J. R.; Brennan, T. P.; Bent, S. F.; Navarro, F.; Bartynski, A.;
Thompson, M. E. Applied Physics Letters 2012, 101, 233903.
50. Tamayo, A. B.; Alleyne, B. D.; Djurovich, P. I.; Lamansky, S.; Tsyba, I.; Ho, N. N.; Bau,
R.; Thompson, M. E. J. Am. Chem. Soc. 2003, 125, (24), 7377–7387.

119

Chapter 4. Fabrication of hybrid organic-inorganic lead based
solar cells with a vapor phase deposition (VPD)

4.1. Introduction

Organometal halide perovskites, i.e., methyl ammonium lead bromide CH3NH3PbBr3
(MAPbBr3) and methyl ammonium lead iodide CH3NH3PbI3 (MAPbI3), were first introduced as
light sensitizers for dye-sensitized solar cells (DSSC) in 2009 i.e., a semiconductor sandwiched
between a photo-sensitized anode and an electrolyte.
1
The perovskite compounds are fabricated
by mixing methylammonium halide (e.g., CH3NH3I or CH3NH3Br) with a lead halide (e.g., PbI2
or PbBr2). After deposition on an anode, i.e., titanium dioxide (TiO2), the perovskite compounds
act as light sensitizers and when they are placed next to a Pt-coated FTO glass (cathode), a solar
cell architecture can be accomplished. To prevent contact thus shorting of the solar cell, a 50μm
thick separator film is placed between the electrodes and the gap filled with an organic electrolyte
solution. The MAPbBr3 based solar cell electrolyte is 0.4 M lithium bromide (LiBr) and 0.04 M
bromide (Br2) dissolved in acetonitrile, while the MAPbI3 based solar cell electrolyte is 0.15 M
lithium iodide (LiI) and 0.075 M iodide (I2), dissolved in methoxyacetonitrile. While efficiencies
of 3.13% and 3.81% were obtained for MAPbBr3 and MAPbI3, respectively, it was not until the
hybrid organometal halide film was used in a solid state heterojunction solar cell, achieving
efficiencies close to 11%, that strong interest sparked in the scientific community
2, 3
thus pushing
the efficiency of hybrid organic-inorganic lead based solar cells to roughly 20% in only a few
years.
4

120

Perovskites are a group of materials that share the crystal structure ABX3 with calcium
titanate (CaTiO3), known as the perovskite structure, Figure 4.1 (a).
5
The predominant three
dimensional perovskite utilized for solar cells have methylammonium as organic cations
(A – CH3NH3
-
), lead for metal cations (B – Pb
2+
) and halides (X – I
-
, Br
-
, Cl
-
or combinations).
However, among MAPbBr3 and MAPbI3, the latter is the perovskite most commonly used to
fabricate solar cells, mainly because its absorbance extends to nearly 800nm, Figure 4.1 (b). A
broader absorbance means the solar cell is capable of collecting more photons that can be turned
into carriers thus producing a more efficient device.

There are two main deposition techniques used to fabricate perovskite films for solar cells,
namely, the one-step and two-step methods. Further, one common stage both techniques have is
annealing of the perovskite film. Annealing is accomplished to increase crystallinity of the films
which in turn improves exciton diffusion and thus device efficiency. Normally, temperatures
ranging between 70°C – 130°C have been used during the annealing step. The one-step method
and also the simplest preparation, relies on simultaneous deposition of both precursors, e.g.,
A (CH
3
NH
3
+
)
B (Pb
2+
)
X (Cl
-
, Br
-
, I
-
)
(a)
400 600 800

(b)
Absorbance (a.u.)
Wavelength (nm)
CH
3
NH
3
PbBr
3
CH
3
NH
3
PbI
3

Figure 4.1.- (a) crystal structure of perovskites and (b) absorbance spectra of
CH3NH3PbBr3 and CH3NH3PbI3
121

methylammonium iodide (MAI) and lead iodide (PbI2). This technique has been used to spin cast
or co-sublime both precursors to deposit perovskite films onto a substrate. However, poor control
over the film morphology and surface coverage has proven problematic when using the one-step
technique, specifically, when spin casting is employed.
6-8
Conversely, the two-step method, in
which consecutive deposition of the two precursors occurs, offers additional control over film
morphology. There are several variations of the two-step method and in the following examples,
the first step refers to the lead halide precursor and the second step refers to the organic halide
compound. Some examples of the two-step technique are, solution dipping of both precursors, spin
casting followed by solution dipping, sublimation followed by solution dipping, sublimation of
both precursors, sublimation followed by spin casting, solution dipping followed by spin casting,
solution dipping followed by sublimation, among others.
9-25
Nevertheless, both deposition
methods have produced solar cells with efficiencies between 10 – 20% and will continue to be
used to understand the science behind the hybrid organic-inorganic perovskite solar cells.

4.2. Results and discussion
4.2.1. Hybrid organic-inorganic perovskite films characterization
4.2.1.1. One-step perovskite films

As mentioned earlier, a one-step perovskite film is fabricated when both materials, i.e.,
MAI and PbI2, are deposited concurrently. Perovskite films with a 330nm thickness were made in
the VPD by fixing the deposition rate of PbI2, i.e., 1.0 – 1.2 Å/s, a sublimation temperature of
363°C ± 3, 20 standard cubic centimeters per minute (sccm), substrate holder temperature
122

of -90°C ± 5 and 0.6torr, while changing the deposition rate of MAI. High sublimation temperature
gradient conditions were used for the fabrication of these films, Table 4.8. Using different
deposition rates of MAI while maintaining the rate fixed for PbI2 helped us find the best ratio to
fabricate fully converted perovskite films. The sublimation temperature of MAI was changed
between 120°C to 135°C (0.5 – 3.5Å/s) to control deposition rates during fabrication of the films.
Not surprisingly, we found that a one to one ratio of PbI2 and MAI produced the best films i.e.,
deposition rates of 1.0 – 1.5Å/s (equal to sublimation temperatures of 125°C – 130°C).
Conversely, below or above those temperatures excess of MAI or PbI2 was observed and verified
by absorbance and XRD measurements of the films, Figure 4.2. Interestingly, a peak at ~700nm
and significant more scattering is observed in the absorbance of the perovskite film when excess
of MAI is codeposited with PbI2, i.e., MAI at 135°C and 3.5Å/s. This strange peak is clearly
associated with the excess of MAI in the perovskite film and not only does it provoke greater
scattering, it also enhances the absorption of the non-annealed film. Fatefully, the strange peak
appears to have a shoulder similar to the other films at ~ 750nm, thus indicating that excess of
MAI not only produces greater scattering but it also causes a stronger absorbance of the vibrational
transitions of the film, causing this strange behavior. Perovskite films with a 160nm thickness were
also made in the VTE and characterized. The deposition rates used to obtain fully converted films
in the VTE were 0.4Å/s and 1.2Å/s for PbI2 and MAI, respectively. A one to three ratio of PbI2 to
MAI, respectively, was necessary to compensate for the low sublimation temperature of MAI
which prevented the organic material from efficiently adhering/condensing onto the substrate,
contrariwise to the VPD, where the adhesion/condensation of the organic compound is unity.
Analysis of the films via absorbance and XRD confirmed perovskite films were accomplished,
Figure 4.3. Both methods, VPD and VTE, produced fully converted perovskite films as seen from
123

the prominent diffraction peaks at approximately 17° and 33° 2θ, corresponding to the (110) and
(220) planes of MAPbI3. Absorbance of the films close to 800nm further confirms MAPbI3 films
were prepared. Further, the films that underwent annealing were annealed inside a glove box in
the presence of an inert gas, i.e., N2 at 90°C for one hour.

400 600 800
0.0
0.5
1.0
1.5
2.0
2.5
CH
3
NH
3
PbI
3
Films on ITO
(a)
Not annealed
Annealed

Absorbance
Wavelength (nm)
10 20 30 40 50 60
(b)

2 Theta (degrees)
Not Annealed
Annelaled
CH
3
NH
3
PbI
3
Films on ITO
Intensity (a.u.)

Figure 4.3.- Annealed and non-annealed films one-step 160nm perovskite films on ITO made
in the VTE, (a) absorbance spectra and (b) XRD patterns
400 600 800 1000
0.0
0.4
0.8
1.2
1.6
2.0
2.4
2.8
(a)
CH
3
NH
3
PbI
3
Films on Glass
MAI @ 135C
MAI @ 130C
MAI @ 125C
MAI @ 120C

Absorbance
Wavelength (nm)
10 20 30 40 50 60
(b)

CH
3
NH
3
PbI
3
Films on glass
Intensity (a.u.)
2 Theta (degrees)
MAI @ 135C
MAI @ 130C
MAI @ 125C
MAI @ 120C

Figure 4.2.- Non-annealed one-step 330nm perovskite films with different CH3NH3I
deposition rates on glass made in the VPD, (a) absorbance spectra and (b) XRD patterns
124

Further, scanning electron microscope (SEM), x-ray photoelectron spectroscopy (XPS),
atomic force microscopy (AFM) and inductively coupled plasma optical emission spectrometry
(ICP-OES) were also be used to characterize the above films. SEM images show full coverage
and crystal like growth of the MAPbI3 films, Figure 4.4. These images are analogous to SEM
images reported in literature.
9-11, 14, 17-20, 22, 24-29
XPS core level spectra of the MAPbI3 films are
shown on Figure 4.5 and are similar to core level peaks previously reported in literature.
30
High
binding energy (BE) peaks for Pb 4f5/2 and 4f7/2 can be observed at approximately 142.5 and 137.5,
respectively. Peaks for I 3d5/2 and 3d7/2 are shown at approximately 630 and 618, respectively.
These high BE peaks suggest strong interfacial charge transfer at the MAI/PbI2 contact surface.
Further, no peaks are detected for Pb0, usually located 4-5 eV below both Pb 4f5/2 and 4f7/2 due to
a weaker binding energy of the reduced metal. Three dimensional images and roughness of
15nm ± 3 and 8nm ± 4 were obtained for VPD and VTE, respectively, via AFM analysis, Figure
4.6. Lastly, inductively coupled plasma optical emission spectrometry (ICP-OES) was used to
measure lead metal traces of perovskite films deposited on glass. Emission from the metal, i.e.,
lead (Pb), was measured following appropriate digestion of the perovskite films in a solution of
5% nitric acid in deionized water. The following wavelengths, 220.353nm, 216.999nm, 261.418,
and 182.205nm were used to measure elemental lead via ICP-OES, shown in Table 4.1. A linear
fit from the emission intensity versus the concentration of the standards (5.97, 14.93 and
29.85 ppm), allowed us to calculate the concentration of elemental lead present in the digested
perovskite films, Figure 4.7. From the XPS analysis, we know the perovskite films have no
presence of elemental lead thus the measured lead is specifically from the lead iodide deposited
on the substrate. Weighing the substrates with and without lead iodide, Table 4.2, the measured
125

ICP-OES concentration of elemental lead is in agreement with the measured concentration
calculated by weight and thickness.

Figure 4.4.- SEM images of a 330nm MAPbI3 film on glass made in the VPD, (a) magnified
x 400 and (b) magnified x 2k
146 144 142 140 138 136
(a)
Pb 4f
4f
7/2
4f
5/2

Intensity (a.u.)
Binding Energy (eV)
640 635 630 625 620 615
(b)
Intensity (a.u.)
Binding Energy (eV)
I 3d
3d
5/2
3d
7/2

Figure 4.5.- High resolution XPS core level spectra of a 330nm MAPbI3 film on ITO made in
the VPD, (a) Pb 4f spectra from PbI2 and (b) I 3d spectra from the PbI2 & MAI
126

(b)

Figure 4.6.- 3D AFM images of a 330nm MAPbI3 film on glass, (a) VPD and (b) VTE
(a)
(b)
0 5.97 14.92 29.85
0
1000
2000
3000
4000
(a)
Target wavelength 220.355nm
x = (y - 120.83) / 148.1

Intensity
Concentration (ppm)
0 5.97 14.92 29.85
0
200
400
600
(b)
Target wavelength 216.999nm
x = (y - 24.3) / 20.67

Intensity
Concentration (ppm)

Figure 4.7.- Inductively coupled plasma optical emission spectrometry (ICP-OES) lead metal
traces concentration study of MAPbI3 films made in the VPD and VTE. (a) Standards target
wavelength of 220.335nm and (b) Standards target wavelength of 216.999nm
(a) (b)
127

4.2.1.2. Two-step perovskite films

A two-step perovskite film is fabricated when both materials, i.e., MAI and PbI2, are
deposited separately. As mentioned in the introduction, there are several approaches to fabricate a
two-step perovskite film, however, only two two-step methods were used for the fabrication of
films. The first technique involved deposition of PbI 2 via sublimation, for both VPD and VTE,
followed by dipping, for three minutes, in a solution of MAI (10mg/mL) dissolved in 2-propanol.
The second technique involved sequential sublimation of PbI2 and MAI in the VPD. Neat films of
PbI2 were fabricated in the VPD at 1 Å/s, 20sccm, sublimation temperature of 363°C ± 3, substrate
holder temperature of -90°C ± 5 and a pressure of 0.6torr. Conversely, VTE films were fabricated
Sample
Target wavelengths
220.353nm
(Intensity and ppm)
216.999nm
(Intensity and ppm)
261.418nm
(Intensity and ppm)
182.205nm
(Intensity and ppm)
Blank
2
(-1.21 ppm)
2
(-1.05 ppm)
0
(-1.16 ppm)
1
(-1.07 ppm)
Standard #1
5.97 ppm
843
(5.75 ppm)
141
(5.63 ppm)
146
(5.55 ppm)
222
(5.69 ppm)
Standard #2
14.92 ppm
2285
(17.69 ppm)
388
(17.59 ppm)
413
(17.92 ppm)
609
(17.51 ppm)
Standard #3
29.85 ppm
3593
(28.58 ppm)
615
(28.58 ppm)
641
(28.43 ppm)
972
(28.61 ppm)
One-step VPD
3031
(23.86 ppm)
523
(24.16 ppm)
532
(23.41 ppm)
829
(24.24 ppm)
Two-step VPD
with dipping
2721
(21.29 ppm)
473
(21.72 ppm)
466
(20.34 ppm)
748
(21.75 ppm)
Two-step VPD
2566
(20.07 ppm)
450
(20.59 ppm)
448
(19.51 ppm)
708
(20.54 ppm)
One-step VTE
3579
(28.39 ppm)
627
(29.19 ppm)
619
(27.44 ppm)
991
(29.20 ppm)
Table 4.1.- ICP-OES intensity emission of various wavelengths targeting traces of lead
metal. 240nm MAPbI3 films made in the VPD and VTE were digested in a 5% nitric acid
solution in DI water
128

with deposition rates of 1 Å/s and a base pressure of 3.0 x 10
-6
torr. High sublimation temperature
gradient conditions were used for the fabrication of the VPD films, Table 4.8. Images of the neat
PbI2 films can be seen in Figure 4.8. A yellow/green film was obtained for both deposition systems.
Both methods displayed full coverage of the substrate. One film from each method, i.e., VPD#1
and VTE#1, was annealed before dipping in the MAI solution. Submerging the neat PbI2 films into
the MAI solution turned the films into a dark brown color, Figure 4.9. However, it’s clear the VPD
films are much darker than the VTE films. Analysis of the films via absorbance and XRD indicates
full conversion of the VPD films to MAPbI3 while the VTE films show little to no conversion,
clearly shown on the XRD diffraction peaks, Figure 4.10. Similar to the one-step method,
prominent diffraction peaks at approximately 17° and 33° 2θ, consistent with the (110) and (220)
planes of MAPbI3, can be seen for the VPD films. A strong absorbance shift can also be observed
after the neat PbI2 films are exposed to MAI, red shifting the absorbance to 800nm. However, it
appears only the surface of the VTE films was converted to MAPbI3, leaving mostly neat PbI2 on
the substrate. A diffraction peak at approximately 15° 2θ, corresponding to the (001) plane of neat
PbI2 evidently confirms this. Longer dipping times, e.g., 1 – 24h, can potentially provide enough
time to allow MAI intercalation thus forming the perovskite films, however, that experiment was
not performed. Denser and more crystalline films are believed to be responsible for the poor
intercalation of MAI during the dipping step. Additionally, the techniques used to characterize
one-step films were also used for the two-step films, i.e., SEM, XPS, AFM and ICP-OES, and the
results are analogous as those shown for the one-step films. Further, the films that underwent
annealing were annealed inside a glove box in the presence of an inert gas, i.e., N2 at 90°C for one
hour.
129

400 600 800 1000
0
1
2
3
(a)
CH
3
NH
3
PbI
3
Films on ITO
VTE
VTE Annealed
VPD
VPD Annealed
PbI2
Absorbance
Wavelength (nm)
10 20 30 40 50 60
(b)
VPD 3min MAI-Annealed
VPD Annealed-3min MAI-Annealed
VTE 3min MAI-Annealed
VTE Annealed-3min MAI-Annealed

2 Theta (degrees)
CH
3
NH
3
PbI
3
Films on ITO
Intensity (a.u.)

Figure 4.10.- Two-step 330nm perovskite films made in the VPD and VTE, (a)
absorbance spectra and (b) XRD patterns

VTE#1 VTE#2 VPD#1 VPD#2
Figure 4.8.- 330nm PbI2 films made in the VPD and VTE. Substrate #1 of both
VPD and VTE were annealed for 1h at 90°C before dipping into a CH3NH3I
solution (10mg/mL) in 2-propanol

VTE#1 VTE#2 VPD#1 VPD#2
Figure 4.9.- 330nm perovskite films made in the VPD and VTE. PbI2 films
were dipped into a CH3NH3I solution (10mg/mL) in 2-propanol for 3min, then
annealed for 1h at 90°C
130

The second two-step method films were fabricated using identical conditions as those used
for the neat PbI2 films previously described and shown in Figure 4.8, thus attaining analogous
films. Exposing the neat films to gaseous MAI in the VPD allowed intercalation of MAI thus
making the perovskite films, Figure 4.11. Low sublimation temperature gradient conditions were
used during deposition of MAI, Table 4.8. It can clearly be seen a shift in absorbance, from 500nm
to 800nm, after the neat PbI2 film is exposed to MAI as well as a shift in the diffraction peaks from
15° 2θ to 17° and 33° 2θ, corresponding to the (110) and (220) planes of MAPbI3. Additional,
SEM, XPS, AFM and ICP-OES obtained for these films were equivalent to the one-step films.
Calibration of all the films made in both VPD and VTE, was accomplished using a J.A. Woollam
Co., Inc. VASE variable-angle ellipsometer and an Ambios XP-2 stylus profilometer, Figure 4.12.
Lastly, relative density measurements were performed on both one-step and two-step
deposition techniques, Table 4.2. Each substrate was weighed before and after deposition of the
films. The cell volume of the tetragonal MAPbI3, from XRD diffraction patterns, was used to
calculate the density of a single crystal and found to be 4.15g/cm.
3, 31, 32
A relative ratio, i.e., the
calculated XRD density of the single crystal divided by the density of the films, was analyzed to
assess how compact/dense the VPD and VTE films were. A relative density of unity would indicate
a film with analogous density to a single crystal while a relative density smaller than unity would
indicate a less dense film. The best relative densities were the one-step VTE and the two-step VPD,
realizing relative densities close to unity. Conversely, VPD one-step and two-step dipping in a
solution of MAI exhibited less dense films, resulting in relative densities of 0.54 and 0.8,
respectively.
131

400 600 800 1000
0
1
2
3
CH
3
NH
3
PbI
3
PbI
2
(a)
Absorbance
Wavelength (nm)
10 20 30 40 50 60
CH
3
NH
3
PbI
3
PbI
2
(b)

Intensity (a.u.)
2 Theta (degrees)

Figure 4.11.- Two-step 330nm perovskite films on ITO fully made in the VPD, (a)
absorbance spectra and (b) XRD patterns
0 500 1000 1500 2000
-400
-300
-200
-100
0
100
(a)

Height (nm)
Tracing Length (nm)
0 500 1000 1500 2000
-300
-200
-100
0
(b)

Height (nm)
Tracing Length (nm)

Figure 4.12.- Profilometer profiles of two-step 330nm perovskite films on ITO made in,
(a) VPD and (b) VTE
Film
Bare
Substrate
(g)
Substrate
+
Perovskite
(g)
Perovskite
weight (g)
Full
area
(inch^2)
Clip
Area
(inch^2)
MAPI
coverage
area
(inch^2)
Thickness
(nm)
Volume
(cm^3)
Density
(g/cm^3)
Relative
Density
VPD Codep PbI2
& MAI
1.27 1.27 2.9E-04 0.85 0.02 0.83 240 1.29E-04 2.26 0.54
VPD PbI2 &
10mg/mL MAI
1.32 1.32 4.2E-04 0.84 0.03 0.82 240 1.26E-04 3.33 0.80
VPD PbI2 & MAI
1.29 1.29 7.5E-04 1.20 0.02 1.18 240 1.83E-04 4.11 0.99
VTE Codep PbI2
& MAI
1.40 1.40 8.9E-04 1.40 0.01 1.39 240 2.16E-04 4.12 0.99
Bare substrate
1.16 1.17 2.0E-04 0.88 0.00 0.88 0 0.00E+00
Table 4.2.- Relative density calculations of 240nm perovskite films on glass made in the VPD
and VTE
132

4.2.2. Hybrid organic-inorganic perovskite solar cells fabricated in the VPD

Hybrid organic-inorganic perovskite devices were fabricated with several different
architectures. Methylammonium lead iodide (MAPbI3 – 160 – 330nm), bathocuproine
(BCP – 10 nm), buckminsterfullerene (C60 – 40nm), molybdenum trioxide (MoO3 – 140nm),
N,N′Di[(1naphthyl)N,N′diphenyl]1,1′biphenyl)4,4′diamine (NPD – 140nm), silver
(Ag – 100nm), and aluminum (Al – 100nm) were the materials used in the fabrication of the solar
cells. The perovskite films were made in the VPD while the rest of the layers were deposited in
the VTE. The device structures were ITO/MAPbI3(330nm)/C60(40nm)/BCP(10nm)/cathode (S1)
and ITO/MAPbI3(330nm)/NPD-MoO3(140nm @ 50% doping concentration)/cathode (S2). It’s
worth nothing that while MAPbI3 can move holes and electrons, in device S1 it transfers electrons
to C60 while in device S2 it transfers holes to NPD-MoO3. Generally, device S2 is called inverted
while device S1 is called conventional. The goal of having both inverted and conventional devices,
was to examine which architecture was better extracting carriers from the perovskite film. The
concentration of NPD doped with MoO3 used in device S2 was selected because it has proven to
be a successful hole transport layer when used in organic light emitting diodes (OLEDs).
33

The perovskite films were fabricated using the two-step method, i.e., both compounds were
sublimed in the VPD sequentially at 1 Å/s, 20sccm, sublimation temperature of 363°C ± 3,
substrate holder temperature of -90°C ± 5 and a pressure of 0.6torr. Solar cell performance
parameters can be seen in Table 4.3. Device IV curves and EQE spectra can be seen in Figure 4.13.
Efficiencies of 3.3 ± 0.3% and 4.1 ± 0.4% were obtained for architectures S1 and S2, respectively.
A larger Voc of 0.8 ± 0.1V was obtained for device S2. The result was expected if we consider that
the Voc has a correlation dependence, Figure 4.14, on the energy difference between the highest
133

occupied molecular orbital (HOMO) energy of the donor (NPD-MoO3) and the lowest occupied
molecular orbital (LUMO) energy of the acceptor (MAPbI3), ΔEDA/2.
34

Another important correlation while testing solar cells is to achieve a comparable response
of the EQE and the Jsc. Commonly, differences of 5% – 20% between EQE and Jsc are obtained,
as is the case of device S1. However, that is not the case for device S2 where the EQE is markedly
smaller than the Jsc, 1.85 vs 12.4, respectively. Characterization of the films, via XRD, ICP-OES,
Uv-Vis, SEM, AFM, etc. as shown above, demonstrate fully converted perovskites were
fabricated, thus this strange behavior could be associated with the ITO/perovskite or
perovskite/donor or acceptor interface. Regrettably, it is not possible to evaluate whether
degradation exists at one or both interfaces thus preventing us from understanding and explaining
this strange behavior. Further, while perovskite devices have proven to achieve efficiencies above
10%, as discussed in the introduction, we discovered the devices we fabricated, regardless of the
architecture used, were unreliable and highly unreproducible, i.e., one in approximately ten solar
cells would rectify correctly while the rest exhibited poor-rectification. Again, it is possible this
abnormal behavior is related to the weak EQE values, however, a clear understanding of why it is
taking place has not been established.
Device
J sc
(mA/cm
2
)
V oc
(V)
FF η (%) EQE n
J s
(μA/cm
2
)
R sa
(Ω/cm
2
)
S1
9.54 0.67 0.56 3.6 7.3 3.8 3.75E-5 0.03642
S2
12.4 0.83 0.42 4.3 1.85 9.6 0.017 -0.08824
Table 4.3.- 330nm MAPbI3 device performance made in the VPD, (S1)
ITO/MAPbI3/C60/BCP/Al and (S2) ITO/MAPbI3/NPD-MoO3/Ag
134

Conversely, allowing the perovskite films to have excess of PbI2 and/or MAI affected
adversely the efficiency of the devices. A series of devices were fabricated with the following
architecture, ITO/MAPbI3(330nm)/C60(40nm)/BCP(10nm)/Al, allowing some of the films to have
-6
-5
-4
-3
-2
-6
-5
-4
-3
-2
MAPbI
3
Energy (eV)
Al
ITO
NPD
C
60
Energy (eV)
HOMO LUMO Gap Diagram

Figure 4.14.- HOMO-LUMO gap diagram for MAPbI3 with C60
(acceptor) and NPD (donor)
-0.5 0.0 0.5 1.0
-20
0
20
(a)
Dark
S1
S2

Current Density (mA/cm
2
)
Voltage (V)
400 600 800
0
10
20
30
40
S1
S2
(b)

EQE (%)
Wavelength (nm)

Figure 4.13.- 330nm MAPbI3 devices made in the VPD, (a) current vs voltage (IV)
curves and (b) external quantum efficiency, S1 circle – ITO/MAPbI3/C60/BCP/Al and
S2 square – ITO/MAPbI3/NPD-MoO3/Ag
135

excess of either PbI2 or MAI. The perovskite films were made in the VPD, two-step method and
under the same conditions mentioned above, while the rest of the layers were deposited in the
VTE. The two parameters that were impacted greatly were current (Jsc) and fill factor (FF), Table
4.4. Efficiencies of 0.56 ± 0.1% and 0.15 ± 0.04% were obtained when the films contained small
excess of PbI2 (S3) or MAI (S4), respectively.

Greater excess of PbI2 (S6) led to a larger current but with the caveat that PbI2 caused traps
and thus poor rectifying IV curves and fill factors, as illustrated in Figure 4.15. Device S3 and S6
exhibited a weak diffraction peak at approximately 15° 2θ, corresponding to the (001) plane of
PbI2 thus supporting excess of PbI2 is present in those films. Lastly, having a fully converted film
(S5), produced efficiencies of 1.3 ± 0.3% in this analysis. The Voc was not influenced by the
presence of PbI2 or MAI and it remained between 0.5 – 0.6V for all four device architectures.
Consequently, it’s clear that having a fully converted perovskite film is critical to achieve diode
like device performance whereas, small excess of either organic/inorganic material is detrimental
to device efficiency by directly affecting Jsc and/or FF.
Device
J sc
(mA/cm
2
)
V oc
(V)
FF η (%) EQE n
J s
(μA/cm
2
)
R sa
(Ω/cm
2
)
S3
2.1 0.60 0.44 0.56 7.7 3.5 2.98E-5 0.00417
S4
1.1 0.51 0.28 0.15 11.3 4.4 1.05E-4 0.02162
S5
3.7 0.59 0.62 1.35 8.53 2.3 4.13E-6 0.00382
S6
7.1 0.58 0.23 0.95 7.41 2.5 3.83E-6 0.05018
Table 4.4.- Device performance parameters of 330nm MAPbI3 devices made in the
VPD, ITO/MAPbI3/C60/BCP/Al, S3 – small excess of PbI2, S4 – excess of MAI,
S5 – ideal film, and S6 – excess of PbI2
136

Further, similar behavior where a significant difference between the measured EQE and Jsc
was also observed on devices S3 – S5, whereas device S6 is a great example of an almost perfect
EQE to Jsc match. However, conversely to device S2, in this instance the Jsc of devices S3 – S5 is
markedly smaller than the EQE. Again, it is unclear why there are discrepancies between the
measured Jsc and EQE, but it appears a possible answer to this conundrum might be in the
perovskite interfaces. Even though the XRD diffraction patterns show fully converted perovskite
films, as shown in Figure 4.16 where diffraction peaks at 17° and 33° 2θ, corresponding to the
(110) and (220) planes of MAPbI3 are clearly shown, it is also possible the perovskites films are
not entirely crystalline, i.e., a blend of amorphous and crystalline phases exists in the films, thus
potentially producing the weaker Jsc and/or EQE response and poor rectifying IV curves observed
on the devices. The latter is less probable since amorphous materials tend to scatter X-rays in many
directions, thus leading to bumps on the collected spectra, whereas the periodicity of crystalline
-0.5 0.0 0.5
-8
-4
0
4
8
(a)
Dark
S3
S4
S5
S6

Current Density (mA/cm
2
)
Voltage (V)
400 600 800
0
10
20
30
40
50
60
S3
S4
S5
S6
(b)

EQE (%)
Wavelength (nm)

Figure 4.15.- 330nm MAPbI3 device made in the VPD, ITO/MAPbI3/C60/BCP/Al, (a)
current vs voltage (IV) curves and (b) external quantum efficiency, S3 circle – small
excess of PbI2, S4 square – excess of MAI, S5 triangle – ideal film, and
S6 diamond – excess of PbI2
137

materials scatters X-rays only in certain directions, consequently, leading to sharp peaks that are
similar to those obtained in the collected XRD spectra.

Additionally, annealed and non-annealed perovskite films were fabricated on ITO and
FTO/TiO2. The goal was to compare ITO and FTO/TiO2 as electrodes and also annealed versus
non-annealed MAPbI3 films. The device structures were
ITO/MAPbI3(160nm)/C60(40nm)/BCP(10nm)/cathode (S7 – S8) and
FTO/TiO2/MAPbI3(160nm)/NPD-MoO3(140nm @ 50% doping concentration)/cathode
(S9 – S10). Solar cell performance parameters can be seen in Table 4.5. Device IV curves and
EQE spectra can be seen in Figure 4.17. We found that annealing had a great impact on device
10 20 30 40 50 60
S3
S4
S5
S6

CH
3
NH
3
PbI
3
Films on ITO
Intensity (a.u.)
2 Theta (degrees)

Figure 4.16.- XRD patterns of 330nm MAPbI3 films made in the VPD,
ITO/MAPbI3/C60/BCP/Al, S3 – small excess of PbI2, S4 – excess of MAI,
S5 – ideal film, and S6 – excess of PbI2

138

performance, specifically on Jsc for devices with ITO electrodes. The Voc remained constant for
ITO devices at 0.75 ± 0.02V while the current (Jsc) fluctuated between 9.2 ± 0.6mA/cm
2
and
2.1 ± 0.4mA/cm
2
for annealed and non-annealed films, respectively. Conversely, smaller current
differences were observed on devices where FTO/TiO2 was used as an electrode. The average Voc
for these devices was 0.65V while the current (Jsc) fluctuated between 16.1 ± 2.7mA/cm
2
and
11.8 ± 2.1mA/cm
2
for annealed and non-annealed films, respectively.

The results suggest that a much rougher TiO2 surface promotes crystallization of the
perovskite film while a flatter electrode, i.e., ITO, does not affect crystallization of the film.
Further, the inherent roughness of TiO2 is what required the use of a thick 140nm NPD doped with
MoO3, to prevent shorting of the device when the electrode was deposited. EQE spectra for the
ITO devices behaved normally while both FTO/TiO2 devices had a common EQE occurrence, the
spectra fluctuated considerably and presented sharp spikes. Further, while the IV curves for
devices S7 – S10 do not have an ideal rectifying performance, a significant amount of current is
present in these devices thus suggesting the poor rectification might be associated to the weak EQE
and/or Jsc response and poor rectifying IV curves observed on devices S3 – S5, previously
discussed.
Device
J sc
(mA/cm
2
)
V oc
(V)
FF η (%) EQE n
J s
(μA/cm
2
)
R sa
(Ω/cm
2
)
S7
9.84 0.74 0.35 2.60 11.12 2.28 5.04E-07
0.02
S8
2.46 0.76 0.20 0.38 9.68 6.27 1.77E-04 -0.83
S9
18.51 0.62 0.22 2.58 6.89 3.92 1.81E-03 0.02
S10
13.65 0.71 0.23 2.28 10.31 5.18 3.33E-03 0.01
Table 4.5.- 160nm MAPbI3 device performance made in the VPD, S7 – S8
ITO/MAPbI3/C60/BCP/Al and S9 – S10 FTO/TiO2/MAPbI3/NPD-MoO3/Al, S7 and S9 are
annealed and S8 and S10 are not annealed
139

Lastly, as mentioned earlier, regardless of the architecture used, the devices fabricated in
the VPD were unreliable and highly unreproducible, displaying poor-rectifying IV curves. Several
device architectures were used to help elucidate this poor-rectifying behavior, from using different
hole transport materials, i.e., MoO3, copper phthalocyanine (CuPc), NPD, and 2,2',7,7'-Tetrakis-
(N,N-di-4-methoxyphenylamino)-9,9'-spirobifluorene (Spiro-OMeTAD), to spin casting PbI2 on
ITO and TiO2. PbI2 was dissolved in N,N-dimethylformamide (DMF) or dimethylsulfoxide
(DMSO). Spin-cast of PbI2 and Spiro-OMeTAD was completed using several recipes, from
varying the revolutions per minute (rpm), i.e., 3K and 4K, using different loading times, i.e., 0s
and 30s, to changing solvent temperatures, i.e., 25°C and 70°C, as reported in literature.
9, 11-17, 19,
21-26, 29, 35, 36
Regardless of the device architecture, ITO/MAPbI3/C60/BCP/cathode,
ITO/MAPbI3/NPD-MoO3/cathode, ITO/MAPbI3/Spiro-OMeTAD/cathode, etc., poor-rectifying
curves and/or small currents were obtained, Figure 4.18 (a). Conversely, XRD patterns displayed
fully converted perovskite films, Figure 4.18 (b). Further, during the fabrication of all of the above
0.0 0.5 1.0 1.5
-20
0
20
(a)
S7 Dark
S7
S8
S9
S10

Current Density (mA/cm
2
)
Voltage (V)
400 500 600 700 800 900
0
20
40
60
80
100
S7
S8
S9
S10
(b)

EQE (%)
Wavelength (nm)

Figure 4.17.- 160nm MAPbI3 device made in the VPD, S7 – S8
ITO/MAPbI3/C60/BCP/Al and S9 – S10 FTO/TiO2/MAPbI3/NPD-MoO3/Al (a)
current vs voltage (IV) curves and (b) external quantum efficiency, S7 and S9 are
annealed and S8 and S10 are not annealed
140

devices (S1 – S10), the perovskite films were exposed to ambient air for a few minutes in order to
transfer to the VTE to deposit organic donors and acceptors and metals. Moreover, exposure to
ambient air of organic films fabricated in the VPD for solar cells has proven to affect the efficiency
of such devices,
37
thus allowing the probability that ambient air is adversely affecting the
perovskite films, consequently causing poor rectifying IV curves and discrepancies between the
EQE and Jsc.

4.2.3. Hybrid organic-inorganic perovskite solar cells fabricated in the VTE

One-step hybrid organic-inorganic perovskite solar cells were also fabricated in the VTE.
The device architecture (S11 – S14) was ITO/MAPbI3(160 – 330nm)/C60(40nm)/BCP(10nm)/Al.
The deposition conditions used to obtain fully converted films in the VTE were deposition rates
of 0.4Å/s and 1.2Å/s for PbI2 and MAI, respectively, and a base pressure of 3.0 x 10
-6
torr. Solar
-1.0 -0.5 0.0 0.5 1.0 1.5
-1.0
-0.5
0.0
0.5
1.0
(a)
Spiro-OMeTAD
C60-BCP
Current Density (mA/cm
2
)
Voltage (V)
10 20 30 40 50 60
(b) 2 Theta (degrees)

2 Theta (degrees)
Spiro-OMeTAD
C
60
-BCP
Intensity (a.u.)

Figure 4.18.- 330nm MAPbI3 device made in the VPD, circle
ITO/MAPbI3/C60/BCP/Ag and square ITO/MAPbI3/Spiro-OMeTAD/Ag (a) current vs
voltage (IV) curves and (b) XRD patterns
141

cell performance parameters for 160nm MAPbI3 films can be seen in Table 4.6. Device IV curves
and EQE spectra can be seen in Figure 4.19. Obtained average values for efficiency, Voc and FF
are 4.7 ± 0.7%, 0.6 ± 0.03V and 0.55 ± 0.04, respectively. Conversely, even though the VTE
perovskite films were also exposed to the same conditions as the VPD perovskite films, e.g.,
ambient air, acceptable rectifying IV curves and comparable EQE and Jsc values were obtained for
devices S11 – S14. Evidently, these devices were not affected like devices S2 – S10 hence
motivating the question, why not? As shown in the one-step and two-step methods discussion,
characterization of VPD and VTE perovskite films show both methods produce similar films.
HOwever, dipping neat films of PbI2 made in the VTE into a solution of MAI demonstrated VTE
films are much denser and possibly more crystalline than films fabricated in the VPD, thus offering
a possible explanation of why solar cells made in the VPD are inconsistent and have poor
reproducibility. A much denser and crystalline film can inhibit oxygen and water diffusion through
the film therefore preventing film damage, poor IV curves and discrepancies between measured
EQE and Jsc values.

Device
Jsc
(mA/cm
2
)
Voc
(V)
FF η (%) EQE n
Js
(μA/cm
2
)
Rsa
(Ω/cm
2
)
S11
14.04 0.63 0.63 5.59 12.16 2.42 2.29E-05 4.50E-03
S12
13.74 0.65 0.58 5.16 12.16 2.81 2.29E-05 3.17E-03
S13
13.26 0.63 0.50 4.20 12.16 3.06 3.01E-04 4.92E-03
S14
13.23 0.61 0.50 4.05 12.16 2.95 2.02E-04 5.01E-03
Average
13.57 0.63 0.55 4.75 12.16 2.81 1.37E-04 4.40E-03
Table 4.6.- 160nm MAPbI3 device performance made in the VTE, S11 – S14
ITO/MAPbI3/C60/BCP/Al
142

Poor EQE response between 600 – 800nm, due to the film thickness of MAPbI3, can be
observed on devices S11 – S14. Therefore, films with 330nm MAPbI3 thicknesses were also
fabricated (S15 – S18). Solar cell performance parameters for the 330nm MAPbI3 films can be
seen in Table 4.7. Device IV curves and EQE spectra can be seen in Figure 4.20. Obtained average
values for efficiency, Voc and FF are 4.6 ± 0.7%, 0.6 ± 0.03V and 0.45 ± 0.05%, respectively. A
current increase was achieved, from 13.5 mA/cm
2
to 16.7 mA/cm
2
, for the 160nm and 330nm
films, respectively. Further, larger EQE values were obtained for the 160nm and 330nm films,
12.2 mA/cm
2
and 15.3 mA/cm
2
, respectively. Regrettably, a decrease in FF also complemented
these devices. Consequently, devices S11 – S14 and S15 – S18 displayed overall analogous
efficiencies. Again, acceptable rectifying IV curves and comparable EQE and Jsc values were
obtained for devices S15 – S18, potentially supporting that having a much denser and crystalline
film does prevent oxygen and water diffusion through the film, therefore avoiding film damage,
poor IV curves and discrepancies between measured EQE and Jsc values.
-0.5 0.0 0.5 1.0
-20
0
20
(a)
S11 Dark
S11
S12
S13
S14

Current Density (mA/cm
2
)
Voltage (V)

400 500 600 700 800 900
0
20
40
60
80
100
(b)

EQE (%)
Wavelength (nm)

Figure 4.19.- 160nm MAPbI3 devices made in the VTE, S11 – S14
ITO/MAPbI3/C60/BCP/Al, (a) current vs voltage (IV) curves and (b) external quantum
efficiency
143

Lastly, solar cells with Ag and Au cathodes were also fabricated in the VTE. The goal of
these devices was to test if using the above metals had an impact on device performance. The
architectures used in those devices are S19 – ITO/MAPbI3/NPD(40nm)/Au,
S20 – ITO/MAPbI3/C60(40nm)/BCP(10nm)/Au, S21 – ITO/MAPbI3/Au, and
S22 – ITO/MAPbI3/NPD-MoO3(140nm)/Au. Film thickness of the perovskite and electrode films
were 330nm and 100nm, respectively. Unexpectedly, devices S19 – S22 exhibited contrasting
-0.5 0.0 0.5 1.0
-20
0
20
(a)
S15 Dark
S15
S16
S17
S18

Current Density (mA/cm
2
)
Voltage (V)
400 500 600 700 800 900
0
20
40
60
80
100
(b)

EQE (%)
Wavelength (nm)

Figure 4.20.- 330nm MAPbI3 device made in the VTE, S15 – S18
ITO/MAPbI3/C60/BCP/Al, (a) current vs voltage (IV) curves and (b) external quantum
efficiency
Device
Jsc
(mA/cm
2
)
Voc
(V)
FF η (%) EQE n
Js
(μA/cm
2
)
Rsa
(Ω/cm
2
)
S15
16.23 0.62 0.32 3.28 15.33 2.18 2.12E-06 1.54E-02
S16
17.03 0.60 0.52 5.39 15.33 2.78 6.13E-05 5.33E-03
S17
17.24 0.63 0.50 5.40 15.33 1.93 2.89E-06 7.43E-03
S18
16.28 0.61 0.43 4.33 15.33 2.31 3.26E-06 1.00E-02
Average
16.70 0.62 0.45 4.60 15.33 2.30 1.74E-05 9.54E-03
Table 4.7.- 330nm MAPbI3 device performance made in the VTE, S15 – S18
ITO/MAPbI3/C60/BCP/Al
144

performance to S11 – S18. Specifically, S19 – S22 essentially displayed similar performance to
VPD devices S2 – S10, i.e., poor-rectifying behavior, as shown in Figure 4.21. Unfortunately,
these devices negate our theory that dense and crystalline films prevents air and water diffusion
therefore avoiding poor-rectifying IV curves and differences between measured EQE and Jsc
values. A possible answer to S19 – S22 performance might lie on the cathode. The cathode used
for S11 – S18 was aluminum while the cathode used for S19 – S22 was gold or silver. Both silver
and gold have a much higher sublimation temperature than aluminum and as mentioned earlier,
perovskite films can experience damage by water and heat. For heat damage to occur, temperatures
above 120°C are necessary in the VTE. While we don’t expect such extreme conditions to ensue
during sublimation of silver or gold, it’s possible such conditions were encountered given
deposition rates of 0.5 Å/s were used to deposit the electrodes. Finally, diffraction peaks at 17°
and 33° 2θ, corresponding to the (110) and (220) planes of MAPbI3 are clearly shown in all four
device architectures, S19 – S22, thus validating full perovskite conversion.

-1.0 -0.5 0.0 0.5 1.0 1.5
-40
0
40
80
(a)
S19
S20
S21
S22
Current Density (mA/cm
2
)
Voltage (V)
10 20 30 40 50 60
(b)
S19
S20
S21
S22

2 Theta (degrees)
Intensity (a.u.)

Figure 4.21.- 330nm MAPbI3 device made in the VTE, (a) current vs voltage (IV)
curves and (b) XRD patterns, S19 square – NPD/Au, S20 circle – C60/BCP/Au,
S21 triangle – Au, and S22 diamond – NPD-MoO3/Au
145

4.3. Conclusions

Hybrid organic-inorganic perovskites solar cells, specifically CH3NH3PbI3 were fabricated in
the VPD and VTE. Films made with both deposition methods were characterized via SEM, AFM,
XRD, Uv-Vis, XPS, and ICP-OES and were found to be analogous to perovskite films reported in
literature. One-step and two-step deposition methods were used to fabricate solar cells in the VPD.
For the two-step method, efficiencies of 3.3 ± 0.3% and 4.1 ± 0.4% and Voc of 0.6 ± 0.1V and
0.8 ± 0.1V were obtained for ITO/MAPbI3(330nm)/C60(40nm)/BCP(10nm)/Ag and
ITO/MAPbI3(330nm)/NPD-MoO3(140nm @ 50% doping concentration)/Ag, respectively.
Further, having excess of PbI2 and/or CH3NH3I in the perovskite films considerably decreased
device performance, i.e., current (Jsc) and fill factor (FF) were impacted greatly. Conversely, fully
converted perovskite films proved to be critical to achieve diode like rectifying IV curves.
Additionally, annealing also had a great impact on device performance. Specifically, smaller Jsc
were observed on non-annealed perovskite films deposited on ITO electrodes, i.e., currents of
9.2 ± 0.6mA/cm
2
and 2.1 ± 0.4mA/cm
2
were obtained

for annealed and non-annealed devices,
respectively. Contrariwise, annealing perovskite films deposited on FTO/TiO2 marginally
impacted the Jsc, thus suggesting that a much rougher TiO2 surface promotes crystallization of the
perovskite film whereas a flatter electrode, i.e., ITO, is not conducive for film crystallization.
However, regardless of the architecture used, the devices fabricated in the VPD were unreliable
and highly unreproducible, displaying poor-rectifying IV curves and noticeably poor match
between Jsc and EQE. Several possible explanations as to why this strange behavior happened were
discussed. First, perovskite films were exposed to ambient air for its completion so it is conceivable
that ambient air adversely affected the perovskite films, i.e., air and water diffused through the
146

film creating problems at the ITO/perovskite interface. This possibility is validated with devices
made in the VTE, which exhibited acceptable rectifying IV curves and comparable EQE and Jsc
values. Second, unreacted (excess) PbI2 and/or MAI influenced solar cells performance, delivering
poor-rectifying IV curves and noticeable discrepancies between Jsc and EQE values, whereas
having a fully converted perovskite film lead to acceptable rectifying IV curves and comparable
EQE and Jsc values. Third, it is also possible the perovskites films made in the VPD are not entirely
crystalline, i.e., a blend of amorphous and crystalline phase exists in the film. The latter is less
probable since amorphous materials tend to scatter X-rays in many directions, thus leading to
bumps on the collected spectra. Whereas, the periodicity of crystalline materials scatters X-rays
only in certain directions, consequently, leading to sharp peaks that are similar to those obtained
in the collected XRD spectra. Regrettably, while it is most likely that perovskite film damage at
the interfaces is what caused the previously mentioned problems, currently there are no
methods/systems that can help us confirm our theory.
Likewise, one-step method was also utilized to fabricate solar cells in the VTE. The
architecture used was ITO/MAPbI3(160 – 330nm)/C60(40nm)/BCP(10nm)/Al. Solar cells with
160nm films exhibited efficiency, Voc and FF of 4.7 ± 0.7%, 0.6 ± 0.03V and 0.55 ± 0.04%,
respectively, whereas devices with 330nm films displayed efficiency, Voc and FF of 4.6 ± 0.7%,
0.6 ± 0.03V and 0.45 ± 0.05%, respectively. Acceptable rectifying IV curves and comparable EQE
and Jsc values were obtained for these devices. Lastly, using silver or gold as electrodes on VTE
solar cells also caused similar behavior to devices made in the VPD. Both silver and gold have
much higher sublimation temperatures than aluminum and it’s possible that low deposition rates
i.e., 0.5 Å/s, produced extreme conditions inside the VTE, consequently harming the perovskite
147

films. While we don’t expect such extreme conditions to ensue inside the VTE, i.e., temperatures
120°C, low deposition rates could allow sufficient time for this to occur.

4.4. Experimental
4.4.1. Characterization of perovskite films

Hybrid organic-inorganic perovskite film morphology, roughness and crystallinity was
characterized using a JEOL JSM-7001F scanning electron microscope (SEM), Digital Instruments
Dimension 3100 atomic force microscopy (AFM) and a Co-source Bruker D8 Powder x-ray
Diffractometer (XRD), respectively. XRD measurements were performed at 10 degrees per
minute, 0.03 step, 0.3 degrees gamma and a 10mm slit. Film thickness were calibrated using an
Ambios XP-2 profilometer stylus and a J.A. Woollam Co., Inc. VASE variable-angle ellipsometer
with a VB-200 control module and a CVI instruments Digikrom 242 monochromator with a 75 W
xenon light source. An Agilent ultraviolet-visible spectrometer was used to obtain the absorbance
spectra, a Kratos Axis Ultra DLD X-ray photoelectron spectrometer (XPS) was used to measure
core level spectra and a thermos scientific iCAP inductively coupled plasma optical emission
spectrometry (ICP-OES) was used to measure lead concentration traces. Lastly, for the calculation
of relative densities the substrates were weight using an OHAUS analytical plus balance model
AP2500.

148

4.4.2. OPV Substrate preparation

Patterned ITO substrates with a resistivity of 1:20 ±5 ohms/sq and an ITO thickness of
2000 ± 50Å were used for the OPVs made in both the VPD and VTE. Substrates were
prepared/cleaned by scrubbing them with Tergitol NP9 (SigmaAldrich Co.)/DI solution followed
by a thorough rinse with DI water. The substrates were then rinsed with acetone (Sigma-Aldrich
Co.) followed by 2-Propanol (SigmaAldrich Co.) and were then blow dried with N2 gas. They
were then washed with tetrachloroethylene (J.T. Baker), acetone (Macron Chemical) and ethyl
alcohol anhydrous reagent (J.T. Baker). The wash consists of placing the substrates, in the order
mentioned above, inside a beaker with each solvent for 10 min while heating the solvent to its
boiling point. After washing, the substrates are blow dried with N2 gas and placed in an oven for
1h at 120°C followed by an ultra-violet ozone cleaning system, model T10X10/OES for 10 min.
Substrates were then transferred to either the VPD or VTE for device fabrication.

4.4.3. VPD device fabrication

A general schematic of the VPD setup can be seen in Figure 4.22, in which the VPD 4” Pyrex
tube/reactor is housed in a Carbolite TVS 12/600/2416CG three zone (1–3) tube furnace.
Preheating of the 4” Pyrex tube (Zone 4) is performed with a 4” x 4” 120 volt, 1100 watts Watlow
mineral insulated band heater. Temperature of the source boats, mineral insulated band heater and
substrate holder was measured and controlled with Omega type K thermocouple probes attached
to Omega CN76000 temperature controllers. Flow of inert gas for each individual source boat,
149

four in total, was measured using individual MKS mass flow controllers (0–500 sccm) while
pressure was measured with a 10 torr Model 626A Baratron Pressure Transducer and adjusted to
desire with a MKS Model 153D Smart Downstream Throttle Valve. All MKS instruments were
monitored and controlled with a MKS Model 647C flow channel controller box while vacuum was
achieved with a Varian IDP3 Oil Free Dry Scroll Pump. The substrate holder, made out of
aluminum, its position approximately 1” away from zone 1 and it can hold two 1” x 1” patterned
substrates. The substrate holder has two shutters which are utilized to control total thickness
deposited onto each individual substrate. The substrate holder was cooled using N2 gas that was
passed through a copper coil submerged in liquid nitrogen. Substrate holder temperature was
controlled by adjusting the flow/volume of cold N2 gas by means of a needle valve. Material
thickness was measured using a 6 MHz Inficon quartz monitor gold coated crystal sensor attached
to an Inficon XTC/2 thin film deposition controller. Proper calibration of the Inficon crystal sensor
was accomplished with a J.A. Woollam Co., Inc. VASE variable-angle ellipsometer with a VB-
200 control module and a CVI instruments Digikrom 242 monochromator with a 75 W xenon light
source and an Ambios XP-2 profilometer stylus, to ensure accurate thickness of films made. The
Inficon crystal sensor was kept at 15 °C at all times using a VWR 1140A chiller.
By introducing the source boat into the hot zone the organic/inorganic molecules were
evaporated into the inert gas stream (N2) which carried the organic compound to the cooled
substrate where the organic/inorganic compounds condensed onto the surface of the substrate.
Organic and inorganic material were deposited at a constant pressure of 0.6 torr, total gas flow
rates of 20 sccm and deposition rates of 1 – 4 Å/s. Further, hybrid organic-inorganic perovskite
films were exposed to air for approximately two minutes while transferring to the VTE to deposit
150

C60, BCP, NPD, MoO3, Al, Ag, and Au. Substrate holder temperatures during deposition of
CH3NH3I and PbI2 in the VPD varied between -100 °C to 100 °C.

4.4.4. VTE device fabrication

Reference devices were made in a Kurt J. Lesker VTE. Film thicknesses was controlled with
a 6MHz Inficon quartz monitor gold coated crystal sensor. All films deposited in the VTE were
performed at pressures ≤4 x 10
-4
Pa, with deposition rates ranging between 0.4 – 2.0 Å/s and at
Crystal Monitor
Substrate Holder
Vacuum
Pump
Cooling Supply
Zone 1 Zone 2 Zone 3 Zone 4
Source Boat
Fused
Silica wall
Carrier Gas
Shutters
Zone 1 Zone 2 Zone 3 Zone 4

Figure 4.22.- Schematic of the vapor phase deposition
Low sublimation temperature (MAI) High sublimation temperature (PbI2)
Zone 1 Pre-Substrate Zone 150°C Zone 1 Pre-Substrate Zone 350°C
Zone 2 Hot Zone 200°C Zone 2 Hot Zone 400°C
Zone 3 Hot Zone 200°C Zone 3 Hot Zone 460°C
Zone 4 Pre-Heating Zone 200°C Zone 4 Pre-Heating Zone 200°C
Table 4.8.-Temperature gradient conditions for sublimation of MAI and PbI2 in the VPD
151

room temperature. Because the VTE is not attached to a glove box, hybrid organic-inorganic films
were exposed to ambient air during the annealing step and electrode mask placement. Proper
calibration of organic-inorganic molecules was accomplished with an Inficon crystal sensor via
spectroscopic ellipsometry with a J.A. Woollam Co., Inc. VASE variable-angle ellipsometer, a
VB-200 control module and a CVI instruments Digikrom 242 monochromator with a 75 W xenon
light source and an Ambios XP-2 profilometer stylus, to ensure accurate thickness of films made.

4.4.5. OPV testing

OPV current density (J) as a function of applied voltage (V) characteristics were measured in
air at room temperature, in the dark and under spectral mismatch corrected 100 mW/cm
2
white
light illumination from an AM-1.5G filtered 300 W Xenon arc lamp (Newport Inc.) and a Keithley
power source meter model 2635A. Routine spectral mismatch correction for ASTM G173-03 was
performed using a filtered silicon photodiode, calibrated by the National Renewable Energy
Laboratory (NREL) to reduce measurement errors. Frequency modulated monochromatic light
(250 Hz, 10 nm FWHM) and lock-in detection was used to perform all spectral responsivity and
spectral-mismatch correction measurements.

4.5. Chapter 4 References

1. Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Journal of the American Chemical
Society 2009, 131, (17), 6050-+.
152

2. Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J. Science 2012,
338, (6107), 643-647.
3. Kim, H. S.; Lee, C. R.; Im, J. H.; Lee, K. B.; Moehl, T.; Marchioro, A.; Moon, S. J.;
Humphry-Baker, R.; Yum, J. H.; Moser, J. E.; Gratzel, M.; Park, N. G. Sci Rep-Uk 2012, 2.
4. Zhou, H. P.; Chen, Q.; Li, G.; Luo, S.; Song, T. B.; Duan, H. S.; Hong, Z. R.; You, J. B.;
Liu, Y. S.; Yang, Y. Science 2014, 345, (6196), 542-546.
5. Wenk, H.-R.; Bulakh, A. G., Minerals : their constitution and origin. Cambridge
University Press: Cambridge ; New York, 2004; p xxii, 646 p.
6. Burlakov, V. M.; Eperon, G. E.; Snaith, H. J.; Chapman, S. J.; Goriely, A. Applied
Physics Letters 2014, 104, (9).
7. Ball, J. M.; Lee, M. M.; Hey, A.; Snaith, H. J. Energ Environ Sci 2013, 6, (6), 1739-1743.
8. Eperon, G. E.; Burlakov, V. M.; Docampo, P.; Goriely, A.; Snaith, H. J. Advanced
Functional Materials 2014, 24, (1), 151-157.
9. Liu, D. Y.; Yang, J. L.; Kelly, T. L. Journal of the American Chemical Society 2014, 136,
(49), 17116-17122.
10. Zhang, T. Y.; Yang, M. J.; Zhao, Y. X.; Zhu, K. Nano Lett 2015, 15, (6), 3959-3963.
11. Liu, M. Z.; Johnston, M. B.; Snaith, H. J. Nature 2013, 501, (7467), 395-+.
12. Im, J. H.; Jang, I. H.; Pellet, N.; Gratzel, M.; Park, N. G. Nat Nanotechnol 2014, 9, (11),
927-932.
153

13. Nie, W. Y.; Tsai, H. H.; Asadpour, R.; Blancon, J. C.; Neukirch, A. J.; Gupta, G.;
Crochet, J. J.; Chhowalla, M.; Tretiak, S.; Alam, M. A.; Wang, H. L.; Mohite, A. D. Science
2015, 347, (6221), 522-525.
14. Aharon, S.; El Cohen, B.; Etgar, L. J Phys Chem C 2014, 118, (30), 17160-17165.
15. Peng, Y. K.; Jing, G. S.; Cui, T. H. J Mater Chem A 2015, 3, (23), 12436-12442.
16. Ito, S.; Tanaka, S.; Nishino, H. J Phys Chem Lett 2015, 6, (5), 881-886.
17. Dong, H.; Wu, Z. X.; Xia, B.; Xi, J.; Yuan, F.; Ning, S. Y.; Xiao, L. X.; Hou, X. Chem
Commun 2015, 51, (43), 8986-8989.
18. You, J. B.; Yang, Y. M.; Hong, Z. R.; Song, T. B.; Meng, L.; Liu, Y. S.; Jiang, C. Y.;
Zhou, H. P.; Chang, W. H.; Li, G.; Yang, Y. Applied Physics Letters 2014, 105, (18).
19. Gao, H.; Ban, C. X.; Li, F. M.; Yu, T.; Yang, J.; Zhu, W. D.; Zhou, X. X.; Fu, G.; Zou, Z.
G. Acs Appl Mater Inter 2015, 7, (17), 9110-9117.
20. Sutherland, B. R.; Hoogland, S.; Adachi, M. M.; Kanjanaboos, P.; Wong, C. T. O.;
McDowell, J. J.; Xu, J. X.; Voznyy, O.; Ning, Z. J.; Houtepen, A. J.; Sargent, E. H. Advanced
Materials 2015, 27, (1), 53-58.
21. Heo, J. H.; Song, D. H.; Han, H. J.; Kim, S. Y.; Kim, J. H.; Kim, D.; Shin, H. W.; Ahn,
T. K.; Wolf, C.; Lee, T. W.; Im, S. H. Advanced Materials 2015, 27, (22), 3424-3430.
22. Wu, Y. Z.; Islam, A.; Yang, X. D.; Qin, C. J.; Liu, J.; Zhang, K.; Peng, W. Q.; Han, L. Y.
Energ Environ Sci 2014, 7, (9), 2934-2938.
154

23. Burschka, J.; Pellet, N.; Moon, S. J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M. K.;
Gratzel, M. Nature 2013, 499, (7458), 316-+.
24. Fu, Y. P.; Meng, F.; Rowley, M. B.; Thompson, B. J.; Shearer, M. J.; Ma, D. W.;
Hamers, R. J.; Wright, J. C.; Jin, S. Journal of the American Chemical Society 2015, 137, (17),
5810-5818.
25. Jeon, N. J.; Noh, J. H.; Kim, Y. C.; Yang, W. S.; Ryu, S.; Il Seol, S. Nature Materials
2014, 13, (9), 897-903.
26. Hao, F.; Stoumpos, C. C.; Liu, Z.; Chang, R. P. H.; Kanatzidis, M. G. Journal of the
American Chemical Society 2014, 136, (46), 16411-16419.
27. Tsai, K. W.; Chueh, C. C.; Williams, S. T.; Wen, T. C.; Jen, A. K. Y. J Mater Chem A
2015, 3, (17), 9128-9132.
28. Papavassiliou, G. C.; Pagona, G.; Karousis, N.; Mousdis, G. A.; Koutselas, I.;
Vassilakopoulou, A. J Mater Chem 2012, 22, (17), 8271-8280.
29. Di, D. W.; Musselman, K. P.; Li, G. R.; Sadhanala, A.; Ievskaya, Y.; Song, Q. L.; Tan, Z.
K.; Lai, M. L.; MacManus-Driscoll, J. L.; Greenham, N. C.; Friend, R. H. J Phys Chem Lett
2015, 6, (3), 446-450.
30. Ng, T. W.; Chan, C. Y.; Lo, M. F.; Guan, Z. Q.; Lee, C. S. J Mater Chem A 2015, 3, (17),
9081-9085.
31. Liu, D. Y.; Gangishetty, M. K.; Kelly, T. L. J Mater Chem A 2014, 2, (46), 19873-19881.
155

32. Stoumpos, C. C.; Malliakas, C. D.; Kanatzidis, M. G. Inorganic Chemistry 2013, 52,
(15), 9019-9038.
33. Shin, W. J.; Lee, J. Y.; Kim, J. C.; Yoon, T. H.; Kim, T. S.; Song, O. K. Organic
Electronics 2008, 9, (3), 333-338.
34. Rand, B. P.; Burk, D. P.; Forrest, S. R. Phys. Rev. B 2007, 75, (11-15), 115327.
35. Wakamiya, A.; Endo, M.; Sasamori, T.; Tokitoh, N.; Ogomi, Y.; Hayase, S.; Murata, Y.
Chem Lett 2014, 43, (5), 711-713.
36. Yoo, E. J.; Lyu, M.; Yun, J. H.; Kang, C. J.; Choi, Y. J.; Wang, L. Advanced Materials
2015, 27, (40), 6170-6175.
37. Francisco F. Navarro, P. I. D., and Mark E. Thompson. Organic Electronics 2014, 15,
(11), 3052–3060.

Abstract (if available)

Abstract The majority of my research focuses on fabrication and characterization of organic and inorganic optoelectronics using a vapor phase deposition (VPD). The VPD, formerly called organic vapor phase deposition (OVPD), was developed to allow sublimation and deposition of organic, inorganic and metal molecules that exhibit high sublimation temperatures, ≤1000℃. Conversely, the OVPD was specifically developed to allow sublimation and deposition of organic‐inorganic compounds that display sublimation temperatures below 400℃. Chapter 1 provides an introduction to the original OVPD method and its accomplishments fabricating organic light emitting diodes (OLEDs), organic photovoltaics (OPVs) and organic thin film transistors (OTFTs). Further, chapter 1 also discusses how the OVPD evolved into the VPD, specifically to enable deposition of thin metal films for optoelectronic devices (discussed in chapter 3). ❧ In chapter 2 we present analysis and characterization of a novel molecule to fabricate OPVs using the OVPD. Specifically, we measured the exciton diffusion length (LD) of platinum tetra 1, 3‐di‐tert‐butylphenyl tetrabenzoporphyrin Pt(ᵗᵇᵘTPBP), by use of the spectrally resolved photoluminescence quenching (SR‐PLQ) method and the optical electric field distribution technique for optimization of OPVs. SR‐PLQ measurements were performed under vacuum (0.08torr), where an average LD of 12.1nm ± 2.4nm was found for Pt(ᵗᵇᵘTPBP). Concurrently, analysis of the films under nitrogen was also performed. However, the LD varied significantly with values ranging from 10nm to 120nm. Significant emission spectra intensity fluctuations were observed during the measurements under nitrogen suggesting oxygen quenching of the porphyrin. Clearly, oxygen quenching altered the LD thus revealing a weakness of the SR‐PLQ method. Likewise, using a program written in MATLAB language for the optical electric field distribution technique, an LD of 16.9nm ± 4nm and 14.8nm ± 3.5nm was obtained for Pt(ᵗᵇᵘTPBP) and C₆₀, respectively. Further, modeling of the optical electric field, photocurrent generation, exciton diffusion profile density, absorption, reflection, among other was also completed using the optical electric field distribution method. Both techniques proved to be great tools for the design and optimization of OPVs. Photovoltaic cells were fabricated in the OVPD and in a vacuum thermal evaporation (VTE) (reference). Fabricated OPVs had the following architecture: ITO/Pt(ᵗᵇᵘTPBP)¹⁰⁻⁶⁰ⁿᵐ/C₆₀¹⁰⁻⁴⁰ⁿᵐ/BCP¹⁰ⁿᵐ/Al¹⁰⁰ⁿᵐ. Open circuit voltages (Voc) in the order of 0.5V and 0.65V were obtained for devices made in the OVPD and VTE, respectively. The results suggest intermixing of the donor/acceptor layers during deposition of C₆₀ in the OVPD devices, thus leading to strong intermolecular interactions. ❧ Further, intermixing is common when films are annealed and strong intermolecular interactions are known to increase dark current, consequently, decreasing the Voc. The efficiencies obtained for the Pt(ᵗᵇᵘTPBP) devices were found to be 0.6% and 1.2% for the OVPD and VTE, respectively. The efficiency discrepancy is due to poor current contribution from the C₆₀ layer (10nm for OVPD vs 40nm for VTE). The external quantum efficiencies (EQE) from the OVPD devices clearly show weak contribution from the acceptor (C₆₀), whereas a stronger current contribution from C₆₀ can be observed on the VTE devices. The results suggest efficiency can be improved for devices made in the OVPD if a thicker C₆₀ film is employed. Similarly, the Voc can also be improved with proper substrate cooling to prevent intermixing of the donor/acceptor layers. Further, a similar Jsc trend was observed on the Pt(ᵗᵇᵘTPBP) versus the Pt(TPBP) devices, confirming a short LD in both molecules is accountable for the decrease in current when a thicker donor film is employed. ❧ In chapter 3 we introduce fully fabricated OPVs and OLEDs with the VPD, an improved version of the OVPD. The objective of the improved/modified VPD was to enable deposition of compounds with high sublimation temperatures e.g., metals and inorganics such as calcium, zinc, cadmium, magnesium, antimony, bismuth, indium, silver, aluminum, zinc sulfide, and manganese, among others. Of the above metals, magnesium (Mg) and zinc (Zn) were selected to fabricate metal films and devices, as they have sublimation enthalpies that are comparable to organic materials commonly used for OLEDs and OPVs. Magnesium and zinc films were deposited in the VPD to prepare optoelectronic devices under low vacuum conditions, i.e. 1 torr. The thin metal films, analyzed via scanning electron microscope (SEM), atomic force microscopy (AFM), X‐ray diffraction (XRD) and four‐point probe resistivity measurements, revealed comparable characteristics to metal films deposited in a VTE. Magnesium cathodes were fabricated for OLEDs and OPVs. OLEDs were fully made in either the VPD or VTE employing aluminum tris‐(8 hydroxyquinoline) [Alq₃] as the green fluorescent emitter or fac‐tris(2‐phenylpyridine)iridium [Ir(ppy)₃] as the green emitting phosphor. Analysis of the OLED devices made in the VPD showed external quantum efficiencies (EQE = 0.9 ±0.1%) and (EQE = 7.6 ±0.6%) at a luminance of 100 cd/m² for the fluorescent and phosphorescent devices, respectively. In addition, OPVs were fully fabricated by both methods employing copper phthalocyanine (CuPc) and C₆₀ as the donor/acceptor materials. Analysis of the OPV devices made in the VPD showed a power efficiency of 0.5 ±0.1%, an open circuit voltage of 0.45 ±0.05% and a fill factor of 0.50 ±0.05%. ❧ The final chapter introduces what we thought was the next logical subject, after studying organics (chapter 2) and metals (chapter 3), inorganic compounds. Chapter 4 presents hybrid organic‐inorganic lead based solar cells grown with the VPD, specifically, perovskite films i.e., methyl ammonium lead iodide (MAPbI₃) as the active layer. Perovskite films were fabricated with both the VPD and VTE and characterized via scanning electron microscopy (SEM), atomic force microscopy (AFM), X‐ray diffraction (XRD), ultraviolet‐visible spectroscopy (Uv‐Vis), X‐ray photoelectron spectroscopy (XPS), profilometry, spectroscopic ellipsometry, and inductively coupled plasma optical emission spectrometry (ICP‐OES). The films were found to be analogous to perovskite films previously reported in literature. One‐step and two‐step deposition methods were used to fabricate solar cells in the VPD. For the two‐step method, efficiencies of 3.3 ± 0.3% and 4.1 ± 0.4% and Voc of 0.6 ± 0.1V and 0.8 ± 0.1V were obtained for ITO/MAPbI₃(330nm)/C₆₀(40nm)/BCP(10nm)/Ag and ITO/MAPbI₃(330nm)/NPD‐MoO₃(140nm @ 50% doping concentration)/Ag, respectively. Further, we found that having excess of PbI₂ and/or CH₃NH₃I in the perovskite films considerably decreased device performance, i.e., current (Jsc) and fill factor (FF) were impacted greatly. Conversely, fully converted perovskite films proved to be critical to achieve diode like rectifying IV curves. Additionally, annealing also had a great impact on device performance were a decreased on Jsc was observed on films deposited on ITO electrodes, i.e., currents of 9.2 ± 0.6mA/cm² and 2.1 ± 0.4mA/cm² were observed for annealed and non‐annealed devices, respectively. Contrariwise, annealing marginally impacted films deposited on FTO/TiO₂ thus suggesting that a much rougher TiO₂ surface promotes crystallization of the perovskite film while a flatter electrode, i.e., ITO, is not conducive for film crystallization. ❧ However, regardless of the architecture used, we found that the devices fabricated in the VPD were unreliable and highly unreproducible, displaying poor‐rectifying IV curves and noticeably weaker Jsc and/or EQE response from the devices. Several possible explanations as to why this strange behavior happened were discussed. First, perovskite films were exposed to ambient air for its completion so it is conceivable that ambient air adversely affected the perovskite films, i.e., air and water diffused through the film creating problems at the ITO/perovskite interface. This possibility is validated with devices made in the VTE, which exhibited acceptable rectifying IV curves and comparable EQE and Jsc values. Second, unreacted (excess) PbI₂ and/or MAI influenced solar cells performance, delivering poor‐rectifying IV curves and noticeable discrepancies between Jsc and EQE values, whereas having a fully converted perovskite film lead to acceptable rectifying IV curves and comparable EQE and Jsc values. Third, it is also possible the perovskites films made in the VPD are not entirely crystalline, i.e., a blend of amorphous and crystalline phase exists in the film. The latter is less probable since amorphous materials tend to scatter X‐rays in many directions, thus leading to bumps on the collected spectra. Whereas, the periodicity of crystalline materials scatters X‐rays only in certain directions, consequently, leading to sharp peaks that are similar to those obtained in the collected XRD spectra. Regrettably, while it is most likely that perovskite film damage at the interfaces is what caused the previously mentioned problems, currently there are no methods/systems that can help us verify our theory.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Scale-up of vapor-phase deposition of polymers: towards large-scale processing

PDF

Vapor phase deposition of dense and porous polymer coatings and membranes for increased sustainability and practical applications

PDF

Fabrication of polymer films on liquid substrates via initiated chemical vapor deposition: controlling morphology and composition

PDF

Organic photovoltaics: from materials development to device application

PDF

The patterning of polymer thin films on porous substrates via initiated chemical vapor deposition

PDF

Inorganic/organic hybrid materials and grafted coatings via vapor phase deposition processes

PDF

Vapor phase deposition of polymers in the presence of low vapor pressure liquids

PDF

Controlling polymer film patterning, morphologies, and chemistry using vapor deposition

PDF

Porphyrin based near infrared‐absorbing materials for organic photovoltaics

PDF

Chemical vapor deposition of graphene: synthesis, characterization, and applications

PDF

Formation of polymer gels, films, and particles via initiated chemical vapor deposition onto liquid substrates

PDF

Fabrication of silicon-based membranes via vapor-phase deposition and pyrolysis of organosilicon polymers

PDF

Molecular design of pyridine and quinoline based compounds for organic optoelectronics

PDF

Solution‐phase synthesis and deposition of earth‐abundant metal chalcogenide semiconductors

PDF

Fabrication, deposition, and characterization of size-selected metal nanoclusters with a magnetron sputtering gas aggregation source

PDF

Simple complexes: synthesis and photophysical studies of luminescent, monovalent, 2-coordinate carbene-coinage metal complexes and higher coordination geometries

PDF

Integration of cost effective luminescent materials into organic light emitting diodes (OLEDs) and molecular approaches to improve the device efficiency

PDF

Expanding the library of surface ligands for semiconductor nanocrystal synthesis and photovoltaic applications

PDF

Characterization, process analysis, and recycling of a benzoxazine-epoxy resin for structural composites

PDF

Solution processing of chalcogenide functional materials using thiol–amine “alkahest” solvent systems

Asset Metadata

Creator Navarro, Francisco Fabian (author)

Core Title Fabrication and study of organic and inorganic optoelectronics using a vapor phase deposition (VPD)

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemical Engineering

Publication Date 02/03/2016

Defense Date 12/09/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,OLEDs,OPVs,OVPD,solar cells,thin films,VPD

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Thompson, Mark E. (committee chair), Lee, Charles Ted Lee, Jr. (committee member), Malmstadt, Noah (committee member)

Creator Email ffnavarr26@gmail.com,fnavarro@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-204973

Unique identifier UC11278357

Identifier etd-NavarroFra-4081.pdf (filename),usctheses-c40-204973 (legacy record id)

Legacy Identifier etd-NavarroFra-4081.pdf

Dmrecord 204973

Document Type Dissertation

Format application/pdf (imt)

Rights Navarro, Francisco Fabian

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA