Embed Size (px)
Citation preview
TiO2 AND ITS DERIVATIVES: SYNTHESIS, CHARACTERIZATION AND
APPLICATION IN H2 PRODUCTION VIA WATER
SPLITTING AND IN BULK HETEROJUNCTION SOLAR
CELLS
Thèse
THI THUY DUONG VU
Doctorat en génie chimique
Philosophiae doctor (Ph.D.)
Québec, Canada
© Thi Thuy Duong Vu, 2015
iii
Résumé
Dans un contexte de crise environnementale et d'épuisement des ressources énergétiques
conventionnelles, le modèle énergétique obsolète fondé sur les combustibles fossiles doit être
redéfini et redessiné. Malgré plusieurs types d'énergies alternatives renouvelable en développé et en
cours de développé, en sachant qu'elles jouent un rôle important à moyen et long terme, l'utilisation
de l'énergie solaire présente actuellement un grand intérêt aux scientifiques. La production
d'hydrogène par la dissociation de l'eau et le dispositif photovoltaïque en convertit directement la
lumière solaire en électricité devient plus compétitifs mais son coût ne cesse de diminuer en
parallèle du progrès de la technologie. En conséquence, cette thèse concentre sur la synthèse et la
modification de nanoparticules de dioxyde de titane (TiO2) et aussi parlant de la fabrication et de
l'optimisation des dispositifs basés sur ces nanoparticules pour des applications photovoltaïque et de
la photo-catalyse par la dissociation de l’eau.
La synthèse et la modification des nanoparticules de TiO2 ont été optimisées pour contrôler
la morphologie des particules, spécialement leur taille et leur forme, en utilisant différents types de
surfactants. Ceci nous a permis de développer des nanoparticules de TiO2 avec différentes formes,
telles que les nanosphères, les nanotiges, les nanorhombiques, et différentes tailles allant de 3 x 40
nm à 3 x 20 nm. L’effet du surfactant sur la morphologie des nanoparticules de TiO2 a été
soigneusement caractérisé et analysé. La modification de la surface des nanoparticules de TiO2 ainsi
développées par du sulfure de cadmium (CdS) a été optimisée dans le but de les utiliser dans les
cellules solaires hybrides à hétérojonction volumique (BHJs) et aussi pour la production
d’hydrogène via la dissociation de l’eau.
Il a été démontré que l’efficacité de conversion de la puissance énergétique des BHJs a été
augmentée de l'ordre de 17 fois en utilisant les nanotiges modifiées TiO2/CdS comparativement au
nanotiges TiO2 non modifiées. Finalement, il a été démontré que la modification en surface des
nanoparticules de TiO2 par du CdS et du Nickel menait à une nette amélioration dans la
performance production d’hydrogène via la dissociation de l’eau. Cette réaction de dissociation
présentait une stabilité.
v
Abstract
In a context of environmental crisis and depletion of conventional energy resources, the
current energy model based on fossil fuels is obsolete and needs to be redefined and redesigned.
Even though, there are many different renewable alternatives developed or under developing, which
are expected to take a main role in the middle and long term. The use of energy from the sun is
currently attracting much attention from the scientists. For example, hydrogen generation via water
splitting and photovoltaic devices that convert directly sunlight into electricity become more
competitive as the cost continues to decrease with the technology advancement. Taking this into
account, this thesis is focused on the synthesis and modification of titanium dioxide nanoparticles
(TiO2 NPs) and the development and optimization of devices based on these nanoparticles for
photovoltaic applications and photocatalyst water splitting.
The synthesis of TiO2 NPs was mainly emphasized on controlling the morphologies,
especially their shape and size, by using different types of capping agents. TiO2 NPs with various
shapes, such as nanosphere, nanorod, nanorhombic, and various sizes from 3 x 40 nm to 3 x 20 nm
were achieved. The effects of capping agent on TiO2 NPs morphologies were characterized and
analyzed carefully. Based on the developed TiO2 NPs, cadmium sulfide (CdS) was deposited on the
surface of TiO2 NPs, and then was optimized for the hybrid bulk heterojunction solar cells (BHJs)
and photocatalytic hydrogen production via water splitting.
Especially, with the use of TiO2-based nanocomposites in BHJs systems, it showed
improvement of around 17 times in power efficiency conversion compared to the system used
unmodified TiO2 NPs. On the other hands, with the use of a new non-noble metal-nanocomposites
composed of CdS/TiO2, and Nikel clusters, the performance of the photocatalytic hydrogen
production via water splitting system was enhanced and it showed that the reaction is stable up to
15h.
vii
Table of Contents
Résumé……. ......................................................................................................................... iii
Abstract…… ........................................................................................................................... v
Table of Contents ................................................................................................................. vii
Index of Tables ................................................................................................................... xiii
Index of Schemes .................................................................................................................. xv
Index of Figures ................................................................................................................. xvii
Abbreviations .................................................................................................................... xxiii
Symbols xxv
Acknowledgements .......................................................................................................... xxvii
Foreword xxix
Chapter 1. Introduction .......................................................................................................... 1
1.1. Overview ..................................................................................................................... 1
1.2. Objectives of the thesis ............................................................................................... 2
1.3. Summary of the Articles ............................................................................................. 3
1.4. References ................................................................................................................... 5
Chapter 2. Literature Review ................................................................................................. 9
2.1. Description of Solar Spectrum .................................................................................... 9
2.2. Titanium dioxide and its derivatives for alternative energy ..................................... 10
2.2.1. Titanium Dioxide Nanoparticles ................................................................... 10
2.2.2. Coupled Colloidal Structures ........................................................................ 12
2.3. H2 Production via Photocatalysis Water Splitting .................................................... 12
2.3.1. Working Principle ......................................................................................... 13
viii
2.3.2. State-of-the-art of H2 production based on TiO2 NPs and its derivatives .... 14
2.4. Photovoltaic Application .......................................................................................... 24
2.4.1. Working Principle ......................................................................................... 27
2.4.2. Solar Cell Characteristic ............................................................................... 28
2.4.3. State-of-the-art of BHSCs based on TiO2 NPs and their derivatives ........... 33
2.5. Reference .................................................................................................................. 44
Chapter 3. Experimental ..................................................................................................... 53
3.1. Experimental Tools .................................................................................................. 53
3.1.1. Microscopy ................................................................................................... 53
3.1.2. X-Ray Diffraction ......................................................................................... 54
3.1.3. X-Ray Photoelectron Spectroscopy .............................................................. 55
3.1.4. Fourier Transform Infrared Spectroscopy .................................................... 57
3.1.5. Ultraviolet-Visible Spectroscopy ................................................................. 58
3.1.6. Photoluminescence (PL) ............................................................................... 59
3.1.7. Zeta (ζ) - Potential Analysis ......................................................................... 60
3.1.8. Thermogravimetric analysis (TGA) ............................................................. 61
3.1.9. Brunauer–Emmett–Teller (BET) Specific Surface Area Analysis ............... 62
3.1.10. Gas Chromatography Analysis ..................................................................... 64
3.2. Techniques ................................................................................................................ 64
3.2.1. Wet Chemical Processing ............................................................................. 64
3.2.2. Spin-coating .................................................................................................. 66
3.2.3. Thermal evaporation ..................................................................................... 67
3.3. Reference .................................................................................................................. 68
Chapter 4. Synthesis of Titanium Dioxide/Cadmium Sulfide Nanosphere Particles for
Photocatalyst Applications ................................................................................ 69
ix
Abstract ............................................................................................................................ 70
Résumé ............................................................................................................................. 71
4.1. Introduction ............................................................................................................... 72
4.2. Experimental ............................................................................................................. 73
4.2.1. Materials ........................................................................................................ 73
4.2.2. Synthesis of length-controlled TiO2 nanorods using oleic acid and 6-
aminohexanoic acid as surfactants ................................................................ 73
4.2.3. Development of TiO2 nanorods by Ligand Exchange Reaction ................... 74
4.2.4. Synthesis of Colloidal Hybrid TiO2/CdS nanocomposite ............................. 74
4.2.5. Synthesis of Ni-TiO2/CdS by a Photodeposition method ............................. 74
4.2.6. Characterization ............................................................................................ 75
4.2.7. Photocatalysis characterization (Photocatalytic H2 evolution) ..................... 76
4.3. Results and Discussions ............................................................................................ 76
4.3.1. TEM, FTIR and BET characterization .......................................................... 77
4.3.2. XRD characterization .................................................................................... 84
4.3.3. XPS and SEM-EDX characterization ........................................................... 86
4.3.4. UV/Vis and Photoluminescence (PL) characterizations ............................... 87
4.3.5. Thermal Gravimetric (TGA) and ζ-potential characterization ...................... 89
4.3.6. Photocatalytic activity ................................................................................... 91
4.4. Conclusions ............................................................................................................... 95
4.5. References ................................................................................................................. 95
Chapter 5. Synthesis of capped TiO2 nanocrystals of controlled shape and their use with
MEH-PPV to develop nanocomposite films for Photovoltaic applications ....... 99
Abstract .......................................................................................................................... 100
Résumé ........................................................................................................................... 101
5.1. Introduction ............................................................................................................. 102
x
5.2. Experimental........................................................................................................... 103
5.2.1. Materials ..................................................................................................... 103
5.2.2. Synthesis of TiO2 nanoparticles ................................................................. 103
5.2.3. Synthesis of MEH-PPV/TiO2 nanocomposites .......................................... 104
5.2.4. Characterization .......................................................................................... 104
5.3. Results and Discussion ........................................................................................... 105
5.3.1. Synthesis and characterization of capped TiO2 nanoparticles .................... 105
5.3.2. Development and characterization of MEH-PPV/TiO2 nanocomposite films
.................................................................................................................... 113
5.4. Conclusions ............................................................................................................ 121
5.5. Reference ................................................................................................................ 121
Chapter 6. The effect of surfactants on the photovoltaic properties of hybrid bulk
heterojunction solar cells based on MEH-PPV and TiO2-based materials ..... 125
Abstract .......................................................................................................................... 126
Résumé .......................................................................................................................... 127
6.1. Introduction ............................................................................................................ 128
6.2. Experimental........................................................................................................... 129
6.2.1. Materials ..................................................................................................... 129
6.2.2. Synthesis of OA and OM or 6-AHA Capped TiO2 nanorods ..................... 129
6.2.3. Synthesis of CdS modified TiO2 nanocomposite ....................................... 130
6.2.4. Preparation of MEH-PPV/capped-TiO2 and MEH-PPV/CdS/TiO2 blend
solutions ...................................................................................................... 130
6.2.5. Fabrication of BHJ solar cell devices ......................................................... 131
6.2.6. Characterization .......................................................................................... 132
6.3. Results and Discussions ......................................................................................... 132
6.3.1. Analysis of synthesized capped-TiO2 nanorods ......................................... 132
xi
6.3.2. Analysis of the synthesized CdS modified TiO2 nanorods ......................... 138
6.3.3. Characterization of BHJSCs with active layers based on MEH-PPV/capped-
TiO2 or MEH-PPV/CdS/TiO2 ..................................................................... 141
6.4. Conclusion .............................................................................................................. 149
6.5. Reference ................................................................................................................ 150
Chapter 7. Conclusion ....................................................................................................... 153
7.1. General conclusions ................................................................................................ 153
7.2. Prospects ................................................................................................................. 155
Annex A - Aminoacid-asisted Synthesis of TiO2 Nanocrystals with Controllable Shape and
Size: A Novel Agent for the Fabrication of Polymer/TiO2 Photovoltaic
Materials .......................................................................................................... 157
Abstract .......................................................................................................................... 158
Résumé ........................................................................................................................... 159
A1. Introduction ............................................................................................................. 160
A2. Experimental ........................................................................................................... 160
A3. Results and Discussions .......................................................................................... 161
A4. Conclusion .............................................................................................................. 165
A5. Reference ................................................................................................................ 166
xiii
Index of Tables
Table 2.1. Photovoltaic properties of hybrid MEH-PPV/TiO2 NRs capped by different
ligands ................................................................................................................ 36
Table 2.2. Photovoltaic properties of BHJs based on MEH-PPV and various nanocrystals
under the illumination of AM1.5, 80 mW/cm−2................................................. 43
Table 3.1. Characteristic frequencies in FTIR7 ..................................................................... 58
Table 6.1. Summary of the photovoltaic parameters of BHJSC devices with active layer
blends A1, A2, S1, and S2 ............................................................................... 149
xv
Index of Schemes
Scheme 4.1. Sketch for the preparation of TiO2/CdS nanocomposites................................77
Scheme 6.1. Architecture scheme of MEH-PPV:CdS/TiO2 hybrid solar cell device …....131
xvii
Index of Figures
Figure 2.1. Solar radiation spectrum (Image created by Robert A. Rohde)6 .......................... 9
Figure 2.2. Crystal structures of TiO2 rutile and TiO2 anatase phase19 ................................ 11
Figure 2.3. Principle of water splitting using semiconductor photocatalysts ....................... 14
Figure 2.4. Relationship between the band structure of semiconductors and the redox
potential of water splitting38 ............................................................................... 15
Figure 2.5. H2 evolution by water splitting over TiO2 catalysts (a) without any sacrificial
agents and UV light irradiation; (b) without any sacrificial agents and visible
light irradiation; (c) using ethanol as sacrificial agent and visible light
irradiation; (d) Mechanism of H2 evolution by water splitting over a Fe–
Ni/TiO2 photocatalyst under visible light irradiation ......................................... 19
Figure 2.6. Schematic illustration of the photo-induced charge injection process that occurs
upon excitation of the CdS component of a CdS/TiO2 colloid in the presence of
a sacrificial electron donor D. ............................................................................ 22
Figure 2.7. (a) TEM image of CdS/TiO2 nanotube; (b) The average rate of H2 evolution and
(c) the amount of H2 evolved vs irradiation time on various photocatalysts: (a)
CdS/TiO2 containing 13.44 wt% CdS ; (b) CdS/TiO2 containing 8.32 wt% CdS;
(c) the physical mixture of 20 wt% CdS/80 wt% TiO2 nanotube; (d) a pure CdS
powder. ............................................................................................................... 22
Figure 2.8. Mechanism of Z-scheme system for water-splitting.68 ...................................... 23
Figure 2.9. Current state of solar cell efficiencies (Reprint from National Renewable
Energy Laboratory (NREL) website) ................................................................. 26
Figure 2.10. (a) Structure of BHJ solar cells (b,c) Scheme drawing of the working principle
of an organic photovoltaic cell. .......................................................................... 27
Figure 2.11. Air mass measurement ..................................................................................... 29
Figure 2.12. Schematic illustration of carriers flow in short-circuited external circuit. ....... 30
Figure 2.13. Illumination energy band diagrams of p–n junction in (a) the short-circuited
and (b) open-circuited current. ........................................................................... 31
Figure 2.14. Current–voltage characteristics of p–n junction under illumination and
darkness. ............................................................................................................. 31
xviii
Figure 2.15. Chemical structures of conjugated polymers used as donors in BHJ solar cells
........................................................................................................................... 34
Figure 2.16. (a)The PL spectra from MEH-PPV: TiO2 layers of 70% TiO2 content with
different capped ligands. (b) The J-V characteristics of the PV devices under
AM 1.5 solar simulator (100 mW/cm2). ............................................................ 36
Figure 2.17. TEM of TiO2 nanorods (a) and dots (b), obtained by hydrolysis method at
100°C: (a) OLEA 35 g, TTIP 5 mmol, 2M TMAO 5 ml; (b) OLEA 35 g, EG
3.2 g, TTIP 1 mmol, TMAO 4 mmol 109. .......................................................... 37
Figure 2.18. (a) J–V characteristics for P3HT:TiO2 cells for different TiO2 concentrations.
(b) The comparison between external quantum efficiency (EQE) for
nanocomposite and pure P3HT cells. (c) J–V characteristics of P3HT:TiO2 cells
with P3HT:TiO2 films spin-coated from different solvents. Inset:g shows the
energy diagram of the devices. .......................................................................... 38
Figure 2.19. TEM of PbS/TiO2 nanocomposites fabricated under different conditions (a)
high concentration of OA results in the formation of small-diameter (d < 3 nm)
PbS domains. (b) The formation of single, large-diameter PbS NCs (d > 4)
when the concentration of OA in the solution is low. (c) The formation of
multiple large-diameter PbS NCs per single nanorod occurs when the
concentration of OA is low, and concentrations of Pb and S precursors are high.
(d−f) HRTEM images of nanocomposites shown in (a−c), respectively. ......... 41
Figure 2.20. Optical properties of PbS/TiO2. (a−b) Absorbance of PbS/TiO2
nanocomposites representing several structural types. (c) Fluorescence intensity
decay and (d) emission profile of PbS/TiO2 heterostructures containing 4.2 nm
PbS NCs. ........................................................................................................... 42
Figure 2.21. (LEFT) J–V curves of the BHJs based on MEH-PPV and the nanocrystals of
Na1, Na2, Na3 and CdSe under illumination of AM1.5, 80 mW/cm−2. (RIGHT)
Band gap energy level positions of MEH-PPV and NC determined by CV
measurements. ................................................................................................... 43
Figure 3.1. (a) Illustration of X-ray diffraction structure; (b) Schematic illustration of the
Bragg’s law. ....................................................................................................... 55
Figure 3.2. The mechanism of photoelectron emission in XPS process .............................. 56
xix
Figure 3.3. Zeta potential in colloid systems. ....................................................................... 60
Figure 3.4. (a) Sample of TGA curve. Note the plateau of constant weight (region A), the
mass loss portion (region B), and another plateau of constant mass (region C);
(b) Typical shape of TGA where 1 - no change; 2 - desorption/drying; 3 –
single stage decomposition; 4 - multi-stage decomposition; 5 - as 4, but no
intermediates or heating rate too fast; 6 - atmospheric reaction; 7 – as 6, but
product decomposes at higher temperature. ....................................................... 62
Figure 3.5. Scheme of ITO etching process .......................................................................... 66
Figure 3.6. The vacuum thermal evaporation deposition system ......................................... 67
Figure 4.1. TEM image of the synthesized TiO2 nanorods before sonication. ..................... 77
Figure 4.2. TEM images of synthesized TiO2 nanorods after sonication a) 3x40 nm
nanorods for TB:OA:6AHA molar concentration of 1:7:3, and b) 3x10nm
nanorods for TB:OA:6AHA molar concentration of 1:7:10. ............................. 78
Figure 4.3. (a) Surfactant-capped TiO2 nanorods dissolved in toluene; (b) TiO2 nanorods
after NOBF4 treatment dissolved in DMF. ........................................................ 79
Figure 4.4. FTIR of (a) capped-TiO2 nanorod synthesized using OA and 6AHA as
surfactants; and (b) TiO2/CdS nanoparticles. ..................................................... 81
Figure 4.5. (a) TEM image of TiO2/CdS nanocomposite, and (b) BET characterization of
TiO2, CdS, and TiO2/CdS nanocomposite with the inset is their corresponding
pore size distribution .......................................................................................... 83
Figure 4.6. XRD characterization of a) TiO2 nanorod b) TiO2/CdS nanocomposite. .......... 85
Figure 4.7. (a) XPS characterization of Ni-TiO2/CdS nanocomposite (b) High-resolution
XPS of Ni ........................................................................................................... 86
Figure 4.8. SEM-EDX characterization of Ni-TiO2/CdS nanocomposite ............................ 87
Figure 4.9. (a) UV-Vis spectra of TiO2, CdS and TiO2/CdS (b) Photoluminescence (PL)
emission spectra under excitation at a wavelength of 380 nm for CdS and
TiO2/CdS nanocomposite. .................................................................................. 88
Figure 4.10. TGA characterization of (black) TiO2 nanorods (blue) CdS NPs (red)
TiO2/CdS nanocomposite. .................................................................................. 89
xx
Figure 4.11. ζ-Potential distributions in aqueous solution at pH~5 of TiO2 nanorods before
and after treatment with NOBF4 surfactant; CdS NPs, and TiO2/CdS
nanocomposite. .................................................................................................. 91
Figure 4.12. (a) Comparison of the activity of H2 evolution using different photocatalysts;
(b) H2 production from TiO2/CdS-Ni photocatalyst monitored over 18 h. Each
4.5 h, the reaction system is bubbled with N2 to remove the H2 inside. ........... 93
Figure 4.13. Mechanism illustration of the activity of Ni-TiO2/CdS under visible light for
the production of H2, inset is the potential redox energy corresponding to CdS,
TiO2 and H+/H2 .................................................................................................. 94
Figure 5.1. TEM of synthesized TiO2 NPs with different shapes: (a) nanosphere, (b)
nanorhombic, and (c) nanorod. ........................................................................ 105
Figure 5.2. XRD of synthesized TiO2 NPs with different shapes. ..................................... 106
Figure 5.3. FTIR spectra of capped- TiO2 NPs with different shapes; inset [1] in the region
1400-1700cm-1; in set[2] in the region 2800-3200 cm-1. ................................. 108
Figure 5.4. TGA curves of TiO2 NPs characterized at a heating rate of 10 oC/min under O2
atmosphere. ...................................................................................................... 110
Figure 5.5. UV-vis absorption spectra of the three synthesized TiO2 NPs of different shapes
in CHCl3 solvent. ............................................................................................. 111
Figure 5.6. Band gaps of the three synthesized TiO2 NPs determined from the plot of versus
photon energy: (a) nanosphere, (b) nanorod, and (c) nanorhombic. ............... 112
Figure 5.7. TEM of composite of MEH-PPV and synthesized TiO2 NPs with different
shapes: (a) nanosphere, (b) nanorhombic, and (c) nanorod ............................ 114
Figure 5.8. TGA curves of (a) pure MEH-PPV (b) MEH-PPV/TiO2 nanocomposites
characterized at a heating rate of 10 oC/min under air atmosphere. ................ 115
Figure 5.9. UV-vis absorption spectra of MEH-PPV/TiO2 nanocomposites: (a) different
TiO2 shapes, and (b) TiO2 nanospheres of different concentrations. .............. 116
Figure 5.10. FTIR of MEH-PPV and MEH-PPV/TiO2 nanocomposites. Bottom inset: FTIR
spectra of MEH-PPV and MEH-PPV/TiO2 nanocomposites using TiO2
nanospheres. .................................................................................................... 118
Figure 5.11. Photoluminescence (PL) emission of MEH-PPV/ TiO2 nanocomposites: (a)
different TiO2 shapes, and (b) TiO2 nanorods of different concentrations. .... 119
xxi
Figure 6.1. TEM of TiO2 nanorods synthesized using (a) OA/OM, and (b) OA/6AHA
surfactants combinations .................................................................................. 133
Figure 6.2. Powder XRD patterns of OA-OM-capped-TiO2 and OA-6AHA-capped-TiO2
nanoparticles. The diffraction pattern of TiO2 anatase is also reported as a
reference ........................................................................................................... 135
Figure 6.3. FTIR spectra of OA-OM-capped-TiO2 (A1), OA-6AHA-capped-TiO2 (A2)
nanoparticles, pure OA, OM, and 6AHA ........................................................ 135
Figure 6.4. TGA spectra of OA-OM-capped-TiO2 nanoparticles (A1) and OA-6AHA-
capped-TiO2 (A2) nanoparticles (heating rate: 10ºC/min, O2 atmosphere) ..... 136
Figure 6.5. UV-vis characterization of capped-TiO2 nanoparticles (a) OA-OM-capped-
TiO2, (b) OA-6AHA-capped-TiO2. The insets show their respective band gap
energy plots. ..................................................................................................... 138
Figure 6.6. FTIR curves of the two developed CdS/TiO2 nanocomposites, together with
those of pure OA, OM, 6AHA and NOBF4. .................................................... 139
Figure 6.7. TGA spectra of CdS/OA-OM-capped-TiO2 (A1) and CdS/OA-6AHA-capped-
TiO2 (A2) (heating rate: 10ºC/min, O2 atmosphere). ....................................... 140
Figure 6.8. UV-vis characterization of capped-TiO2 nanoparticles, CdS, and CdS/ TiO2
nanocomposites. ............................................................................................... 141
Figure 6.9. SEM pictures of BHJSC active layer blends: A1 (a), A2 (b), S1 (c), and S2 (d)
.......................................................................................................................... 142
Figure 6.10. TGA spectra of BHJSC active layer blends A1, A2, and S2 (heating rate:
10ºC/min, O2 atmosphere). .............................................................................. 143
Figure 6.11. UV-vis of polymer composite of MEH-PPV and two different CdS/TiO2
nanocomposites. MEH-PPV/OA-OM-capped-TiO2 NPs (blend A1), MEH-
PPV/OA-6AHA-capped-TiO2 NPs (blend A2), MEH-PPV/CdS/OA-OM-
capped-TiO2 NPs (blend S1) and MEH-PPV/CdS/OA-6AHA-capped-TiO2
(blend S2). ........................................................................................................ 144
Figure 6.12. Photoluminescence (PL) of pure MEH-PPV and BHJSC active layer blends
A1, A2, S1, and S2. .......................................................................................... 146
xxii
Figure 6.13. J-V Characterization under light illumination (1.5 AM) of BHJSC devices with
the active layer blends (a) A1 and A2, and (b) S1 and S2. The insets are their
corresponding log J–V properties in the dark and under light. ....................... 148
Figure A.1. TEM images of TiO2 nanoparticles (a) TiO2 nanospheres with an average size
of 5 nm (b) TiO2 nanobars with an average size of 10 nm x 20 nm. ............... 162
Figure A.2. SEM images of (a) nanobars TiO2/PS and (b) nanosphere TiO2/PS films ..... 163
Figure A.3. UV-vis absorption spectra for the samples of glass, pure PS film, TiO2
nanobars/PS and TiO2 nanosphere/PS films. ................................................... 164
Figure A.4. FTIR spectra of (a) nanosphere TiO2/PS film (b)nanobar TiO2/PS film, and (c)
pure PS film. .................................................................................................... 165
xxiii
Abbreviations
AFM Atomic Force Microscopy
Ag Silver
Al Aluminium
AM1.5G Air Mass 1.5 Global
Au Gold
BET Braunauer-Emmer-Teller
BHJ Bulk Heterojunction
BHJSC Bulk Heterojunction Solar Cells
Ca Calcium
CB Conduction Band
CBD Chemical Bath Deposition
Cd Cadmium
Cd-O Cadmium-Oleate
CdS Cadmium Sulphide
CdSe Cadmium Selenide
CdTe Cadmium Telluride
EDS Energy Dispersive X-Ray Spectroscopy
EQE External Quantum Efficiency
eV Electron Volt
FF Fill Factor
FTIR Fourier Transform Infrared Spectroscopy
GC Gas Sorption And Gas Chromatography
HOMO Highest Occupied Molecular Orbital
ITO Indium Tin Oxide
MDMO-PPV Poly[2-Methoxy-5-(3′,7′-Dimethyloctyloxy)-1,4-Phenylenevinylene]
xxiv
MEH-PPV Poly[2-Methoxy-5-(2-Ethylhexyloxy)-1,4-Phenylenevinylene]
NCs Nanocrystals
NPs Nanoparticles
OA Oleic Acid
OM Oleyamine
P3HT Poly(3-Hexylthiophene-2,5-Diyl)
PbS Lead Sulphide
PbSe Lead Selenide
PCE Power Conversion Efficiency
PEDOT:PSS Poly(3,4-Ethylenedioxythiophene)-Poly(Styrenesulfonate)
PL Photoluminsience
PV Photovoltaic
QDs Quantum Dots
SEM Scanning Electron Microscopy
TEM Transmission Electron Microscopy
TGA Thermogravimetric Analysis
TiO2 Titanium Dioxide
UV-vis Ultraviolet-Visible Spectroscopy
VB Valence Band
XPS X-Ray Photoelectron Spectroscopy
XRD X-Ray Diffraction
ZnO Zinc Oxide
ZnS Zinc Sulphide
xxv
Symbols
Eg Band gap
λ Wavelength
E Photon energy
h Planck's constant
Hz Hertz
ν Frequency of electromagnetic radiation
J Current density
V Voltage
Pmax Maximum power
Jsc Short-circuit current density
Voc Open-circuit voltage
η Conversion efficiency or Dynamic viscosity
J Joule
Å Angstrom
d Lattice plane distance
θ Angle of incidence
D Particle size
Ek Kinetic energy of photoelectron
ϕ Spectrophotometer work function
α Absorption coefficient
kB Boltzmann’s constant
T Temperature
Rq Gas constant
xxvii
Acknowledgements
Firstly, I would like to thank Professor Frej Mighri for his excellent supervision and
guidance, support and patience throughout my Ph.D studies, as well as for the great
opportunity that he offered me to work in his lab during the last four years. I would also
like to thank to my co-director, Professor Trong-On Do who gave me invaluable thoughtful
insights, advices, support, discussions and encouragements.
I am very grateful for the generous and contentious helps from present and former
members of professors Mighri and Do research groups and from my other colleagues at the
Department of chemical engineering, Université Laval. I would especially thank Mr. Jayesh
Patel and Mr. Yann Giroux for their friendship and generous help in performing the thesis
work.
I am also grateful to the Natural Science and Engineering Research Council of
Canada (NSERC) and Laval University for their financial support.
To my deceased grandpas, deceased grandmas, my uncles, my aunties, and my
cousins, I would like to send my warmest thanks for their love, encouragement and support
during all the years of my education.
Most important of all, especially I would like to thank my parents and my sister.
Finally, I would like to express my gratitude to all my Vietnamese friends at Laval
University for their friendship, support and encouragement.
xxix
Foreword
This thesis is written in the form of a collection of three publications with two
introductory chapters. The purpose of these introductory chapters is to provide an
introduction to the field of solar cell research, motivate the topic of the thesis, and provide
additional theory and experimental details that are not presented in the papers.
The thesis is divided into seven chapters. A brief general introduction is presented
in Chapter 1. Chapter 2 presents literature review related to the objectives of the
dissertation and an introduction to renewable energy technologies. Chapter 3 gives the
fundamental backgrounds of characterizations and techniques that were used in this PhD
work.
Chapters 4, 5 and 6 correspond to the main body of the thesis and were presented as
published or submitted. Chapter 4 presents a new hybrid photocatalytic system for the
production of H2 under visible light illumination using ethanol as a sacrificial agent. This
hybrid system was based on TiO2 nanorods, CdS nanoparticles and Ni cluster cocatalyst.
The corresponding results were published in Industrial & Engineering Chemistry Research
2014, 3888-3897. Chapter 5 reports the synthesis details of TiO2 nanoparticles of different
morphologies using oleic acid (OA) and oleyl amine (OM) as capping agents, as well as the
promising properties of polymer nanocomposites based on these synthesized TiO2 NPs. The
corresponding results were published in Journal of Nanoscience and Nanotechnology,
2012, 2815-2824. Chapter 6 reports the evaluation of effects of different surfactants on
photovoltaic power conversion efficiency (PEC) of BHJ solar cells which based on MEH-
PPV and TiO2-based materials. The paper will be appeared in Green Processing and
Synthesis Journal, issue 2, 2015 March.
Finally, Chapter 7 summarizes the main finding of this work, presents some general
conclusions and recommendations for future work.
1
Chapter 1. Introduction
1.1. Overview
As the global energy demand of our rapidly growing population continues to
increase (it is expected to be doubled within next 50 years), a significantly larger fraction of
our energy supply will need to be sourced from renewable sources in the very near future.
Besides, excessive greenhouse gas emissions from carbon-based fuels, coupled with
environmental concerns, have placed a greater demand on the clean energy sector. To
decrease the use of fossil fuel, several alternative energies have been developed. Solar
energy, together with energies from wind, geothermal, hydropower, biofuel and biomass,
are called renewable energies. These energies provide a much cleaner and environmentally
benign source of power. However, they are still limited due to their high production cost
and low energy conversion efficiency. Given the increase in energy consumption, a number
of recent studies suggest that the direct use of hydrogen as a fuel or the direct use of solar
energy may provide much cleaner and less expensive fuel alternatives.
Hydrogen generation via water splitting by using solar energy in the presence of
semiconductor photocatalysts is considered as an ideal solution in the near future.1,2 The
sun and the wind contribute by around 5% the production of commercial H2 production,
primarily via water electrolysis, while the other 95% are mainly produced from fossil
fuels.3,4 Even though the renewable H2 production is not popular yet, current active
progress in photocatalytic water-splitting using semiconducting materials (especially TiO2,
ZnO, CdSe, CdS) offers a promising way for clean, low-cost and environmentally friendly
production of hydrogen by solar energy. 5–15
The development of solar cells has become an active field of research in recent
years. In the market, inorganic silicon thin film solar cells using nanocrystalline and
amorphous silicon have already achieved power conversion efficiency (PEC) of 12-15%.
Thick crystalline silicon solar cells achieved PEC of 22-24% and cadmium telluride (CdTe)
and copper indium gallium selenide (CIGS) inorganic solar cells also achieved PEC of 15-
20%.16–18 However, the production cost and toxicity of CdTe have become one of the major
disadvantages of those cell devices. Organic solar cells, especially hybrid bulk
2
heterojunction solar cells, which are based on the combination of conjugated polymers and
inorganic materials, have an advantage over its inorganic counterparts in its facile
production and flexibility in materials properties. The usage of bulk industrial production
techniques, such as doctor-blading, spray coating, inkjet printing, and roll to roll fabrication
techniques give the possibility to produce BHJ solar cells in automated manufacturing
units, which would reduce the production cost to meet the standard of the market.19–23
However, one fundamental disadvantage of polymer based solar cells is their low power
efficiency conversion, short life time and instability when exposed to the ambient
atmosphere as well as solar illumination. Recent progress in morphology control of
inorganic constituents in BHJs led to a significant improvement in their photovoltaic
efficiency. As key components, many inorganic semiconductors, such as TiO2, ZnO, PbS,
CuInSe2, CdSe, CdS, and many others, have been used for BHJs.16,24–28
Since its discovery in 1891, TiO2 has gained considerable attention in the energy
and environment sectors due to their brilliant prospects in photocatalysis, environmental
pollution treatment, batteries, sensors, ultraviolet blockers, pigments, surface coating,
paints, solar cells and in solar water splitting for the production of H2.29–37 Moreover, TiO2
NPs are easily produced, inexpensive and showed good stability under illumination in most
environment conditions. However, with the intrinsic wide band gap energy of 3.2 eV for
anatase phase, TiO2 allows to adsorb UV light only, which accounts for merely 5 % of the
incoming solar energy on the earth’s surface.38 To improve the performance of TiO2, it is
desirable to red-shift the absorption onset to also include the less energetic but more intense
visible part of the solar spectrum. A significant part of research on TiO2 has been
performed, and a number of reviews on various aspects of TiO2 have been published to
understand and improve the performance of TiO2 as well as TiO2-based materials.
1.2. Objectives of the thesis
The present Ph.D. work has been undertaken with the aim of studying the
performance and application of semiconductor materials based on TiO2 inorganic
semiconductor nanoparticles in photocatalytic H2 production via water splitting process as
well as in hybrid bulk heterojunction solar cells.
3
The first objective of this research was to synthesize and characterize TiO2 NPs with
shape and size controlled by combining different surfactants, including oleic acid, oley
amine and 6-aminohexanoic acid. Then, by using surface treatment techniques, the surface
of archived TiO2 NPs were treated with NOBF4 then deposited with CdS NPs in order to
form CdS/TiO2 semiconductor nanocomposites.
The second objective of this research was to prepare new non-noble metal-
nanocomposites (NCs) as highly efficient and stable in visible-light driven photocatalysis.
These NCs are composed of CdS/TiO2, and Ni clusters. An important advantage of TiO2
nanorod-based nanocomposites is that CdS NPs are evenly-dispersed on TiO2 nanorod
surface with strong bonding, and co-catalyst Ni clusters are selectively deposited on the
surface of these nanorods. This configuration could improve the efficiency of electron
transfer from the sensitized CdS NPs to TiO2 and then to Ni clusters, hence it enhances H2
production from water under visible light using ethanol as a sacrificial agent.
Finally, the third objective of the thesis was to develop a new concept of low cost
high performance HBSCs based on the conjugated polymer (MEH-PPV) and CdS/TiO2
semiconductor nanocomposites. The characterization of the developed solar cell properties,
including morphology, optical properties, and energy conversion efficiency were then
carefully done and discussed.
1.3. Summary of the Articles
This section presents an overview on the papers included in this thesis.
Paper 1: Thi Thuy Duong Vu, Frej Mighri, Abdellah Ajjia, Trong-On Do, “Synthesis
of Titanium Dioxide/Cadmium Sulfide Nanosphere Particles for Photocatalyst
Applications”. The paper reported semiconductor nanocomposites, which are composed of
TiO2 nanorods, CdS nanoparticles and Ni clusters. The synthesis consisted of a three steps
process: (i) surfactant-capped TiO2 nanorods with controlled length were synthesized in
autoclave using oleic acid and amino hexanoic acid as surfactants. By using a ligand-
exchange procedure, in which NOBF4 was used to replace the original surfactants,
hydrophilic NOBF4-TiO2 nanorods were obtained; (ii) the resulting nanorods were
deposited with CdS nanoparticles and (iii) then deposited selectively with Ni clusters (as
4
co-catalyst) on the nanocomposite surface. Under visible-light illumination, the generated
electrons from the conduction band of CdS of the obtained nanocomposite are transferred
to TiO2 via TiO2/CdS interface, then to the metallic Ni cluster. As a result, the charge
(electron/hole) separation was highly enhanced owing to the electrons to be separated from
the holes. This gives to the achieved Ni-TiO2/CdS nanocomposite a high photocatalytic
performance for the production of hydrogen (H2).
Paper 2: Thi Thuy Duong Vu, Frej Mighri, Trong-On Do, Abdellah Ajji, “Synthesis
of capped TiO2 nanocrystals of controlled shape and their use with MEH-PPV to develop
nanocomposite films for Photovoltaic applications”. This paper presents the synthesis
details of TiO2 nanoparticles of different morphologies using oleic acid and oleyl amine as
capping agents. Different shapes of NPs, such as nanospheres, nanorods, and
nanorhombics, were achieved. In order to develop nanocomposite thin films for
photovoltaic cells, these TiO2 NPs were carefully dispersed in 2-methoxy-5-(2’-
ethylhexyloxy)-p-phenylene vinylene (MEH-PPV) matrix. The properties of the
synthesized TiO2 NPs and MEH-PPV/TiO2 nanocomposites were characterized using
transmission electron microscopy (TEM), thermogravimetric analysis (TGA), UV-Visible
spectroscopy, and Photoluminescence techniques. Obtained results showed promising
properties for photovoltaic devices, especially solar radiation absorption properties and
charge transfer at the interface of the conjugated MEH-PPV matrix and TiO2 dispersed
NPs.
Paper 3: Thi Thuy Duong Vu, Frej Mighri, Abdellah Ajji, Trong-On Do, “The effect
of surfactants on the photovoltaic properties of hybrid bulk heterojunction solar cells based
on MEH-PPV and TiO2-based materials”. This paper focused on the evaluation of the
effects of different surfactants on photovoltaic power conversion efficiency (PEC) of BHJ
solar cells. Different kinds of surfactants were studied during the synthesis of TiO2
nanorods. The active layer, which is the most important layer of BHJ solar cells, was
fabricated by using MEH-PPV conjugated polymer and TiO2 nanorods or TiO2-based
nanocomposites. Solar cell devices characterization showed that the PEC increases by
around 3.2 times when using OA-6AHA-capped-TiO2 nanorods was used instead of using
5
OA-OM-capped TiO2 nanorods. Further optimization the PEC of BHJ solar cell devices
were done by enhancing the properties of the inorganic nano-semicondutors in the active
layer. CdS NPs were doped on the surface of TiO2 nanorods by using the similar method
presented in chapter 4. The PEC of devices using an active layer composed of MEH-PPV
and CdS/OA-6AHA-capped-TiO2 or CdS/OA-OM-capped-TiO2 nanocomposites were
respectively increased by around 12 and 17 times compared to devices with an active layer
respectively composed of MEH-PPV and only OA-6AHA-capped-TiO2 NPs or OA-OM-
capped-TiO2. It was also found that the PEC of BHJs using CdS/OA-6AHA-capped-TiO2
was around 2.3 times higher than that of BHJs using CdS/OA-OM-capped-TiO2
nanocomposites.
1.4. References
(1) Kudo, A.; Miseki, Y. Chem. Soc. Rev. 2009, 38, 253–278.
(2) Maeda, K.; Domen, K. J. Phys. Chem. Lett. 2010, 1, 2655–2661.
(3) Ni, M.; Leung, M.; Sumathy, K.; Leung, Y. Proc. Int. Hydrog. Energy
Forum 2004, 1, 475–480.
(4) Ni, M.; Leung, M. K. H.; Leung, D. Y. C.; Sumathy, K. Renew. Sustain.
Energy Rev. 2007, 11, 401–425.
(5) St. John, M. R.; Furgala, A. J.; Sammells, A. F. J. Phys. Chem. 1983, 87,
801–805.
(6) Bamwenda, G. R.; Tsubota, S.; Nakamura, T.; Haruta, M. J. Photochem.
Photobiol. A Chem. 1995, 89, 177–189.
(7) Sakthivel, S.; Shankar, M. V; Palanichamy, M.; Arabindoo, B.;
Bahnemann, D. W.; Murugesan, V. Water Res. 2004, 38, 3001–3008.
(8) Wu, N.-L.; Lee, M.-S. Int. J. Hydrogen Energy 2004, 29, 1601–1605.
(9) Alenzi, N.; Liao, W.-S.; Cremer, P. S.; Sanchez-Torres, V.; Wood, T.
K.; Ehlig-Economides, C.; Cheng, Z. Int. J. Hydrogen Energy 2010, 35,
11768–11775.
6
(10) Paunović, P.; Gogovska, D. S.; Popovski, O.; Stoyanova, A.; Slavcheva,
E.; Lefterova, E.; Iliev, P.; Dimitrov, A. T.; Jordanov, S. H. Int. J.
Hydrogen Energy 2011, 36, 9405–9414.
(11) Anpo, M. J. Catal. 2003, 216, 505–516.
(12) Steinfeld, A. Int. J. Hydrogen Energy 2002, 27, 611–619.
(13) Lin, Y.-G.; Hsu, Y.-K.; Chen, Y.-C.; Chen, L.-C.; Chen, S.-Y.; Chen,
K.-H. Nanoscale 2012, 4, 6515–6519.
(14) Ohno, T.; Bai, L.; Hisatomi, T.; Maeda, K.; Domen, K. J. Am. Chem.
Soc. 2012, 134, 8254–8259.
(15) Shao, M.; Ning, F.; Wei, M.; Evans, D. G.; Duan, X. Adv. Funct. Mater.
2014, 24, 580–586.
(16) Zeng, X.; Gan, Y. X. In Advances in Composite Materials for Medicine
and Nanotechnology; Attaf, B., Ed.; InTech, 2011; p. 648.
(17) Chopra, K. L.; Paulson, P. D.; Dutta, V. Prog. Photovoltaics Res. Appl.
2004, 12, 69–92.
(18) Dittrich, T.; Belaidi, A.; Ennaoui, A. Sol. Energy Mater. Sol. Cells 2011,
95, 1527–1536.
(19) Shah, V. Proc. IMAPS 37th Annu. Int. Symp. Microelectron. 2004.
(20) Shaheen, S. E.; Radspinner, R.; Peyghambarian, N.; Jabbour, G. E. Appl.
Phys. Lett. 2001, 79, 2996.
(21) Lee, H.; Leventis, H. C.; Moon, S.; Chen, P.; Ito, S.; Haque, S. A.;
Torres, T.; Nüesch, F.; Geiger, T.; Zakeeruddin, S. M.; Grätzel, M.;
Nazeeruddin, M. K. Adv. Funct. Mater. 2009, 19, 2735–2742.
(22) Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 394–412.
(23) Han, J.; Kim, H.; Kim, D. Y.; Jo, S. M.; Jang, S. ACS Nano 2010, 4,
3503–3509.
7
(24) Verma, D.; Ranga Rao, A.; Dutta, V. Sol. Energy Mater. Sol. Cells 2009,
93, 1482–1487.
(25) Huynh, W. U.; Dittmer, J. J.; Alivisatos, a P. Science 2002, 295, 2425–
2427.
(26) Wang, L.; Liu, Y.; Jiang, X.; Qin, D.; Cao, Y. J. Phys. Chem. C 2007,
111, 9538–9542.
(27) Roberson, L. B.; Poggi, M. A.; Kowalik, J.; Smestad, G. P.; Bottomley,
L. A.; Tolbert, L. M. Coord. Chem. Rev. 2004, 248, 1491–1499.
(28) Chandrasekaran, J.; Nithyaprakash, D.; Ajjan, K. B.; Maruthamuthu, S.;
Manoharan, D.; Kumar, S. Renew. Sustain. Energy Rev. 2011, 15, 1228–
1238.
(29) Bach, U.; Lupo, D.; Comte, P.; Moser, J. E.; Weissortel, F.; Salbeck, J.;
Spreitzer, H.; Gratzel, M. Nature 1998, 395, 583–585.
(30) Kuang, D.; Brillet, J.; Chen, P.; Takata, M.; Uchida, S.; Miura, H.;
Sumioka, K.; Zakeeruddin, S. M.; Grätzel, M. ACS Nano 2008, 2, 1113–
1116.
(31) Wang, H.; Oey, C. C.; Djurišić, A. B.; Man, K. K. Y.; Chan, W. K.; Xie,
M. H.; Leung, Y. H.; Pandey, A.; Nunzi, J.; Lavoisier, B.; Cedex, A.
Proc. 2005 5th IEEE Conf. Nanotechnol. 2005, 1–4.
(32) Wu, G.; Wang, J.; Thomas, D. F.; Chen, A. Langmuir 2008, 24, 3503–
3509.
(33) Holland, B. T. Science (80-. ). 1998, 281, 538–540.
(34) Landau, O.; Rothschild, A. Sensors Actuators B Chem. 2012, 171-172,
118–126.
(35) Pfaff, G.; Reynders, P. Chem. Rev. 1999, 99, 1963–1982.
(36) Águia, C.; Ângelo, J.; Madeira, L. M.; Mendes, A. Polym. Degrad. Stab.
2011, 96, 898–906.
8
(37) Grätzel, M. Inorg. Chem. 2005, 44, 6841–6851.
(38) Qiu, X.; Burda, C. Chem. Phys. 2007, 339, 1–10.
9
Chapter 2. Literature Review
2.1. Description of Solar Spectrum
The sun is a complex radiator with a spectrum that can be approximated by the
spectrum of a 5525K (5250°C) black body. This spectrum is then modified and affected by
many variation factors such as temperature across the sun’s disk, Fraunhofer absorption
lines, and the path length through the earth’s atmosphere. It was reported that, up to about
70% of energy within the light arriving is absorbed by clouds, oceans and land masses.
Figure 2.1 shows the solar radiation spectrum for direct light at both the top of the Earth's
atmosphere and at sea level, as a function of wavelength, where the red part is the energy
absorbed on sea level.
Figure 2.1. Solar radiation spectrum (Image created by Robert A. Rohde)6
When the sun is shining near its peak, with a relative modest 10% overall
efficiency, 1kW of electricity would be generated for every 10m2 of active area, which is
equal to the average electricity per residence. However, as the sun doesn’t shine at its peak
intensity for whole day, electrical storage devices are required for all solar cell devices.
10
Based on quanta theory, the particle of light was called photon. Photon energy, E, is
proportional to its frequency ν, and is given by the following Planck–Einstein equation:
hcE h
(2-1)
where h is Planck's constant (h= 6.626068×10-34 m2kg/s), and c is the speed of light
(c=299,792,458 m/s). According to the wavelength, the light is including ultraviolet
radiation (100 - 400 nm), visible light (400 - 700 nm) and infrared radiation (700 nm - 1
mm).
2.2. Titanium dioxide and its derivatives for alternative energy
2.2.1. Titanium Dioxide Nanoparticles
Titanium dioxide (TiO2), which is also called as titania or titanium (IV), was first
discovered in 1891, and was commercialized in 1961 as white pigment. Still then, it
remains as one of the most promising and interesting materials due to its high
photostability, high oxidation efficiency, non-toxicity, chemical inertness, biocompatibility,
environmentally friendly nature, and low cost production. Since 1972, the phenomenon of
photocatalytic splitting of water on TiO2 electrode by Fujishima and Honda was firstly
reported, then an exponential growth of research activities on TiO2 and its derivatives have
been seen in various applications, such as photovoltaics, photocatalysis, batteries, sensors,
ultraviolet blockers, pigments, surface coating, and paints.1–10
TiO2 belongs to the family of transition metal oxides. In nature, TiO2 has four
polymorphs: rutile (tetragonal), anatase (tetragonal), brookite (orthorhombic) and TiO2 (B)
(monoclinic).11 In addition, four more structures were synthesized under high pressure,
which are TiO2 (II) with a PbO2 structure, TiO2 (H) with a hollandite structure, baddelleyite
and cotunnite.12–15 Among them, the two polymorphs anatase and rutile are mostly
manufactured in chemical industry as crystalline materials. In fundamental studies, the
anatase and rutile TiO2 structures both have tetragonal structure but the distortion of
interconnected TiO6 octahedron is slightly larger for anatase phase.16 In addition, each
11
octahedron of TiO2 anatase is connected to 10 surrounding octahedrons, while those of
TiO2 rutile are connected to 8 surrounding octahedrons.2 These differences in lattice
structures are responsible for the mass densities and different electronic band gap energy
structures between these two forms of TiO2. The band gap energy of TiO2 anatase phase is
reported to be 3.2 eV, while the band gap energy of TiO2 rutile phase is 3.0 eV. This
relatively wide band gap means that both TiO2 forms could be stimulated only under UV
irradiation, and have low conversion efficiency under visible light.17,18
Figure 2.2. Crystal structures of TiO2 rutile and TiO2 anatase phase19
In general, the rutile is thermodynamically more stable than the anatase and
brookite. Both anatase and brookite phases are converted to rutile phase at high
temperature, around 750-800oC.20 TiO2 rutile phase is the mostly used form in the pigments
industry. However, the activity of TiO2 rutile phase as a photocatalyst under UV
illumination is generally very poor. Recently, according to Sclafani et al., these activities
12
could be improved by changing its preparation conditions.21 Differently, the TiO2 anatase
phase is metastable at low temperature, and it was reported to be preferred over other
polymorphs for photocatalyst as well as for photovoltaic applications because of its higher
surface area, higher electron mobility, lower dielectric constant and lower density.22,23
2.2.2. Coupled Colloidal Structures
Since TiO2 NPs can only be excited by high energy UV irradiation with a
wavelength shorter than 387 nm due to its relatively high energy band gap (3.2 eV), many
investigations confirmed that the coupled colloidal structures, in which TiO2 NPs is
coupled with different semiconductor particles, would extend the light absorption range of
TiO2 from UV to visible light. This leads to an increase in charge separation, hence they
result in higher activities in both photovoltaic and photocatalyst applications. Several
coupled colloidal structures of TiO2, such as CdS/TiO2, ZnO/TiO2, Fe2O3/TiO2, SiO2/TiO2,
SnO2/TiO2, Bi2S3/TiO2, WO3/TiO2, and MoO3/TiO2 have been reported.24–32 Among them,
the coupled structure of CdS quantum dot and TiO2 NPs has received the most attention.
CdS is a visible-light-driven photo-absorption with a narrow band gap of 2.4 eV. It
has an absorption band between 450 - 470 nm for CdS nanoparticles, and at about 515 nm
for the bulk crystalline CdS. Hence, CdS becomes an attractive candidate for photo-
absorption under solar light. However, CdS is subjected to photoanodic corrosion in
aqueous environment and has low quantum efficiency.33,34 To overcome this stability
problem and improve the photovoltaic and photocatalytic activity, CdS has been combined
with a wide band gap semiconductor, such as ZnO or TiO2, and this coupling gives reduced
photogenerated electron-hole recombination.
2.3. H2 Production via Photocatalysis Water Splitting
To replace or reduce the use of fossil fuels, another alternative ideal candidate for
the energy generation is hydrogen which has to be produced from water using natural
energies, such as sunlight. Hydrogen is the most abundant element and it exists in both
13
water and biomass. Its energy yield is high and is reported to be up to 122 kJ/g, which is
largely higher than that of other fuels, such as gasoline (40 kJ/g).
Hydrogen obtained via solar water splitting is generally categorized in four different
groups, which includes (i) water biophotolysis, (ii) organic biophotolysis, (iii) thermochemical
water splitting and (iv) photocatalytic water splitting. Thermochemical water splitting system
typically works at around 2000oC with the presence of a catalyst, such as ZnO35, in order to
perform water-splitting reaction, hence in large-scale production, this technique is often costly.
In water biophotolysis, hydrogen is generated from water in the presence of light by
cyanobacteria or green algae and special enzyme, such as hydrogenase or nitrogenase. This
technology presents some difficulties in designing and scaling up the bioreactor for the process,
and also in increasing the hydrogen yield production. Different from water biophotolysis,
organic biophotolysis generates hydrogen by photosynthetic anoxygenic bacteria under light
irradiation and anaerobic condition. Although organic biophotolysis gives a high hydrogen
yield, this reaction will generate CO2 as the by-product, hence it makes this technology less
environmentally friendly compared to other technologies.
Compared to those three above technologies, hydrogen generated from photocatalytic
water splitting has many advantages, such as production efficiency. Moreover, H2 production
from solar water splitting is environmentally friendly and has a great potential for low-cost
and clean hydrogen production. In addition, H2 can be easily distributed over large
distances through pipelines or via tankers. It can also be stored in gaseous, liquid or metal
hydride forms, and thus providing a huge market potential.
2.3.1. Working Principle
Photocatalytic water splitting to generate H2 using solar energy is defined as the
chemical reaction induced by photo-irradiation in the presence of semiconductor
photocatalysts, where the electronic structure of semiconductor plays an important role in
the reaction. When the semiconductors are excited by photons with energy higher than their
band gap energy level, electrons are promoted from valance band (VB) to conduction band
(CB). Separated electrons and holes migrate to the surface of the semiconductors and can
respectively reduce/oxidize the reactants adsorbed by semiconductors.
14
Figure 2.3. Principle of water splitting using semiconductor photocatalysts
Under the irradiation of light with energy greater than the bandgap of a
semiconductor photocatalyst, electrons in the VB are excited and jump into the CB,
resulting to the formation of an electron (e−)/hole (h+) pair. These photogenerated electrons
and holes can participate in redox reactions on the surface of the photocatalyst, unless they
recombine to give no net chemical reaction (Figure 2.3). To achieve overall water splitting,
the top of the VB of a semiconductor photocatalyst must be more positive than the
oxidation potential of H2O to O2 (0.82 V vs NHE at pH 7), and the bottom of the CB must
be more negative than the reduction potential of H+ to H2 (−0.41 V vs NHE at pH 7).
Therefore, the minimum photon energy thermodynamically required to drive the reaction is
equal to 1.23 eV.
2.3.2. State-of-the-art of H2 production based on TiO2 NPs and its derivatives
In a photocatalytic water splitting reaction, the photocatalyst plays a crucial role.
Most recently, extensive studies have been performed to split water under light irradiation,
but the number of photocatalyst materials known is yet limited, and the activity efficiency
is still low.36,37
15
Figure 2.4. Relationship between the band structure of semiconductors and the redox
potential of water splitting38
Metal oxides, such as TiO2 and ZnO, have been extensively studied as
photocatalysts for one-step water splitting, and some of them have achieved high quantum
efficiencies as high as several tens of percent; however, these materials are inactive in the
visible-light region. Beside, few metal chalcogenides, including CdS and CdSe, appear to
be suitable photocatalysts for photocatalytic water splitting. They exhibit band gap energies
sufficiently small to allow absorption of visible light and at the same time have conduction
and valence bands at potentials appropriate for water reduction and oxidation. However,
these chalcogenides are not stable in water, the S2− and Se2− anions are easier to oxidation
than water, causing the CdS or CdSe catalyst itself to be oxidized and degraded before
water.39,40
2.3.2.1. Modified-TiO2 NPs-based Photocatalysts for H2 Production Water Splitting
In general, TiO2 has been widely used as photocatalyst for photocatalytic water
splitting because it is stable, non-corrosive, environmentally friendly, abundant, and cost-
effective. More importantly, its energy levels are appropriate to initiate the water-splitting
reaction. However, pure TiO2 NPs cannot easily split water into H2 and O2 in the simple
aqueous suspension system due to the undesired electron-hole recombination reaction. In
g-C3N4
2.6
6eV
16
addition, the wide band gap (3.0 eV for the rutile phase and 3.2 eV for the anatase phase)
makes TiO2 only active under UV region, which only covers less than 5% of the solar
energy spectrum. So in order to utilize the visible light, which accounts for the major part
of the solar spectrum (~45%), extensive investigations have been carried out to extend the
photo-response of TiO2 into the visible light region. It is also important to prevent the
electron-hole recombination process during the photocatalytic water splitting. Effective
approaches to achieve this goal have included noble metal loading, metal-ion implanting,
non-metal doping, and organic dye sensitizing, and composite semiconductors. 41–53
Several noble metals, including Pt, Au, Pd, Rh, Cu and Ag, have been reported to be
very effective for enhancement of TiO2 photocatalysis in H2 production.41–47 These selected
noble metals normally have the Fermi levels lower than the CB of TiO2, which would
enhance the mobility of the photo-excited electrons transferred from the CB of TiO2 to the
metal particles.50 Anpo et al.47 found that the photocatalytic reactivity of semiconducting
TiO2 powder was dramatically enhanced by adding small amounts of Pt. By analyzing the
Electron Spin Resonance (ESR) signals to investigate the electron transportation, the results
indicated the occurrence of an effective electron transfer from TiO2 to Pt particles. As the
electrons accumulated on the noble metal particles then can be transferred to protons
adsorbed on the surface and further reduce the protons to hydrogen molecules, thus this
would be beneficial for water-splitting hydrogen production.48,49 Bamwenda et al.42 studied
the hydrogen production activity from water-ethanol solution using Au and Pt loaded TiO2
photocatalyst, which were prepared by deposition-precipitation, impregnation,
photodeposition and colloidal mixing methods. The roles of Au and Pt on TiO2 is to
generate the attraction and trapping of photogenerated electrons, the reduction of protons
and the formation and desorption of hydrogen. H2 yield was observed to be dependent on
the metal content on TiO2 and showed a maximum in the ranges 0.3–1 wt.% Pt and 1–2
wt.% Au. However, the overall activity of Pt samples was generally about 30% higher than
that of Au samples, which is probably a result of the more effective trapping and pooling of
photogenerated electrons by Pt and/or because platinum sites have a higher capability for
the reduction reaction. Sakthivel et al.43 investigated the photo-oxidation of leather dye,
acid green 16 in aqueous solution using Pt, Au and Pd deposited on TiO2 NPs as
photocatalyst. The photonic efficiency of Pt deposited on TiO2 is almost comparable to the
17
efficiency of Au/TiO2 but higher than that of Pd/TiO2. In addition, the effect of metal
contents on the photocatalytic activity was observed with metal deposition level of less than
1%. Increasing metal dopants resulted in a decrease of the surface area of TiO2, the
blockage of fine capillaries of parent TiO2 surface, a reduction of photon absorption by
TiO2, and electron-hole recombination, leading to a lower water splitting efficiency. In
additional, due to the high cost of Pt, Au, more research is needed to identify low-cost
metals with enhanced photocatalytic activity, such as Cu and Ag. Sakata et al.50 first
showed that Cu-TiO2 catalyst exhibit enhanced H2 production from a water/methanol
solution with photon energies within the visible-light region. Wu et al.44 found that, by
optimizing the loading of Cu, the hydrogen production activity was increased up to 10-fold
times.
Another common practice for modifying the bandgap of the photocatalyst is the so-
called metal ion doping practice, in which a small percentage of metal ions are incorporated
into the crystal lattice of the photocatalyst.51–55 Transitional metal ion doping and rare-earth
metal ion doping have been extensively investigated for enhancing photocatalytic activities
under visible light. Ikeda et al.52 synthesized transition-metal (V, Cr, Fe, Co, Mo, or W)
doped TiO2 which displayed a higher visible light absorption intensity and a higher water
splitting activity than pure TiO2 under visible light irradiation. Peng et al.53 carried out a
systematic study the effect of Be metal ions doped TiO2 on photocatalytic hydrogen
production in the presence of ethanol as electron donors. It was found that the doping of
metal ions could expand the photo-response of TiO2 into visible spectrum, and could
enhance the hydrogen production up to 75% compared to undoped-TiO2. However, in case
of deep doping, metal ions likely behave as recombination centers, which is unfavorable for
the photocatalytic reactions. Therefore, metal ions should be doped near the surface of TiO2
particles for a better charge transfer. Dholam et al.51 synthesized Cr- or Fe-ion-doped TiO2
thin films by radio-frequency magnetron sputtering and a sol–gel method to study hydrogen
generation by photocatalytic water-splitting under visible light irradiation. H2 production
rates were recorded higher with Fe-doped TiO2 (15.5 μmol·h-1) than with Cr-doped TiO2
(5.3 μmol·h-1) because Fe ions trap both electrons and holes thus avoiding recombination.
On the other hand, Cr can only trap one type of charge carrier. Other low-cost metals, such
as Ni and Co,54,55 were also found to be effective for photocatalytic activity enhancement.
18
These low-cost but effective metals are expected to be promising materials to improve
photocatalytic activities of TiO2 for practical applications. In recent years, many
researchers have focused on TiO2 with the double element co-doped TiO2, which shows
apparently higher photocatalytic activity than that of a single doped TiO2. Ryo et al. 56
synthesized (Ni, Ta or Ni, Nb) co-doped TiO2 photocatalysts, which displayed a higher
visible light absorption intensity and a higher water splitting activity than pure TiO2 under
visible light irradiation. Recently, Sun et al.54 prepared a single anatase phase of the Fe–Ni
co-doped TiO2 photocatalysts by alcohol-thermal method. The photocatalytic activities on
H2 evolution from water with ethanol as the sacrificial agent are studied in detail (Figure
2.5). The 5.0% Fe–4.0% Ni/TiO2 particles displayed a good absorption of the visible light,
and showed the average H2 evolution rate is 361.64 μmol·h-1·g-1, which is higher than pure
and single doped TiO2 as a result of the large amount of H+ and low recombination rate of
electron–hole pairs in the reaction systems. The mechanism of H2 evolution by water
splitting over Fe–Ni/TiO2 under visible light irradiation was proposed and showed on
Figure 2.5d
19
Figure 2.5. H2 evolution by water splitting over TiO2 catalysts (a) without any sacrificial
agents and UV light irradiation; (b) without any sacrificial agents and visible light
irradiation; (c) using ethanol as sacrificial agent and visible light irradiation; (d) Mechanism
of H2 evolution by water splitting over a Fe–Ni/TiO2 photocatalyst under visible light
irradiation
Beside the use of metal doping, anion doping is also used to improve the
photocatalytic activity under visible light. It was reported that the doping of anions (N, F,
C, S etc.) in crystalline TiO2 could shift its photo-response into visible-light spectrum.
Different from metal ions doped TiO2, anion doped TiO2 are less likely to form
recombination centers; hence they are more effective at improving the hydrogen production
activity. However, the ionic radius of S was reported to be too large to be incorporated into
the lattice of TiO2, and dopants P were found to be less effective as the introduced states
were so deep that photo-generated charge carriers were difficult to be transferred to the
d
d
20
surface of the catalyst. Therefore, S- and P-doped TiO2 were being less attractive as
photocatalyst for the hydrogen production compared to N-/C-doped TiO2.57 Wang et al.58,59
recently investigated that N-doped TiO2 film with a narrow band gap of 2.65 eV was
fabricated by RF magnetron sputtering and was successful applied as photocatalyst in
hydrogen production without the assistance of metal cathode, bias, or loading noble metal.
The H2 production rate of the N-doped TiO2 film was reported to be about 601 μmol·h-1·g-1,
far higher than that of the undoped TiO2 film and even about 50 times higher than that of
dispersive TiO2 P25 powder. Krengvirat et al.60 studied the incorporation of C with TiO2
and found that C-incorporated TiO2 photoelectrodes with nanotubular structures provided
higher photo-conversion efficiency (η) and hydrogen (H2) evolution capability than those
with irregular structures. The photoelectrode with an aspect ratio of ~142.5 had the
remarkable ability to generate H2 at an evolution rate of up to ∼508.3 μL.min-1.cm-2 and η
of ∼2.3%.
The combination of TiO2 and organic dyes sensitizing is a widely technique used in
photocatalyst systems. The benefits of adopting dye-sensitized photocatalyst systems
include the inhibition of charge recombination by improving electron-hole separation, the
increase of spectrum response range of photocatalyst, and a change in the selectivity or
yield of a particular product. Some of the frequently used dyes include Thionine (TH+),
Toluidine blue (Tb+), Methylene blue (MB), Phenosafranin (PSF), Rhodamin B (Rh. B),
Acridine orange (AO), Methyl violet, etc. Dhanalakshmi et al.61 carried out a study to
understand the effect of using [Ru(dcpy)2(dpq)]2+ as a dye sensitizer on photocatalytic
hydrogen production from water under visible light irradiation. It was found that hydrogen
production rate was enhanced by adsorbing dye molecules to the TiO2; moreover, the
hydrogen production rate did not further increase when additional Pt or dye loading beyond
the optimal values.
The use of composite semiconductors is another strategy to increase the
photocatalytic activity by achieving efficient charge separation and by expanding the
absorption spectrum of the photocatalyst at the same time. This strategy is based on the
coupling of a wide band gap semiconductor (non-oxide photocatalyst) with a narrow band
gap semiconductor having a more negative CB level. With the difference in energy gap
between two CB, the electrons can be injected from the smaller band gap semiconductor to
21
the larger band gap semiconductor; in our case from the CB of smaller band gap
semiconductor to the CB of TiO2. This would allow the extent in the absorption capacity of
the mixed photocatalyst. Successful coupling of TiO2 with other smaller band gap
semiconductors for photocatalytic water splitting hydrogen production under visible light
irradiation can be achieved when (i) the smaller band gap semiconductor should be able to
be excited by visible light; (ii) the CB of the smaller band gap semiconductor should be
more negative than that of TiO2; (iii) and finally, the electron injection should be fast and
efficient. Currently, coupled samples such as TiO2/CdS, Bi2S3/TiO2, TiO2/WO3,
TiO2/SnO2, TiO2/MoO3, and TiO2/Fe2O3 have been reported.
Sasikala et al.62 presented the TiO2/SnO2 mixed oxide in which SnO2 is in a
dispersed phase on TiO2, which have been synthesized by a polyol-mediated
route. Photocatalytic activity of these samples for hydrogen generation from water using
methanol as sacrificial reagent was studied under sunlight type radiation. The results
showed that mixed oxide enhanced the photocatalytic activity for hydrogen generation
compared to bare TiO2 and the activity decreases with increasing SnO2 concentration in
TiO2.
Similarly, it has been reported that coupling CdS with TiO2 could improve the
visible light response of TiO2.63–66 In this system, the photogenerated electrons move from
CdS to TiO2, whereas photogenerated holes remain in CdS. This charge-carrier separation
stops charge recombination, therefore improves the photocatalytic activity of TiO2. Optical
absorption spectra analysis showed that CdS/TiO2 could absorb photons with wavelength
up to 520 nm. Under visible light illumination (Xe lamp), CdS/TiO2 composite
semiconductors produced hydrogen at a higher rate than CdS and TiO2 used separately.63
22
Figure 2.6. Schematic illustration of the photo-induced charge injection process that occurs
upon excitation of the CdS component of a CdS/TiO2 colloid in the presence of a sacrificial
electron donor D.
Li et al.64 conducted photocatalytic hydrogen production using CdS/TiO2 composite
semiconductors, which consist of CdS nanoparticles incorporated into TiO2 nanotubes. The
composite photocatalyst exhibited an unprecedented high rate of hydrogen production with
an aqueous solution containing 0.35 M Na2SO3 and 0.25 M Na2S as sacrificial reagents,
and the apparent quantum yield for hydrogen production reached about 43.4% under visible
light irradiation (Figure 2.7).
Figure 2.7. (a) TEM image of CdS/TiO2 nanotube; (b) The average rate of H2 evolution and
(c) the amount of H2 evolved vs irradiation time on various photocatalysts: (a) CdS/TiO2
containing 13.44 wt% CdS ; (b) CdS/TiO2 containing 8.32 wt% CdS; (c) the physical mixture
of 20 wt% CdS/80 wt% TiO2 nanotube; (d) a pure CdS powder.
c b a
23
Wu et al.65 reported uniform and large-volume TiO2 nanowires, which were
successfully grown by a facile thermal treatment of titanium substrates assisted by KF in
the presence of a H2O vapor flow. The as-synthesized TiO2 nanowires were further
modified with hexagonal CdS QDs. The CdS/TiO2 composite photocatalyst exhibited a
very strong visible light response, and had the photocurrent density enhanced by over than
60% compared to the unmodified TiO2 nanowires, which is promising for photocatalytic
applications and hydrogen generation using the solar energy.
However, despite the improved activity of composite photocatalysts, most of the
narrow bandgap non-oxide photocatalysts involved may encounter photo-corrosion
problems in aqueous solution, which greatly confines their applications in hydrogen
production photocatalytic water splitting. To overcome these photo-corrosion problems, a
photocatalytic system called Z-scheme has been developed to generate H2 and O2
simultaneously. Basically, the Z-scheme consists of H2 and O2 photocatalysts to perform
water reduction and oxidation, respectively (Figure 2.8).67,68
Figure 2.8. Mechanism of Z-scheme system for water-splitting.68
Since electron donors are consumed in the photocatalytic reaction, continuous
addition of electron donors (sacrificial reagents or hole scavengers) is also required to
sustain hydrogen production. It can also help to control the electron-hole recombination
process.69,70 Various compounds, such as lactic acid, methanol, ethanol, ethylene diamine
tetra acetic acid derivative (EDTA), formaldehyde, Na2S, Na2SO4, or ions, such as I-, IO3-,
24
CN-, and Fe3+ have been used as sacrificial reagents.70–73 In their research, Nada et al.71
carried out a qualitative investigation to study the effect of different electron donors on
hydrogen production. It was found that the degree of hydrogen production was increased in
the following order: lactic acid < ethanol < methanol < EDTA. Li et al.74 added organic
pollutants, such as formaldehyde, oxalic acid and formic acid, as electron donors into the
photocatalytic reaction system. Decomposition of the organic pollutants was reported to be
consistent with hydrogen production.
Besides their use as sacrificial reagents, the addition of carbonate salts was also
found to improve the photocatalytic hydrogen production. Sayama et al.50 found that the
addition of carbonate salts to Pt-loaded TiO2 suspensions led to highly efficient
stoichiometric photocatalytic decomposition of liquid water into H2 and O2. These
carbonate species, which covered the TiO2 surface, can effectively suppress the back
reaction of water splitting to form water and alleviate the photoabsorption of oxygen on the
TiO2.
2.4. Photovoltaic Application
Solar cell or photovoltaic technology is one of many alternative renewable energies,
such as wind, biomass and water. Solar cells present three unique properties: i) direct
generation of electricity from solar radiation without the need of generators, ii) supplying
electrical power in form of portable modules, and iii) it is the only energy that can be
customized according to the need of uses. Thus, it is not surprising that since its first
discovery, photovoltaic solar cells (PV) are becoming a great potential solution to the
growing energy challenge and essential components of future global energy production.
However, the big drawback of current PV technologies is their rather high production cost
compared to other types of energies. They are about 10 times more expensive than energy
from fossil fuel and about 3 times more expensive than other renewable energies.
The term ‘photovoltaic’ is derived from the combined Greek words for light,
photos, and voltaic, named after Alessandro Volta. The development of photovoltaic cells
began with the work of the French physicist, Antonie-Cesar Becquerel, in 1839.75
Becquerel discovered the photovoltaic effect while experimenting with a solid electrode in
25
an electrolyte solution. He observed a small voltage and current when light fell upon the
electrode. About 50 years later, in 1877, Charles Fritts constructed the very first solar cell
device using a junction composed of semiconductor selenium layer and an ultra-thin, nearly
transparent layer of gold.76 However, the efficiency of the developed device transforming
the absorbed light into electrical energy was less than 1%. By 1927, solar cells made of
copper and the semiconductor copper oxide had been developed but still had energy
conversion efficiency of less than 1%. In 1941, with the invention of silicon solar cells
made by Russell Ohl, the energy conversion efficiency had been largely improved. In 1954,
Pearson et al. 76 opened a new era of semiconductor photovoltaic material when he obtained
a silicon solar efficiency of about 6%. In 1989, concentrator solar cells (types of cells
where sunlight is concentrated onto the cell surface by means of lenses) achieved an
efficiency of around 37% due to the increased intensity of the collected energy.
In the energy market, the competitive position of each solar technology is mainly
determined by the three factors: efficiency, lifetime and cost. As an alternative and
effective energy source, a solar cell must generate at least enough energy in its operating
lifetime in order to payback the financial and energy cost required to produce the cell. It is
estimated that an operating lifetime of a cell of about 20 years would be a workable value.
The operating lifetime may be affected by many external factors, such as physical damage,
corrosion, deterioration of cell support structures, etc. Also, it could be affected by internal
material-related factors like materials degradation, diffusion, photogeneration of defects,
etc. Especially, for those solar cells that are used in space, radiation damage is also a major
factor in degrading cell performance.
26
Figure 2.9. Current state of solar cell efficiencies (Reprint from National Renewable Energy
Laboratory (NREL) website)
Up-to now, solar cells have been classified into four generations based on the
materials and the processing technologies used to fabricate the devices. The most recent
generation is the fourth generation, which includes hybrid solar cells, where the electron
acceptor and transporter are grown in self-organized structures on a substrate, filled with a
conjugated polymer as hole transporter. In general, this type of solar cells is based on
inorganic semiconductor nanoparticles and polymer materials; hence it combines the
unique properties of inorganic semiconductor nanoparticles, high electron mobility, and
organic materials flexibility and their easy solution processing. Recently, various hybrid
bulk heterojunction solar cells have been reported. Highest efficiency for these devices had
been obtained with CdSe nanoparticles and polythiophene.77
Despite the low energy conversion efficiency, the strongest argument for the newest
generation solar cells is certainly their promising ultralow cost. The vision for solar cell
materials is based on thin film plastic carriers, using materials like solution-processable
organic and inorganic semiconductors, which are generally manufactured by coating and
printing techniques that are highly attractive from an economic standing point. The new
27
generation of solar cells are also attracted by their unique features like the potential to be
flexible and semitransparent and their potential to be manufactured in large area coating by
continuous printing processes.
2.4.1. Working Principle
In hybrid organic/inorganic solid-state devices, as the polymer is illuminated by
photons of energy higher than the band gap, electron-hole pairs are generated 78. An
electron is then promoted from the highest occupied molecular orbital (HOMO) to the
lowest unoccupied molecular orbital (LUMO), forming an exciton. The formed excitons
diffuse into the organic material and can reach the depletion layer where the internal
electric field can induce the separation of the charge carriers. The photogenerated holes can
thus migrate along the polymer, while the electrons can move along the nanocrystalline
network, then collected via the respective electrical contacts (Figure 2.10).
Figure 2.10. (a) Structure of BHJ solar cells (b,c) Scheme drawing of the working principle of
an organic photovoltaic cell.
28
In general, for a successful hybrid bulk heterojunction solar cell, four important
processes must be optimized to obtain a high conversion efficiency of solar energy into
electrical energy, which include the absorption of light, charge transfer and separation of
the opposite charges, charge transport and charge collection. In PVs, charge recombination
of the photogenerated electron-hole pairs is the major disadvantage in the use of conjugated
polymers as an active layer. In conjugated polymers, the diffusion length of excitons is
typically about 5−15 nm, so the light excitation occurring far from the interfaces will decay
without any charge transfer from the polymer to the nanocrystals.79,80 Charge separation
can be enhanced at the interface with a material of higher electron affinity, so that carriers
can be easily transferred because of the favorable energetic states in the junction energetic
diagram. Besides, a large interface between the two materials is also needed in order to
achieve an efficient photoconductivity.
To overcome this limitation, blending between conjugated polymers and nanosize
crystal oxides (especially particle sizes in the range of 2–10 nm) has been recently
proposed.81 This could create a large interface between the polymer matrix and the
dispersed nanoparticles, leading to an enhancement of charge transfer inside the
nanocomposite. This condition facilitates the diffusion of the photogenerated excitons to
the interface, where the separated charge carriers may travel to the respective contacts, thus
delivering current to the external circuit.81
2.4.2. Solar Cell Characteristic
2.4.2.1. Solar Irradiance Air Mass
The path length through the atmosphere is of fundamental importance. This path
length can be conveniently described in terms of air mass, mr. Basically, it is the ratio of the
path length of the sun rays through the atmosphere when the sun is at a given angle θ to the
zenith.
29
Figure 2.11. Air mass measurement
The equation that is provided to calculate the air mass is given by the following
equation:
1.6364
1
cos 0.50572(96.07995 )AM
(2-2)
The reference solar spectral irradiance AM0 (Air Mass 0) represents the irradiance
at the top of the atmosphere with a total energy of 1353W/m2. In characterization, an air
mass distribution of AM1.5 corresponds to the spectra power distribution observed when
the sun’s radiation is coming from an angle to over head of about 48.2o and the total energy
equals 1000W/m2.
An ideal and perfect solar cell that would be expected to cover the entire spectrum
and to convert all this energy into electricity would have an efficiency of 100 %. However,
in reality, depending on the semiconductor used, only a part of the solar spectrum is
covered and utilized ( Figure 2.1).
In addition to the direct irradiance, we also have to consider the diffused irradiance,
which is predominant on a cloudy day, and also the reflected irradiance. Reflected
irradiance is dependent on the albedo, which is a measure of the reflectivity of the Earth’s
surface. Fresh snow has an albedo of around 80 %, desert sand 40 % and grass between 5
and 30 %.
30
2.4.2.2. The short-circuit current (ISC) and the Open-circuited voltage (VOC)
The current to voltage curve of a solar cell has a very characteristic shape and can
be described by the mathematical models of an ideal or real photovoltaic generator.82 When
the p–n junction is illuminated by the sunshine, an electron–hole pair is generated by the
photons that have energy greater than the energy bandgap. The number of electron–hole
pairs is proportional to the light intensity. Because of the electric field in the depletion
region due to the ionized impurity atoms, the drift of electrons toward the n-side and that of
holes toward the p-side occur in the depletion region. This charge separation results in the
current flow from n- to p-side when an external wire is short-circuited (Figure 2.12).
Figure 2.12. Schematic illustration of carriers flow in short-circuited external circuit.
When the p- and n-sides are short-circuited, the current is called short-circuit current
ISC and is equals to the photogenerated current IL if the series resistance is zero. When the
p- and the n-sides are isolated, electrons move toward the n-side and holes move toward the
p-side, resulting in the generation of a current potential. The voltage developed is called the
open-circuit voltage VOC.
31
Figure 2.13. Illumination energy band diagrams of p–n junction in (a) the short-circuited and
(b) open-circuited current.
Assuming that the area of the solar cell is unity, the current characteristic of the
illuminated p–n junction is given by the following equation:
/
0( 1)qV nkT
SCI I e I
(2-3)
where I0 is the reverse saturation current (A).
In the open-circuit, which is obtained for I = 0, the voltage is given by equation:
0
ln( 1)SCOC
InkTV
q I
(2-4)
When the solar cell is operated under a condition that gives the maximum output
power, the voltage Vm and the current Im at the optimal operation point are shown in the
following Figure 2.14.
Figure 2.14. Current–voltage characteristics of p–n junction under illumination and
darkness.
32
In between these two points where in both cases the retrieved power is zero, there is
a working point, called the maximum power point, where the power that can be retrieved is
the highest and equals to: Pm = Vm Im. It is precisely at this point that the cells should be
used and the ratio between Pmax and the light intensity represents precisely the cell
efficiency. However, the curve, and thus this point, are not fixed and vary depending on
many parameters.
The relationship between the ratio of the maximum output power with the product
of ISC and VOC is defined as the fill factor (FF).
mIm
OC SC
VFF
V I
(2-5)
For a simple single-semiconductor photovoltaic model, the FF lies between 0.25 and
1.0.
2.4.2.3. Power conversion efficiency (PCE or ηe) and Quantum efficiency (QE)
The power conversion efficiency is the essential parameter for solar cell with
respect to energy power and cost. The overall efficiency ηe of a solar cell is calculated by
the following equation:
OC SCe
in
V I FF
P
(2-6)
where SCI is the short circuit current density in A/m2 (current for V=0) and inP (W/m2) is
the light incident solar radiation on the device as measured by a calibrated reference cell.
The last important parameter experimentally accessible is the variation of ISC with
the wavelength () of the incident light. This value is called external quantum efficiency
(EQE) or incident photon to collected electron (IPCE), which gives the ratio of the
collected charge carriers per incident photons:
33
1240 SC
in
IEQE
P
(2-7)
where is given in nanometer (nm), ISC in amperes per meter squared (A/m2), and Pin in
watts per meter squared (W/m2). As the short circuit current density (ISC) does not
necessarily increase linearly with the incident light power (Pin), the EQE values generally
depend on the Pin of the monochromatic light. The value of EQE can be further corrected
to take into account the different losses, like the reflection from the glass surface and the
absorption by different nonphotoactive layers involved in the device. The IQE is related to
the EQE by the reflectance (R) and the transmittance (TTr) of the solar cell by the following
equation:
(1 )Tr
EQEIQE
R T
(2-8)
2.4.3. State-of-the-art of BHSCs based on TiO2 NPs and their derivatives
In general, the bulk heterojunction ensures a higher interfacial area and thus an
optimal donor-acceptor contact. Over the last decade, bulk heterojunction solar cells (BHJs)
based on an interpenetrating network of electron donors and acceptors prepared using
solutions of conjugated polymers have become attractive for use in inexpensive large area
and low weight devices.83–85 The reported power conversion efficiencies of polymer
organic solar cells were about 4 to 5%; however, for practical applications, there are several
factors that limit the efficiency, which include the poor stability of the active layer under
the illumination, the poor overlap between the absorption spectrum of the polymer and the
solar spectrum, phase segregation and the low mobility of charge carriers, especially
electrons transportation.86–89 One potential solution is the use of PVs based on inorganic
semiconductor nanocrystals and conjugated polymers, due to the possibility of combining
the superior conductivity of inorganic nanoparticles and optoelectronic properties of
organic polymers. These systems are advantageous because they combine an inorganic
material, which performs the task of electron transport, with a conjugated polymer, which
is able to absorb the solar light as well as to conduct holes. Several kinds of inorganic
34
nanocrystals, CdSe, ZnO, CdS, and TiO2 are reported as charge acceptors.90–93 Also,
various conjugated polymers are good candidates for BHJ solar cells, such as (2-methoxy,
5-(2-ethyl-hexy-loxy)-p-phenyl vinylene) (MEH-PPV), poly(3 -hexylthiophene) (P3HT),
poly(alkyl-thiophenes) (PATs), poly[2-methoxy - 5 - (30, 70- dimethyloctyloxy )-p-
phenylenevinylene] (OC1C10-PPV) and poly(2-methoxy-5-(3,7-dimethyloctyloxy)-1,4-
phenylene- vinylene) (MDMO-PPV).94–97
Figure 2.15. Chemical structures of conjugated polymers used as donors in BHJ solar cells
Conjugated polymers were first reported in 1958 by Hoegel et al. 98 who proposed
its practical use as an electro-photographic agent. In the 1970s, it was discovered that
certain conjugated polymers, notably poly(sulphur nitride) and polyacetylene could be
made highly conductive in the presence of certain dopants. In 1982, Weinberger et al. 99
investigated the use of polyacetylene as the active material in an Al/polyacetylene/graphite
cell. The cell had a low open-circuit voltage of only 0.3 V and a low QE of only 0.3%.
Later, Glenis et al.100 investigated different polythiophenes.
2.4.3.1. TiO2 NPs and Conjugated Polymers-based BHJ solar cells
Nanostructured TiO2 has been studied as a photovoltaic material since the 1980s,
when the first observations of efficient photoinduced charge injection from dyes into TiO2
were reported 101. These studies established the basis for dye-sensitized solar cells102. The
sensitization of TiO2 by conjugated polymers or molecular films rather than by chemically
adsorbed dye monolayer became of interest in the late 1990s following the first reports of
photocurrent generation from conjugated polymer-based heterojunctions. Several studies
35
established that efficient photoinduced electron transfer from conjugated polymers into
TiO2 was possible 102–104. Compared to dye-sensitized solar cells, the solid nanostructured
TiO2-polymer solar cell has the advantage of utilizing the complete heterostructure for
exciton dissociation, potentially leading to thinner devices, since the entire polymer-filled
pore volume is available for exciton generation rather than only a dye monolayer at the
TiO2 surface. Further, the rigid structure of TiO2 offers better mechanical stability
compared to the organic PVs.
Examples of blends from TiO2 nanoparticles and conjugated polymers have shown
only moderate external quantum efficiencies of a few percent and short circuit currents of
tens of microamperes.105–108 Devices efficiencies have been reported recently for blends of
isotropic TiO2 particles with P3HT (η = 0.42%, AM 1, 100 mW/cm2),106 and elongated
TiO2 rods in MEH–PPV (η = 0.49%, AM 1.5, 100 mW/cm2).107
Despite promising EQE values, the power conversion efficiency values of devices
have were low compared to those achieved using the same polymers in polymer–fullerene
blends. The main factors limiting the performance include: incomplete distribution of the
nanoparticles into the conjugated polymer matrix; sub-optimum nanostructure morphology
where small quantity of the polymer volume lies within an exciton-diffusion length of the
interface; poor charge transport in the metal oxide component itself; and a less than
optimum photovoltage as a result of an unnecessarily large driving force for interfacial
charge separation.
Fabricating TiO2/polymer bulk heterojunction structures is an effective way to
improve the excitation dissociation in hybrid PV cells. Petrella et al.105,109 reported the
photoinduced charge transfer and the recombination of MEH-PPV and TiO2 nanorods
(NRs) capped with oleic acid (OLA), but the power transfer efficiency was not given.
Subsequently, Su et al.107 reported a PCE of 0.49% for MEH-PPV/TiO2 hybrid PV device
by inserting a thin layer of TiO2 NRs on the top of TiO2/MEH-PPV hybrid layer. For
further improve of the property of the hybrid polymer/TiO2 NRs, it is very important to
choose an appropriate ligand to exchange the OLA at the surface of TiO2 NRs.
36
(a) (b)
Figure 2.16. (a)The PL spectra from MEH-PPV: TiO2 layers of 70% TiO2 content with
different capped ligands. (b) The J-V characteristics of the PV devices under AM 1.5 solar
simulator (100 mW/cm2).
While studying the PV properties of bulk heterojunction devices from MEH-PPV
and TiO2 NRs modified by different ligands (OLA, n-octyl-phosphonic (OPA), thiophenol
(TP)) and TiO2 with thoroughly cleaned surface, Liu et al.108 reported that TiO2 NRs
modified with thiophenol (TP–TiO2) showed best PV performance. They obtained a fill
factor of around 0.34, an open-circuit voltage of approximately 0.70 V, and a power
conversion efficiency of 0.16% at AM 1.5 solar simulator (100 mW/cm2). Compared with
the P-TiO2, OPA- TiO2 and OLA-TiO2 NRs, the most effective exciton dissociation at
MEH-PPV/TP–TiO2 interface is due to the thiophenol capping, which is consistent with the
PL quenching ability (Figure 2.16a).
Table 2.1. Photovoltaic properties of hybrid MEH-PPV/TiO2 NRs capped by different ligands
TiO2 NRs Content of
TiO2 NRs
VOC
(V)
ISC
(mA cm-2)
FF
(%)
PCE
(%)
OLA-TiO2 NRs 70 0.75 0.051 28 0.016
OPA-TiO2 NRs 70 0.7 0.15 31 0.053
P-TiO2 NRs 70 0.5 0.365 37 0.096
TP-TiO2 NRs 70 0.7 0.456 34 0.157
37
Another strategy to improve the morphology of blend devices consists of using
elongated nanocrystals, such as rods through synthetic control of the nanocrystal shape.
The synthesis of TiO2 NPs normally occurs using hydrothermal or sovolthermal process.
Mixing these NPs into organic solvents generally leads to aggregate formation. For this
reason, only 20–40 nm TiO2 nanoparticles and conjugated polymers can be blended from
common organic solvents. Petrella et al.109 performed an extensive optical and photo-
electrochemical study of blended systems composed of organic-capped TiO2 crystals with a
spherical (d ~ 5 nm) or rod-like (d ~ 3–4 nm, l = 25–30 nm) morphology and MEH-PPV.
The blend exhibited higher photocurrents than those obtained with the single components,
in agreement with the enhancement of MEH-PPV photo-excited electron transfer to TiO2.
In general, the use of spherical TiO2 nanocrystals provided higher photo-electrochemical
responses than their rod-like counterparts. The reported results also suggested that such
MEH-PPV/TiO2 heterojunctions may be exploited as potential active layers in photovoltaic
and photo-electrochemical devices.
Figure 2.17. TEM of TiO2 nanorods (a) and dots (b), obtained by hydrolysis method at
100°C: (a) OLEA 35 g, TTIP 5 mmol, 2M TMAO 5 ml; (b) OLEA 35 g, EG 3.2 g, TTIP 1
mmol, TMAO 4 mmol 109.
Kwong et al.106 developed efficient solar cell devices by incorporating 60 wt% TiO2
and a ~100 nm thick TiO2:P3HT film spin coated from xylene. The obtained AM1 power
conversion efficiency was 0.06% for pure P3HT, 0.01% for 20%-30% of TiO2, 0.08% for
40% of TiO2, 0.27% for 50% TiO2, 0.42% for 60% TiO2, and 0.07% for 70% TiO2. For low
38
TiO2 concentration, the cell performance is inferior to that of the pure P3HT, while for the
TiO2 concentration of 50% and 60%, considerable improvement in AM1 power conversion
efficiency was obtained. However, for TiO2 concentration of 70% and higher, a good
quality uniform films could not be produced, so that the device performance worsened and
the efficiency became comparable to that of pure P3HT ( Figure 2.18a,b).
Figure 2.18. (a) J–V characteristics for P3HT:TiO2 cells for different TiO2 concentrations. (b)
The comparison between external quantum efficiency (EQE) for nanocomposite and pure
P3HT cells. (c) J–V characteristics of P3HT:TiO2 cells with P3HT:TiO2 films spin-coated
from different solvents. Inset:g shows the energy diagram of the devices.
Figure 2.18c shows J–V characteristics for P3HT:TiO2 cells with 60% of
nanoparticle concentration prepared from different solvents. The obtained AM1 power
conversion efficiencies are 0.03% for chloroform, 0.09% for THF, 0.17% for
chlorobenzene, and 0.42% for xylene. As the solvent evaporation rates influence the
surface morphology of polymer films 110, THF and chloroform have one order of
magnitude higher vapor pressure compared to xylene, and hence evaporate significantly
faster than xylene and chlorobenzene. In addition to solvent evaporation rate, the solvating
power may significantly affect the morphology, since a good solvent could lead to a more
extended polymer chain in solid state. Thus, a good solvent for P3HT with lower solvent
evaporation rate may favor better mixing of the components, resulting in improved exciton
dissociation and short circuit current density. The best and completed ITO/PEDOT:PSS/nc-
TiO2:P3HT/Al devices gave JSC= 2.759 mA/cm2, VOC = 0.44 V, FF= 0.396, and PEC=
0.424% using xylene as solvent.
39
2.4.3.2. Modified-TiO2 NPs and Conjugated Polymers-based BHJ solar cells
Several reports on various systems agree that the interface between the donor and
the acceptor plays a crucial role for processes of charge separation and recombination. The
commonly used solution processing for fully organic solar cells does not allow direct
control of this interface. In contrast, metal oxide nanostructures can be easily modified in
HSCs. Surface treatments, doping and the application of core-shell structures offer the
potential to increase charge separation yield, reduce the recombination and enhance both
VOC and ISC, resulting in more efficient PV devices.
Besides surface modifications, doping of metal oxides is a versatile method to
influence charge transport properties and the location of valence and conduction bands. By
doping ZnO with Mg, Olson et al. 111 were able to double the VOC for ZnO-P3HT hybrid
devices. An alloy of ZnMgO results in a reduced band offset and therefore allows an
increased potential. For Mg contents up to 25%, they were able to decrease the effective
work function from -4.2 eV to -3.9 eV resulting in an increase of VOC from 0.5 V to more
than 0.9 V. As mentioned above, similar effects have been reported for TiO2 doped with Ta
or N.112,113 For N-doping, Vitiello et al.114 were also able to show enhanced photoactivity of
TiO2 nanotubes in the visible range.
To simultaneously optimize both surfaces of the metal oxide and charge transport
properties of the nanostructure, core-shell morphologies have been considered. Metal oxide
nanostructures are coated with a thin layer of another material thus combining high
mobility of the inner material with high charge selectivity of the coating.
Furthermore, nanometer-sized crystals of inorganic semiconductors are another
interesting class of low-dimensional materials with useful optical and electronic properties.
When the size of the nanocrystal is smaller than that of the exciton in the bulk
semiconductor, quantum dot semiconductors (QDs), the lowest energy optical transition is
significantly increased in energy due to quantum confinement. The absorbed and emitted
energy can thus be tuned by changing the size of the nanocrystal. For example, by changing
the size from 6 to 2 nm, the energy gap can be tuned from 2.6 to 3.1 eV in CdS and from
2.0 to 2.6 eV in CdSe; hence it makes them interesting optical materials of in solar cell
application.90,94,115–118
40
At present, there are two main schemes for the deposition of NC sensitizers onto the
surface of another NC, which include (i) introducing organic linker-molecules that bridge
the colloids between two adjacent NCs, and (ii) growing NCs directly onto the oxide
surface via chemical bath deposition (CBD) or successive ionic layer adsorption and
reaction (SILAR) process119. For example, the former approach has been successfully
employed to the following couples: CdS-TiO2, CdSe-TiO2, CdS-ZnO, CdTe-CdSe, and
PbS-TiO2.120–125 The main drawback of this method is the presence of organic spacers
between the nanocrystals and oxide domains, which increase the tunneling barrier between
excited states of two semiconductors, causing a decrease in electron transfer probability. In
addition, a number of experimental works have demonstrated that organic linkers can also
serve as carrier traps, which further reduce the electron transfer rate.
Acharya et al.125 demonstrated a facile method for developing PbS-sensitized TiO2
films, which combines the benefits of the hot-injection colloidal route to the synthesis of
monodisperse PbS NC sensitizers. The processes allowed a high-temperature growth of the
PbS sensitizer directly onto the surface of TiO2, where a controlled tuning of PbS domain
sizes in the 2−20 nm range with an average dispersion of PbS diameters between 9 and
14% (Figure 2.19). Owing to a sequential two-step approach to the synthesis of TiO2/PbS
NCs, the size and the shape of TiO2 domains can be well tuned, which provides an
additional avenue for optimizing the transport of photoinduced carriers through an array of
TiO2/PbS NPs.
41
Figure 2.19. TEM of PbS/TiO2 nanocomposites fabricated under different conditions (a) high
concentration of OA results in the formation of small-diameter (d < 3 nm) PbS domains. (b)
The formation of single, large-diameter PbS NCs (d > 4) when the concentration of OA in the
solution is low. (c) The formation of multiple large-diameter PbS NCs per single nanorod
occurs when the concentration of OA is low, and concentrations of Pb and S precursors are
high. (d−f) HRTEM images of nanocomposites shown in (a−c), respectively.
The absorption profile in Figure 2.20b shows a small excitonic peak near 600 nm;
its spectral position agrees well with the expected band gap absorption in 2.3 nm PbS NCs.
This feature is absent in the spectrum of mixed PbS/TiO2 heterostructures (green curve),
possibly due to the overlapping absorption of large-diameter PbS NCs. Likewise, there is
no clear indication of band edge transitions in PbS/TiO2 NPs, comprising only large-
diameter PbS NCs (d = 4.2 nm). The absence of the excitonic peak in the latter case cannot
be attributed to the sample inhomogeneity alone because the size dispersion of PbS NCs in
these heterostructures is only 12% and is likely to arise from excitations of intermediate
states that exist at the interface of PbS and TiO2 domains.
42
Figure 2.20. Optical properties of PbS/TiO2. (a−b) Absorbance of PbS/TiO2 nanocomposites
representing several structural types. (c) Fluorescence intensity decay and (d) emission profile
of PbS/TiO2 heterostructures containing 4.2 nm PbS NCs.
PbS/TiO2 heterostructures comprising small-diameter PbS NCs showed very weak
or no fluorescence in the energy range corresponding to 1S(e)-1S(h) carrier recombination,
whereas NPs containing larger PbS domains (d > 4) generally produced somewhat stronger
emission in the near-infrared (λ = 800−1200 nm). This result is consistent with the
prediction that small-diameter PbS are more likely to inject excited carriers into TiO2,
leaving only a small fraction of excited population to decay via radiative channels.
In their study on the effect of Te content in mixed semiconductor CdSexTe1-x, Zhou
et al.124 synthesized a series of ternary tetrapodal nanocrystals of CdSexTe1-x with x = 0
(CdTe), 0.23, 0.53, 0.78, 1 (CdSe) and used them to fabricate hybrid nanocrystal/polymer
solar cell devices. It was found that, under identical experimental conditions, Voc, the Jsc
and power conversion efficiency (η) of the devices increased with increasing Se content in
the CdSexTe1-x nanocrystals. For convenience, CdSe0.23Te0.77, CdSe0.53Te0.47 and
CdSe0.78Te0.22 are respectively denoted as Na1, Na2 and Na3 in the following Table 2.2 and
Figure 2.21.
The photovoltaic properties of the hybrid CdSexTe1-x nanocrystals/MEH-PPV solar
cells were systematically investigated with different compositions of the nanocrystals in
ambient condition. The corresponding I–V curves are shown in Figure 2.21. From these
43
curves, the Voc, Jsc, the FF, and η were calculated (see Table 2.2). The device based on
CdTe without a Se component in the nanocrystals shows a relatively poor photovoltaic
property with Jsc of 0.024 mA/cm− 2, Voc of 0.33 V, and η of only 0.003%. When increasing
the Se content in the nanocrystals, the performance of the devices improves steadily. After
Te was totally substituted by Se, the efficiency increased to 1.13%, almost 400 times
greater than that without Se component, and the Jsc and Voc also significantly increased.
Figure 2.21. (LEFT) J–V curves of the BHJs based on MEH-PPV and the nanocrystals of Na1,
Na2, Na3 and CdSe under illumination of AM1.5, 80 mW/cm−2. (RIGHT) Band gap energy
level positions of MEH-PPV and NC determined by CV measurements.
Table 2.2. Photovoltaic properties of BHJs based on MEH-PPV and various nanocrystals
under the illumination of AM1.5, 80 mW/cm−2
Nanocrystals in the hybrid
solar cells
VOC (V) (JSC)
(mA cm-2)
FF η
(%)
CdTe 0.33 0.024 0.33 0.003
Na1 0.42 0.48 0.3 0.075
Na2 0.63 0.65 0.35 0.18
Na3 0.69 1.57 0.36 0.49
CdSe 0.69 2.86 0.46 1.13
Figure 2.21 (right) indicates that the highest occupied molecular orbital (HOMO)
and lowest unoccupied molecular orbital (LUMO) level positions of ternary nanocrystals
Na1, Na2 and Na3. When comparing the energy levels, all nanocrystal LUMO levels are
44
lower than those of the polymer's. However, only the CdTe HOMO level exceeds the
polymer HOMO level, indicating that change separation is forbidden between the two
components, and allowed for the other nanocrystals. Efficient charge separation requires
optimal energy differences between electron donor (polymer) and acceptor (nanocrystal).
Figure 2.21 (right) also shows that the energy gap between the nanocrystals and the
polymer increased with Se content for both HOMO and LUMO levels, explaining why the
photovoltaic properties of the devices improved with increasing the Se content in
CdSexTe1-x.
2.5. Reference
(1) Fujishima, A.; Honda, K. Nature 1972, 238, 37–38.
(2) (a) Fujishima, A.; Rao, T. N.; Tryk, D. A. J. Photochem. Photobiol. C Photochem.
Rev. 2000, 1, 1–21. (b) Fujiwara, A.; Matsuoka, Y.; Suematsu, H.; Ogawa, N.
Carbon N. Y. 2004, 42, 919–922. (c) Ishibashi, K.; Fujishima, A.; Watanabe, T.;
Hashimoto, K. J. Photochem. Photobiol. A Chem. 2000, 134, 139–142. (d) Ohko, Y.;
Ando, I.; Niwa, C.; Tatsuma, T.; Yamamura, T.; Nakashima, T.; Kubota, Y.;
Fujishima, A. Environ. Sci. Technol. 2001, 35, 2365–2368. (e) Ishibashi, K.;
Fujishima, A.; Watanabe, T.; Hashimoto, K. Electrochem. commun. 2000, 2, 207–
210. (f) Nakata, K.; Sakai, M.; Ochiai, T.; Murakami, T.; Takagi, K.; Fujishima, A.
Langmuir 2011, 27, 3275–3278. (g) Ishiguro, H.; Nakano, R.; Yao, Y.; Kajioka, J.;
Fujishima, A.; Sunada, K.; Minoshima, M.; Hashimoto, K.; Kubota, Y. Photochem.
Photobiol. Sci. 2011, 10, 1825–1829. (h) Liu, B.; Nakata, K.; Zhao, X.; Ochiai, T.;
Murakami, T.; Fujishima, A. J. Phys. Chem. C 2011, 115, 16037–16042. (i)
FUJISHIMA, A.; ZHANG, X.; TRYK, D. Surf. Sci. Rep. 2008, 63, 515–582. (j)
Nakata, K.; Udagawa, K.; Ochiai, T.; Sakai, H.; Murakami, T.; Abe, M.; Fujishima,
A. Mater. Chem. Phys. 2011, 126, 484–487. (k) Zhao, T.; Liu, Z.; Nakata, K.;
Nishimoto, S.; Murakami, T.; Zhao, Y.; Jiang, L.; Fujishima, A. J. Mater. Chem.
2010, 20, 5095.
(3) Bach, U.; Lupo, D.; Comte, P.; Moser, J. E.; Weissortel, F.; Salbeck, J.; Spreitzer,
H.; Gratzel, M. Nature 1998, 395, 583–585.
45
(4) Kuang, D.; Brillet, J.; Chen, P.; Takata, M.; Uchida, S.; Miura, H.; Sumioka, K.;
Zakeeruddin, S. M.; Grätzel, M. ACS Nano 2008, 2, 1113–1116.
(5) Wang, H.; Oey, C. C.; Djurišić, A. B.; Man, K. K. Y.; Chan, W. K.; Xie, M. H.;
Leung, Y. H.; Pandey, A.; Nunzi, J.; Lavoisier, B.; Cedex, A. Proc. 2005 5th IEEE
Conf. Nanotechnol. 2005, 1–4.
(6) Wu, G.; Wang, J.; Thomas, D. F.; Chen, A. Langmuir 2008, 24, 3503–3509.
(7) Holland, B. T. Science (80-. ). 1998, 281, 538–540.
(8) Landau, O.; Rothschild, A. Sensors Actuators B Chem. 2012, 171-172, 118–126.
(9) Pfaff, G.; Reynders, P. Chem. Rev. 1999, 99, 1963–1982.
(10) Águia, C.; Ângelo, J.; Madeira, L. M.; Mendes, A. Polym. Degrad. Stab. 2011, 96,
898–906.
(11) Marchand, R.; Brohan, L.; Tournoux, M. Mater. Res. Bull. 1980, 15, 1129–1133.
(12) Simons, P. Y.; Dachille, F. Acta Crystallogr. 1967, 23, 334–336.
(13) Latroche, M.; Brohan, L.; Marchand, R.; Tournoux, M. J. Solid State Chem. 1989,
81, 78–82.
(14) Sato, H.; Endo, S.; Sugiyama, M.; Kikegawa, T.; Shimomura, O.; Kusaba, K.
Science 1991, 251, 786–788.
(15) Nishio-Hamane, D.; Shimizu, A.; Nakahira, R.; Niwa, K.; Sano-Furukawa, A.;
Okada, T.; Yagi, T.; Kikegawa, T. Phys. Chem. Miner. 2009, 37, 129–136.
(16) Mo, S.; Ching, W. Phys. Rev. B 1995, 51, 13023–13032.
(17) Tanaka, K.; Capule, M. F. V.; Hisanaga, T. Chem. Phys. Lett. 1991, 187, 73–76.
(18) Leary, R.; Westwood, A. Carbon N. Y. 2011, 49, 741–772.
(19) Diebold, U. Surf. Sci. Rep. 2003, 48, 53–229.
(20) Wiggins, M. D.; Nelson, M. C.; Atta, C. R. MRS Proc. 2011, 398, 381.
(21) Sclafani, A.; Palmisano, L.; Schiavello, M. J. Phys. Chem. 1990, 94, 829–832.
(22) Carp, O.; Huismanb, C. L.; Rellerb, A. Prog. Solid State Chem. 2004, 32, 33–177.
(23) Reidy, D. J.; Holmes, J. D.; Morris, M. a. Ceram. Int. 2006, 32, 235–239.
(24) Srinivasan, S. S.; Wade, J.; Stefanakos, E. K. J. Nanomater. 2006, 2006, 1–7.
(25) Wang, B.; Kerr, L. L. J. Solid State Electrochem. 2011, 16, 1091–1097.
(26) Li, Y.; Xie, W.; Hu, X.; Shen, G.; Zhou, X.; Xiang, Y.; Zhao, X.; Fang, P. Langmuir
2010, 26, 591–597.
46
(27) Li, Y.; Lu, P.; Jiang, M.; Dhakal, R.; Thapaliya, P.; Peng, Z.; Jha, B.; Yan, X. J.
Phys. Chem. C 2012, 116, 25248–25256.
(28) Cowan, A. J.; Barnett, C. J.; Pendlebury, S. R.; Barroso, M.; Sivula, K.; Grätzel, M.;
Durrant, J. R.; Klug, D. R. J. Am. Chem. Soc. 2011, 133, 10134–10140.
(29) Ren, Y.; Chen, M.; Zhang, Y.; Wu, L. Langmuir 2010, 26, 11391–11396.
(30) Vinodgopal, K.; Bedja, I.; Kamat, P. V. Chem. Mater. 1996, 8, 2180–2187.
(31) Song, K. Y.; Park, M. K.; Kwon, Y. T.; Lee, H. W.; Chung, W. J.; Lee, W. I. Chem.
Mater. 2001, 13, 2349–2355.
(32) Bessekhouad, Y.; Robert, D.; Weber, J. . J. Photochem. Photobiol. A Chem. 2004,
163, 569–580.
(33) De, G. .; Roy, A. M.; Bhattacharya, S. S. Int. J. Hydrogen Energy 1995, 20, 127–
131.
(34) Li, W.; Li, D.; Chen, Z.; Huang, H.; Sun, M.; He, Y.; Fu, X. J. Phys. Chem. C 2008,
112, 14943–14947.
(35) Steinfeld, A. Int. J. Hydrogen Energy 2002, 27, 611–619.
(36) Kudo, A.; Omori, K.; Kato, H. J. Am. Chem. Soc. 1999, 121, 11459–11467.
(37) Zou, Z.; Arakawa, H. J. Photochem. Photobiol. A Chem. 2003, 158, 145–162.
(38) Kudo, A.; Miseki, Y. Chem. Soc. Rev. 2009, 38, 253–278.
(39) Williams, R. J. Chem. Phys. 1960, 32, 1505.
(40) Ellis, A. B.; Kaiser, S. W.; Bolts, J. M.; Wrighton, M. S. J. Am. Chem. Soc. 1977, 99,
2839–2848.
(41) St. John, M. R.; Furgala, A. J.; Sammells, A. F. J. Phys. Chem. 1983, 87, 801–805.
(42) Bamwenda, G. R.; Tsubota, S.; Nakamura, T.; Haruta, M. J. Photochem. Photobiol.
A Chem. 1995, 89, 177–189.
(43) Sakthivel, S.; Shankar, M. V; Palanichamy, M.; Arabindoo, B.; Bahnemann, D. W.;
Murugesan, V. Water Res. 2004, 38, 3001–3008.
(44) Wu, N.-L.; Lee, M.-S. Int. J. Hydrogen Energy 2004, 29, 1601–1605.
(45) Alenzi, N.; Liao, W.-S.; Cremer, P. S.; Sanchez-Torres, V.; Wood, T. K.; Ehlig-
Economides, C.; Cheng, Z. Int. J. Hydrogen Energy 2010, 35, 11768–11775.
47
(46) Paunović, P.; Gogovska, D. S.; Popovski, O.; Stoyanova, A.; Slavcheva, E.;
Lefterova, E.; Iliev, P.; Dimitrov, A. T.; Jordanov, S. H. Int. J. Hydrogen Energy
2011, 36, 9405–9414.
(47) Anpo, M. J. Catal. 2003, 216, 505–516.
(48) Subramanian, V.; Wolf, E.; Kamat, P. V. J. Phys. Chem. B 2001, 105, 11439–11446.
(49) Jakob, M.; Levanon, H.; Kamat, P. V. Nano Lett. 2003, 3, 353–358.
(50) Sayama, K.; Arakawa, H. J. Chem. Soc. Faraday Trans. 1997, 93, 1647–1654.
(51) Dholam, R.; Patel, N.; Adami, M.; Miotello, A. Int. J. Hydrogen Energy 2009, 34,
5337–5346.
(52) Ikeda, S.; Sugiyama, N.; Pal, B.; Marcí, G.; Palmisano, L.; Noguchi, H.; Uosaki, K.;
Ohtani, B. Phys. Chem. Chem. Phys. 2001, 3, 267–273.
(53) Peng, S.; Li, Y.; Jiang, F.; Lu, G.; Li, S. Chem. Phys. Lett. 2004, 398, 235–239.
(54) Sun, T.; Fan, J.; Liu, E.; Liu, L.; Wang, Y.; Dai, H.; Yang, Y.; Hou, W.; Hu, X.;
Jiang, Z. Powder Technol. 2012, 228, 210–218.
(55) Szabo-Bardos, E.; Czili, H.; Horvath, A. J. Photochem. Photobiol. A Chem. 2003,
154, 195–201.
(56) Niishiro, R.; Kato, H.; Kudo, A. Phys. Chem. Chem. Phys. 2005, 7, 2241.
(57) Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. Science 2001, 293, 269–
271.
(58) Wang, C.; Hu, Q.; Huang, J.; Wu, L.; Deng, Z.; Liu, Z.; Liu, Y.; Cao, Y. Appl. Surf.
Sci. 2013, 283, 188–192.
(59) Wang, C.; Hu, Q.-Q.; Huang, J.-Q.; Deng, Z.-H.; Shi, H.-L.; Wu, L.; Liu, Z.-G.;
Cao, Y.-G. Int. J. Hydrogen Energy 2014, 39, 1967–1971.
(60) Krengvirat, W.; Sreekantan, S.; Mohd Noor, A.-F.; Negishi, N.; Oh, S. Y.;
Kawamura, G.; Muto, H.; Matsuda, A. Int. J. Hydrogen Energy 2012, 37, 10046–
10056.
(61) Dhanalakshmi, K. Int. J. Hydrogen Energy 2001, 26, 669–674.
(62) Sasikala, R.; Shirole, A.; Sudarsan, V.; Sakuntala, T.; Sudakar, C.; Naik, R.;
Bharadwaj, S. R. Int. J. Hydrogen Energy 2009, 34, 3621–3630.
(63) So, W. Int. J. Hydrogen Energy 2004, 29, 229–234.
48
(64) Li, C.; Yuan, J.; Han, B.; Jiang, L.; Shangguan, W. Int. J. Hydrogen Energy 2010,
35, 7073–7079.
(65) Wu, G.; Tian, M.; Chen, A. J. Photochem. Photobiol. A Chem. 2012, 233, 65–71.
(66) Jang, J. S.; Ji, S. M.; Bae, S. W.; Son, H. C.; Lee, J. S. J. Photochem. Photobiol. A
Chem. 2007, 188, 112–119.
(67) Bard, A. J. J. Photochem. 1979, 10, 59–75.
(68) Sasaki, Y.; Iwase, a; Kato, H.; Kudo, a. J. Catal. 2008, 259, 133–137.
(69) Moon, S.; Mametsuka, H.; Tabata, S.; Suzuki, E. Catal. Today 2000, 58, 125–132.
(70) Abe, R.; Sayama, K.; Arakawa, H. Chem. Phys. Lett. 2003, 371, 360–364.
(71) Nada, A. A.; Barakatb, M. H.; Hameda, H. A.; Mohameda, N. R.; Vezirogluc, T. N.
Int. J. Hydrogen Energy 2005, 30, 687–691.
(72) Abe, R.; Sayama, K.; Domen, K.; Arakawa, H. Chem. Phys. Lett. 2001, 344, 339–
344.
(73) Galinska, A.; Walendziewski, J. Energy & Fuels 2005, 19, 1143–1147.
(74) Li, Y.; Lu, G.; Li, S. Chemosphere 2003, 52, 843–850.
(75) Becquerel, A. E. Comptes Rendus L’Academie des Sci. 1839, 9, 561– 567.
(76) Chapin, D. M.; Fuller, C. S.; Pearson, G. L. J. Appl. Phys. 1954, 25, 676.
(77) Liu, J.; Tanaka, T.; Sivula, K.; Alivisatos, a P.; Fréchet, J. M. J. J. Am. Chem. Soc.
2004, 126, 6550–6551.
(78) Wallace, G. G.; Chen, J.; Mozer, A. J.; Forsyth, M.; Macfarlane, D. R.; Wang, C.
Mater. Today 2009, 12, 20–27.
(79) Ginger, D. S.; Greenham, N. C. Synth. Met. 2001, 124, 117–120.
(80) Salafsky, J. Phys. Rev. B 1999, 59, 10885–10894.
(81) Nelson, J. Curr. Opin. Solid State Mater. Sci. 2002, 6, 87–95.
(82) Markvart, T.; Castaner, L. Practical Handbook of Photovoltaics: Fundamentals and
Applications; 1st ed.; Elsevier Science Ltd, 2003; p. 512.
(83) Brabec, C. J.; Sariciftci, N. S.; Hummelen, J. C. Adv. Funct. Mater. 2001, 11, 15–26.
(84) Mihailetchi, V. D.; Koster, L. J. a.; Blom, P. W. M. Appl. Phys. Lett. 2004, 85, 970.
(85) Kim, Y.; Choulis, S. a.; Nelson, J.; Bradley, D. D. C.; Cook, S.; Durrant, J. R. Appl.
Phys. Lett. 2005, 86, 063502.
(86) Padinger, F.; Rittberger, R. S.; Sariciftci, N. S. Adv. Funct. Mater. 2003, 13, 85–88.
49
(87) Reyes-Reyes, M.; Kim, K.; Dewald, J.; López-Sandoval, R.; Avadhanula, A.;
Curran, S.; Carroll, D. L. Org. Lett. 2005, 7, 5749–5752.
(88) Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Nat.
Mater. 2005, 4, 864–868.
(89) Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. Adv. Funct. Mater. 2005, 15,
1617–1622.
(90) Verma, D.; Ranga Rao, A.; Dutta, V. Sol. Energy Mater. Sol. Cells 2009, 93, 1482–
1487.
(91) Huynh, W. U.; Dittmer, J. J.; Alivisatos, a P. Science 2002, 295, 2425–2427.
(92) Wang, L.; Liu, Y.; Jiang, X.; Qin, D.; Cao, Y. J. Phys. Chem. C 2007, 111, 9538–
9542.
(93) Roberson, L. B.; Poggi, M. A.; Kowalik, J.; Smestad, G. P.; Bottomley, L. A.;
Tolbert, L. M. Coord. Chem. Rev. 2004, 248, 1491–1499.
(94) Kang, Y.; Kim, D. Sol. Energy Mater. Sol. Cells 2006, 90, 166–174.
(95) Ram, M. K.; Sarkar, N.; Bertoncello, P.; Sarkar, A.; Narizzano, R.; Nicolini, C.
Synth. Met. 2001, 122, 369–378.
(96) Wu, M.-C.; Lo, H.-H.; Liao, H.-C.; Chen, S.; Lin, Y.-Y.; Yen, W.-C.; Zeng, T.-W.;
Chen, Y.-F.; Chen, C.-W.; Su, W.-F. Sol. Energy Mater. Sol. Cells 2009, 93, 869–
873.
(97) Al-Ibrahim, M. Sol. Energy Mater. Sol. Cells 2004, 85, 13–20.
(98) Heeger, A. J. J. Phys. Chem. B 2001, 105, 8475–8491.
(99) Weinberger, B. R.; Akhtar, M.; Gau, S. C. Synth. Met. 1982, 4, 187–197.
(100) Glenis, S.; Tourillon, G.; Garnier, F. Thin Solid Films 1986, 139, 221–231.
(101) Vlachopoulos, N.; Liska, P.; Augustynski, J.; Graetzel, M. J. Am. Chem. Soc. 1988,
110, 1216–1220.
(102) O’Regan, B.; Grätzel, M. Nature 1991, 353, 737–740.
(103) Van Hal, P. A.; Christiaans, M. P. T.; Wienk, M. M.; Kroon, J. M.; Janssen, R. A. J.
J. Phys. Chem. B 1999, 103, 4352–4359.
(104) Arango, A. C.; Carter, S. a.; Brock, P. J. Appl. Phys. Lett. 1999, 74, 1698.
50
(105) Petrella, A.; Tamborra, M.; Cozzoli, P. .; Curri, M. .; Striccoli, M.; Cosma, P.;
Farinola, G. .; Babudri, F.; Naso, F.; Agostiano, A. Thin Solid Films 2004, 451-452,
64–68.
(106) Kwong, C. Y.; Djurišić, A. B.; Chui, P. C.; Cheng, K. W.; Chan, W. K. Chem. Phys.
Lett. 2004, 384, 372–375.
(107) Zeng, T.; Lin, Y.; Lo, H. Nanotechnology 2006, 17, 5387–5392.
(108) Liu, J.; Wang, W.; Yu, H.; Wu, Z.; Peng, J.; Cao, Y. Sol. Energy Mater. Sol. Cells
2008, 92, 1403–1409.
(109) Petrella, A.; Tamborra, M.; Curri, M. L.; Cosma, P.; Striccoli, M.; Cozzoli, P. D.;
Agostiano, A. J. Phys. Chem. B 2005, 109, 1554–1562.
(110) Strawhecker, K. E.; Kumar, S. K.; Douglas, J. F.; Karim, A. Macromolecules 2001,
34, 4669–4672.
(111) Olson, D. C.; Shaheen, S. E.; White, M. S.; Mitchell, W. J.; van Hest, M. F. A. M.;
Collins, R. T.; Ginley, D. S. Adv. Funct. Mater. 2007, 17, 264–269.
(112) Feng, X.; Shankar, K.; Paulose, M.; Grimes, C. a. Angew. Chem. Int. Ed. Engl. 2009,
48, 8095–8098.
(113) Yu, H.; Zhang, Z.; Han, M.; Hao, X.; Zhu, F. J. Am. Chem. Soc. 2005, 127, 2378–
2379.
(114) Sun, Y.; Fuge, G. M.; Fox, N. a.; Riley, D. J.; Ashfold, M. N. R. Adv. Mater. 2005,
17, 2477–2481.
(115) Vossmeyer, T.; Katsikas, L.; Giersig, M.; Popovic, I. G.; Diesner, K.; Chemseddine,
A.; Eychmueller, A.; Weller, H. J. Phys. Chem. 1994, 98, 7665–7673.
(116) Li, L.; Hu, J.; Yang, W.; Alivisatos, A. P. Nano 2001, 0–2.
(117) Greenham, N.; Peng, X.; Alivisatos, A. Phys. Rev. B 1996, 54, 17628–17637.
(118) Zhou, Y.; Riehle, F. S.; Yuan, Y.; Schleiermacher, H.; Niggemann, M.; Urban, G. a.;
Kruger, M. Appl. Phys. Lett. 2010, 96, 013304.
(119) Pathan, H. M.; Lokhande, C. D. Bull. Mater. Sci. 2004, 27, 85–111.
(120) Fujii, H.; Inata, K.; Ohtaki, M.; Eguchi, K.; Arai, H. J. Mater. Sci. 2001, 6, 527–532.
(121) Peter, L. M.; Riley, D. J.; Tull, E. J.; Wijayantha, K. G. U. Chem. Commun. 2002,
1030–1031.
51
(122) Mora-Sero, I.; Bisquert, J.; Dittrich, T.; Belaidi, A.; Susha, A. S.; Rogach, A. L. J.
Phys. Chem. C 2007, 111, 14889–14892.
(123) Kongkanand, A.; Tvrdy, K.; Takechi, K.; Kuno, M.; Kamat, P. V. J. Am. Chem. Soc.
2008, 130, 4007–4015.
(124) Zhou, Y.; Li, Y.; Zhong, H.; Hou, J.; Ding, Y.; Yang, C.; Li, Y. Nanotechnology
2006, 17, 4041–4047.
(125) Acharya, K. P.; Hewa-Kasakarage, N. N.; Alabi, T. R.; Nemitz, I.; Khon, E.; Ullrich,
B.; Anzenbacher, P.; Zamkov, M. J. Phys. Chem. C 2010, 114, 12496–12504.
53
Chapter 3. Experimental
Synthesis and evaluation of opto-electronic properties of semiconductor
nanoparticles as well as polymer nanocomposites under ambient conditions are the major
parts of this thesis. A number of experimental tools and techniques have been applied. For
the characterization of the developed materials and PV devices, a range of characterization
techniques, such as X-ray diffraction (XRD); X-ray photoelectron spectroscopy (XPS);
electron microscopy techniques including transmission electron microscopy (TEM),
scanning electron microscopy (SEM), and the complementary energy dispersive X-ray
spectroscopy (EDS); Fourier transform infrared spectroscopy (FTIR); ultraviolet-visible
spectroscopy (UV-vis); ζ-potential analysis; thermogravimetric analysis (TGA), J-V
characterization; gas sorption and gas chromatography (GC) have been applied. Also
several techniques include wet chemical processing, spin-coating and thermal evaporation
coating have been used in this thesis. In this chapter, the basic principles and overview of
the main experimental tools and techniques are presented.
3.1. Experimental Tools
3.1.1. Microscopy
While optical microscopy is the simplest small-scale materials characterization
technique, optical microscopy is limited in its resolution by the wavelength of light. With
new developments, where most of materials with particles in the micrometer or nanometer
scale, electron microscopy (EM) is a very important and irreplaceable technique for the
investigation morphology of particles. Basically, there are two main categories of electron
microscopes, divided on the basis of the imaging principles and specimen forms,
transmission electron microscopy (TEM), and scanning electron microscopy (SEM).
TEM was developed by Ruska and his co-workers in the 1930s. In TEM, the
instrument uses high energy electrons, which are accelerated to nearly the speed of light
under vacuum conditions. The electron beam behaves like a wave-front with wavelength
54
about a million times shorter than light-waves. When an electron beam passes through a
thin-section specimen of a material, electrons are scattered and then focused by a
sophisticated system of electromagnetic lenses to produce an image from the sample
volume on a screen or camera beneath the sample.1 Since its discovery, TEM has
incorporated many improvements of the electron gun, magnetic lens, power supply,
vacuum system, etc., to enhance the resolution up to 600K magnification.
On the other hand, in the 1950s a new version of EM with a completely different
principle and design from optical microscopy was developed and commercialized as SEM.
Similar to TEM, a beam of highly accelerated electrons is focused by magnetic lenses onto
a small region on the sample. However in SEM, secondary electrons or backscattered
electrons from the surface of the sample, caused by the electron beam as it is scanned
across the sample, are used for generating images of the sample surface.2 The electron
beams do not need to pass through the sample, therefore larger and thicker samples can be
investigated in SEM.
3.1.2. X-Ray Diffraction
X-ray diffraction (XRD) is an extensively useful technique to analyze and
characterize the crystallographic structure and different phase of nanocrystalline materials.
Its basic structure is composed of the X-ray detector and the x-ray source, which lie on the
circumference of a circle with the specimen (Figure 3.1a). The X-rays source is one which
generates X-rays by directing an electron beam of high voltage at a metal target anode
inside an evacuated X-ray tube. The detector is capable of counting the number of X-rays
photons of a particular energy for each angle 2θ, which is a proportional reflection of the
peak intensity.
55
(a) (b)
Figure 3.1. (a) Illustration of X-ray diffraction structure; (b) Schematic illustration of the
Bragg’s law.
The XRD pattern consists of a series of intensity peaks as a function angle. The
positions of the peaks depend on the crystal structure, includes shape and size of the unit
cell. In addition, each peak in the spectrum corresponds to a specific atomic spacing d. The
diffraction peaks are a result of Bragg’s law of diffraction, which says that the relationship
between the angles measured for each peak and the corresponding spacing d is given as:
nλ = 2dsinθ (3-1)
where n is the order of interference
is the wavelength of X-ray (nm)
d is lattice plane distance (nm)
is the angle of incidence (degree)
3.1.3. X-Ray Photoelectron Spectroscopy
X-ray Photoelectron Spectroscopy (XPS), also known as ESCA, is a powerful
technique that provides information about the nature of many different types of surface,
such as the actual elemental composition, and chemical state of surface.3 In basic principle,
XPS is based on the photoelectric effect arising when an X-Ray beam directs to the surface
of sample, the high energy of the X-ray photon is adsorbed by the core electron of an atom,
and then consequently cause the emission of electrons out of the surface.4 The kinetic
56
energies of ejected photoelectrons are not only characteristic of the atoms from which they
are emitted, but can also provide information on the chemical states of those atoms. The
emitted electron with the kinetic energy of Ek (eV) is referred to as the photoelectron. The
binding energy of the core electron is given by the Einstein relationship:
KE = h EB (3-2)
where h is the Planck’s constant (eV.s)
is the frequency of incident X-rays (s-1)
EB is the binding energy of the electron in a particular level (eV)
φ is the surface work function, which is about 4~5eV (eV)
Figure 3.2. The mechanism of photoelectron emission in XPS process
The exact binding energy of an electron depends not only upon the level from which
photoemission is occurring, but also upon the formal oxidation state of the atom and the
local chemical and physical environment change in either of these gives rise to small shifts
in the peak positions in the spectrum, so called chemical shifts. The peak of electron
emission from p, d and f orbitals, which are characterized by the orbital momentum number
and s the spin momentum number, hence is split into a doublet, with an energy difference
called spin-orbit splitting. The intensity ratio of the two components is determined by the
ratio of the multiplicity (2j+l) of the corresponding levels, where l is the orbital momentum
57
number, s the spin momentum number which is equal to 1/2 or −1/2, and j is a total
momentum number j=l+s. These features correspond to photoelectrons emitted from an
atom in which a second electron in a given orbital goes into an excited state as consequence
of the sudden change in the atom central potential produced by the photoelectron ejection.
The presence of these types of peaks may be quite useful for chemical state determinations.
3.1.4. Fourier Transform Infrared Spectroscopy
Fourier Transform Infrared spectroscopy (FTIR) is a powerful and well-known
application spectroscopy method implemented to identify the chemical function group in
materials.
An infrared (IR) spectrum is obtained by passing IR radiation with frequencies in
the range of 400 - 4000 cm-1 through a material, the IR causes vibrational and rotational
excitation of the molecular bonds in the material. Intensities of vibrations increase when
infrared radiation is absorbed. Radiation is absorbed by a molecule only if the frequency of
the radiation provides energy in the precise amount required by one of the bonds in the
molecule. Specific atomic groups tend to absorb infrared light at particular wavenumbers,
regardless of the response of other chemical bonds in the rest of the molecule.5,6 However,
each chemical bond with several vibrational modes can absorb several IR frequencies with
different intensities, include stretching and bending modes. Stretching absorptions usually
produce stronger peaks than bending, however the weaker bending absorptions can be
useful in differentiating similar types of bonds (e.g. aromatic substitution).
The plot of measured infrared absorbance versus wavenumber is called the IR
spectrum. Consequently, the infrared spectrum can be used as a fingerprint for molecules.
58
Table 3.1. Characteristic frequencies in FTIR7
Chemical bond Type of Vibrations Wavenumber (cm-1)
-OH Free 3650 - 3600
H-bonded 3400-3200
C=C Alkene 1680-1600
Aromatic 1600 and 1475
C=O
Aldehyde 1740-1720
Keton 1725-1705
Carbonxylic acid 1725-1700
Ester 1750-1730
Amides 1680 - 1630
C-O Alcohols, ethers, esters, carbonxylic
acids, anhydrides 1300-1000
C-N Amines 1350-1000
N-H Stretching
3350 and 3180 (primary);
3300 (secondary)
Bending 1640 - 1550
3.1.5. Ultraviolet-Visible Spectroscopy
UV-Visible (UV-vis) absorption spectroscopy is the measurement of the attenuation
of a photon beam after it passes through a sample or after reflection from a sample surface.8
UV-vis spectroscopy is useful not only for quantitative measurements but also for
characterization of absorption, transmission and reflectivity of a variety of materials.
Information extracted from UV-vis spectrum can further be used for estimation of the band
onset (band gap), color and the thickness of photocatalytic films
UV-vis uses light in the visible and adjacent near ultraviolet (UV) ranges. At these
wavelengths, molecules undergo electronic transitions, and excite electrons from their
ground states to higher energy excited states. Therefore, the energy absorbed depends on
the energy difference between ground state and excited state; smaller the difference, larger
the wavelength of absorption. The relationship between the energy absorbed in an
59
electronic transition and the frequency, ν (Hz), wavelength, λ (nm) and wavenumber, of
the radiation producing the transition is
hcE h h c
λ (3-3)
where h is Planks constant (6.626 x 10-34 J.s)
c is the speed of light (3.0 x 108 m/s)
λ is wavelength at absorption edge (nm)
E is the energy absorbed in an electronic transition in a molecule from a ground
state to a excited state (eV). The conversion factor: 1eV = 1.6 x 10-19 J
In crystalline semiconductor with an indirect band gap, the dependence of the
absorption coefficient, α, on the frequency, ν, can be approximated as followed:9
2 1 2(h ) ( )
h
i g
g
B E hE
h
(3-4)
where iB is the absorption constant for indirect transition
Eg is the optical Tauc band gap (eV)
The Tauc gap is found by extrapolating the linear part of a plot of (αE)(1/2) against
the photon energy h , the intercept of this line with the energy axis gives the value of the
Tauc optical gap Eg.6
3.1.6. Photoluminescence (PL)
Photoluminescence (PL) is a non-destructive optical technique used for the
characterization, investigation, and detection of point defects or for measuring the band-
gaps of materials. It involves a simple experiment and needs minimum sample preparation,
and can be performed at various temperatures, thus it can provide useful information on the
temperature dependence of fundamental electronic properties, such as the energy of the
band gap. Such studies are very important for the fabrication of electroluminescence
devices and in semiconductor research.10–12
60
PL involves a process in which a material absorbs photons from external source,
thus jumps to a higher electronic energy state, and then radiates photons, returning to a
lower energy state. A plot of emission against wavelength for any given excitation
wavelength is known as emission spectrum. If the wavelength of the excitation light is
changed and the emission from sample plotted against the wavelength of excited light, the
result is the excitation spectrum. The luminescent signals detected could result from the
band to band recombination, intrinsic crystalline defects, dopant, or other extrinsic defect
levels. The factors that can affect the quantitative accuracy are temperature effects which
affect the viscosity of the medium, pH effects, quenching etc.
3.1.7. Zeta (ζ) - Potential Analysis
Zeta (ζ) - Potential analysis is a technique for determining the surface charge of
nanoparticles in solution. It is also an important tool for understanding the state of the
nanoparticle surface and predicting the long term stability of the nanoparticles.
Figure 3.3. Zeta potential in colloid systems.
61
In a solid and liquid colloid system, an electric double layer (EDL) is formed at the
interface of the particle and the surrounding liquid environment by the surface charge of the
particle and counter ions in solution. The first EDL is called the stern layer, where the ions
are strongly bound to surface charge, whereas the second EDL is the diffuse region, where
the ions are less strongly attached to the surface. The stern layer contains ions opposite in
sign to the surface charge (Figure 3.2). When the particles move, the ion within stern layer
and on the closest part of the diffuse layer also move together with the particle, while the
rest of ions of the diffuse layer stays still with the bulk liquid. The plane that separates the
moving and the sessile layer of ions part is called the slipping or the plane of shear, and the
potential at this plane is called Zeta potential. Zeta potential is one of the main forces that
mediate inter-particle interactions, in which particles with a high zeta potential of the same
charge sign either positive or negative will repel each other. The most important factor that
affects zeta potential is pH of the solution.
3.1.8. Thermogravimetric analysis (TGA)
Thermogravimetric analysis (TGA) is an analytical technique used to measure the
mass or change in mass of a sample as a function of increasing temperature or time or both
in a controlled atmosphere.13 The change of mass generally occurs during sublimation,
evaporation, decomposition, chemical reaction, and magnetic or electrical transformation.
Therefore, the profile of the TGA thermogram would determine a material’s thermal
stability and its fraction of volatile components by monitoring the weight change that
occurs as a specimen is heated. It is especially useful for the study of polymeric materials,
including thermoplastics, thermosets, elastomers, composites, films, fibers, coatings and
paints.
The TGA measurement is normally carried out in air or in an inert atmosphere, such
as Helium or Argon. Sometimes, the measurement is performed in a lean oxygen
atmosphere (1 to 5% O2 in N2 or He) to slow down oxidation.
62
Figure 3.4. (a) Sample of TGA curve. Note the plateau of constant weight (region A), the mass
loss portion (region B), and another plateau of constant mass (region C); (b) Typical shape of
TGA where 1 - no change; 2 - desorption/drying; 3 – single stage decomposition; 4 - multi-
stage decomposition; 5 - as 4, but no intermediates or heating rate too fast; 6 - atmospheric
reaction; 7 – as 6, but product decomposes at higher temperature.
3.1.9. Brunauer–Emmett–Teller (BET) Specific Surface Area Analysis
The specific surface area (SSA) for powder materials can be determined by using
Brunauer-Emmett-Teller (BET) method. This measurement technique is based on the
physical adsorption of nitrogen or helium molecules onto the material surface at low,
constant temperatures to calculate surface area. In detail, this technique measures gas
uptake (adsorption) for increasing partial pressure over a powder sample and the release of
gas (desorption) at decreasing partial pressures. The collecting measurement data produce
adsorption isotherms which relate amount adsorbed to the relative pressure, from which the
SSA of the powder is obtained. The surface area is determined by using the BET
equation:14
63
o
o m m o
P/P 1 c 1 P
n(1 P/P ) n c n c P
(3-5)
where n is the amount adsorbed at the relative pressure P/Po
nm is the monolayer capacity
c is a constant related exponentially to the heat of adsorption in the first adsorbed
layer.
The BET equation gives a linear relationship between o
o
P/P
n(1 P/P ) and P/Po. The new
trend, still linear, can be interpolated with the BET equation:
oo
o m m
P/P 1 c - 1P/P
n(1 P/P ) cn cn
(3-6)
Based on the intercept between m
1
n c and slope
m
c 1
n c
, the values of nm and c could
be estimated. The value of c is derived as:
ads
vap
ΔH
RT
ΔH
RT
ec =
e
(3-7)
where adsΔH is the variation of adsorption enthalpy
vapΔH is the variation of vaporisation enthalpy.
R is the gas constant.
T is the absolute temperature.
Once c and nm are derived, the surface area can thus be calculated from the
monolayer capacity on the assumption of close packing as:
m mA = n α L (3-8)
where m is the molecular cross-sectional area, for N2, m = 0.162 nm2 at 77 K.
L is the Avogadro constant
64
3.1.10. Gas Chromatography Analysis
Gas Chromatography (GC) analysis is a useful analytical separation technique to
analyze volatile substances in the gas phase.15 To separate the compounds in GC, the
mixture to be separated and analyzed is vaporized and injected into a separation column
whose walls are coated by a polymer called stationary phase. The mixture components
traverse the length of the column in a mobile phase (i.e. carrier gas) at rates determined by
their retention in the stationary phase. If the column length and difference in the retention
times are sufficient, a complete separation of components is possible. The separated
components pass over a detector such as a flame ionization detector (FID), which generates
a signal called chromatogram. The position of peak maximum on the chromatogram
qualitatively identifies the component and the peak area is corresponding to the mass of the
component present in the sample.
3.2. Techniques
3.2.1. Wet Chemical Processing
Several different wet chemical processes have been applied either for cleaning or
for structuring or removal of the ITO layer on the substrate before fabricated solar cell.
These will be briefly summarized here.
3.2.1.1. Cleaning
The cleaning processes were performed before etching or depositing materials on
the surface of the ITO glasses in order to remove all the contaminations and dusts. ITO
glasses were cleaned with water and soap, and then followed by sonicator in acetone and
isopropanol for 5 mins, respectively.
65
3.2.1.2. Photolithography
Etching is used in solar cell device fabrication to chemically remove ITO layer from
the surface of a substrate during manufacturing. Etching is a critically important process
module, and every substrate undergoes many steps before it is complete. One of the most
comment methods etching that is used in solar cell fabrication is photolithography.
Photolithography is a process that uses light-activated chemicals to form patterns
on a surface. Fundamentally, a photoresist (S1813TM) layer was spun onto the substrate
surface and exposed to UV with desired pattern. The unmasked area of ITO will be etched
away by either wet etch and dry etch are suitable for ITO patterning according to different
application. In industry, wet chemical etch is widely used for ITO pattern due to its low
cost, excellent selectivity and large yield. Different ITO films require different wet etching
solutions. Conventional etchants are generally composed of strong acids, such as aqua
regia16, and halogen acid.17,18 According to some references mentioned, wet etching ITO
was performed in mixture or hydrochloric acid (HCl) and nitric acid (HNO3) solution for
30 mins.
66
Figure 3.5. Scheme of ITO etching process
3.2.2. Spin-coating
Spin coating is currently the predominant technique employed to generate thin and
homogeneous films with thickness of the order of micrometers and nanometers. Since
1978, spin coating was first studied for coating of paint and pitch.19 Currently, this process
has been widely used in the manufacture of integrated circuits, optical mirrors, color
television screens and magnetic disk for data storage.20
A typical spin process consists of a dispense step in which an excess amount of a
solution is placed on the substrate, which is then rotated at high speed in order to spread the
fluid by centrifugal force. And finally, a drying step to eliminate excess solvents from the
resulting film. A machine used for spin coating is called a spin coater. In order to get
homogeneous films, several different factors are important and have to be considered,
which are included spinning speed, spinning time, concentration of the solution, viscosity
of the fluid, and evaporation rate of the solvent.
67
3.2.3. Thermal evaporation
Thermal evaporation is one of the most widely used thin film deposition. The
vacuum thermal evaporation deposition technique consists of heating the materials until
evaporation. The material vapor then rises above and finally condenses in form of thin film
on the substrate surface and on the vacuum chamber walls. To heating the source material,
two heat sources are predominantly used, the filament source and the electron beam (e-
beam) source. The method that uses the filament source is called as filament evaporation,
wherein the heat source is a simple electrical resistive heat element or filament. There many
different physical configurations available for these filaments, including the “boats”, which
are basically thin sheet metal pieces of appropriate high temperature metals like tungsten
with formed indentations or troughs to place the material.
Figure 3.6. The vacuum thermal evaporation deposition system
In thermal evaporation, usually low pressures are used, about 10-6 or 10-5 Torr, to
avoid reaction between the vapor and atmosphere. The deposited film thickness would
range from angstroms to microns and are composed of a single material or layers of
multiple materials.
68
3.3. Reference
(1) Microscopy, T. E. III-Vs Rev. 2000, 13, 36–40.
(2) Vernon-Parry, K. D. III-Vs Rev. 2000, 13, 40–44.
(3) Van der Heide, P. X-Ray Photoelectron Spectroscopy; John Wiley & Sons, Inc.:
Hoboken, NJ, USA, 2011; Vol. 57, pp. 1–12.
(4) Venezia, A. M. Catal. Today 2003, 77, 359–370.
(5) Griffiths, P. R.; de Haseth, J. a. Fourier Transform Infrared Spectrometry; John
Wiley & Sons, Inc.: Hoboken, NJ, USA, 2007; p. 535.
(6) Murphy, A. Sol. Energy Mater. Sol. Cells 2007, 91, 1326–1337.
(7) Pavia, D. L.; Lampman, G. M.; Kriz, G. S. Introduction to Spectroscopy; Third Edit.;
Harcourt, Inc., 2008; p. 579.
(8) Kisch, H. Angew. Chem. Int. Ed. Engl. 2013, 52, 812–847.
(9) Serpone, N.; Lawless, D.; Khairutdinov, R. J. Phys. Chem. 1995, 99, 16646–16654.
(10) Herman, I. P. Opt. Diagnostics Thin Film Process. 1996, 619–636.
(11) Krawczyk, S. K. Encycl. Mater. Sci. Technol. 2001, 8397–8412.
(12) Shionoya, S. In Luminescence of Solids; Vij, D. R., Ed.; Springer US, 1998; pp. 95–
133.
(13) Coats, A. W.; Redfern, J. P. Analyst 1963, 88, 906.
(14) Brunauer, S.; Emmett, P. H.; Teller, E. J. Am. Chem. Soc. 1938, 60, 309–319.
(15) Visser, T. Handbook of Vibrational Spectroscopy; Chalmers, J. M.; Griffiths, P. R.,
Eds.; John Wiley & Sons, Ltd: Chichester, UK, 2006.
(16) Huang, C. .; Su, Y. .; Wu, S. . Mater. Chem. Phys. 2004, 84, 146–150.
(17) Van den Meerakker, J. E. A. M. J. Electrochem. Soc. 1995, 142, 2321.
(18) Scholten, M. J. Electrochem. Soc. 1993, 140, 471.
(19) Meyerhofer, D. J. Appl. Phys. 1978, 49, 3993.
(20) Yonkoski, R. K.; Soane, D. S. J. Appl. Phys. 1992, 72, 725.
69
Chapter 4. Synthesis of Titanium Dioxide/Cadmium Sulfide
Nanosphere Particles for Photocatalyst Applications
Thi Thuy Duong Vu a,b, Frej Mighri a,b,*, Abdellah Ajjib,c, Trong-On Doa,d
aDepartment of Chemical Engineering, Laval University, Quebec, QC, G1V 0A6 Canada;
bCenter for Applied Research on Polymers and Composites (CREPEC);
cDepartment of Chemical Engineering, École Polytechnique of Montreal, C.P. 6079, Montreal, QC,
H3C 3A7 Canada;
dCentre in Green Chemistry and Catalysis (CGCC).
Published in Industrial & Engineering Chemistry Research, 2014, 53(10), 3888–3897.
70
Abstract
Semiconductor nanocomposites, which are composed of titanium dioxide (TiO2)
nanorods, cadmium sulphide (CdS) nanoparticles (NPs), and Ni clusters, were synthesized.
The following steps were adopted: (i) surfactant-capped TiO2 nanorods with controlled
length were synthesized in autoclave using oleic acid and amino hexanoic acid as
surfactants. By using a ligand-exchange procedure, in which nitrosonium tetrafluoroborate
(NOBF4) was used to replace the original surfactants, hydrophilic NOBF4-TiO2 nanorods
were obtained; (ii) the resulting nanorods were deposited with CdS NPs and (iii) then
deposited selectively with Ni clusters (as cocatalyst) on the nanocomposite surface. Under
visible-light illumination of the nanocomposite, the generated electrons from the
conduction band of CdS are transferred to TiO2 via TiO2/CdS interface, then to metallic Ni
cluster. As a result, the electron/hole separation was highly enhanced leading to a Ni-
TiO2/CdS nanocomposite with high photocatalytic performance for the production of
hydrogen (H2).
71
Résumé
Des nanocomposites semiconducteurs, qui sont composés de nanotubes de dioxyde
de titane (TiO2), de nanoparticules (NPs) de sulfure de cadmium (CdS) et de clusters de Ni,
ont été synthétisés. Les étapes suivantes ont été adoptées: (i) des nanotubes de TiO2
recouverts de surfactant avec une longueur contrôlée, on été synthétisés dans une autoclave
en utilisant l’acide oléique et l’acide aminohexanoïque comme les surfactants. En utilisant
une procédure d’échange de ligands, dans laquelle le nitrosonium tetrafluoroborate
(NOBF4) a été utilisé pour remplacer les surfactants originaux, des nanotubes hydrophiles
de NOBF4-TiO2 ont été obtenus; (ii) les nanotubes obtenus ont été déposés avec des NPs de
CdS et (iii) déposés sélectivement ensuite avec les clusters de Ni (comme cocatalyseur) sur
la surface du nanocomposite. Avec un éclairage à la lumière visible du nanocomposite, les
électrons générés à partir de la bande de conduction du CdS sont transférés au TiO2 via
l’interface TiO2/CdS, ensuite au cluster métallique du Ni. En conséquence, la séparation
électron/trou a été fortement améliorée conduisant à un nanocomposite Ni-TiO2/CdS de
haute performance catalytique pour la production de l’hydrogène (H2).
72
4.1. Introduction
As one of the most abundant element with high energy efficiency, hydrogen
generated via solar water splitting has currently attracted a particular attention. Hydrogen
energy yield is reported up to 122 kJ/g, which is largely higher than that of other fuels, such
as gasoline (40 kJ/g)1. So H2 is presently considered as one on the future ideal fuel
candidates for the energy generation. Moreover, solar water splitting is environmentally
friendly and has a great potential for low-cost and clean hydrogen production. In addition,
H2 can be easily distributed over large distances through pipelines or via tankers. It can also
be stored in gaseous, liquid or metal hydride forms, and thus providing a huge market
potential.
In a photocatalytic H2 production reaction from water, the chemical reaction is
induced by photo-irradiation in the presence of a photocatalyst. With a relative narrow band
gap of 2.4 eV, CdS is one of the sulfide-based semiconductors, which have promising
applications in photocatalysis2–6. However, CdS alone shows very low H2 generation rates
due to the rapid recombination of photogenerated electrons and holes, which causes a lack
of H2 evolution sites. Good performances were mostly achieved in the presence of noble
metal co-catalysts, such as platinum (Pt), palladium (Pd) and nickel (Ni). Among various
strategies to improve the photocatalytic activity of CdS, the most efficient method is to
promote the charge separation of photogenerated electrons and holes by coupling CdS with
other semiconductors with adequate flat potentials, such as TiO27,8, zinc oxide (ZnO)9 or
graphene10,11. In such systems, electrons from the conduction band of CdS can be
transferred to other semiconductors or graphene, leading to improved electron-hole
separation, hence could enhance the generation rate of H2. TiO2 has been widely used as a
photocatalyst due to its high photostability and oxidation efficiency, and its abundance and
non-corrosives. It is also environmental friendly cost effective12. With proper band
structures, TiO2/CdS nanocomposite exhibits good properties in photocatalysis, leading to
an improved photo-production of H2 under visible light13–17.
Herein, we describe new non-noble metal-nanocomposites (NCs) as highly efficient
and stable visible-light driven photocatalysts. These NCs are composed of TiO2 nanorods,
CdS NPs, and Ni clusters. An important advantage of TiO2 nanorod-based nanocomposites
73
is that CdS NPs are evenly-dispersed on nanorod surface with strong bonding, and
cocatalyst Ni clusters are selectively deposited on the surface of these nanorods. This
configuration can improve the efficiency of electron transfer from the sensitized CdS NPs
to TiO2 and then to Ni clusters. As anticipated, Ni-TiO2/CdS nanocomposites developed in
the present work exhibit enhanced H2 production from water under visible light using
ethanol as a sacrificial reagent.
4.2. Experimental
4.2.1. Materials
All chemicals were used as received without further purification or distillation.
Titanium (IV) butoxide (TB, 97%), oleic acid (OA, 90%), 6-aminohexanoic acid (6AHA),
cadmium acetate dehydrate, thioamide and nitrosonium tetrafluoroborate solution (NOBF4)
were purchased from Aldrich. Absolute ethanol, N,N-dimethylformamide (DMF),
dichloromethane, hexane and toluene, were respectively purchased from Brampton Canada,
Fisher Scientific Canada, and Anachemia Canada. All of them were of analytical grade.
4.2.2. Synthesis of length-controlled TiO2 nanorods using oleic acid and 6-
aminohexanoic acid as surfactants
Capped-TiO2 nanorods were synthesized at low temperature using solvothermal
method. Oleic acid (OA), and 6-aminohexanoic acid (6AHA) were used as surfactants with
various molar ratio. Mixture of 1mmol TB, 6AHA, OA and absolute ethanol (EtOH) with
desired precursor molar ratios were mixed well and stirred for 30 mins under room
temperature before being transferred into a Teflon-lined stainless steel autoclave. The
autoclave also contained about 5-10 ml EtOH in order to keep equilibrium in the mixture
and to avoid any change in EtOH concentration during the crystallization process. The
synthesis process was set at 140oC for 18 h. After that, the autoclave was cooled down
74
slowly to room temperature, and samples were collected and washed several times using
ethanol and toluene.
4.2.3. Development of TiO2 nanorods by Ligand Exchange Reaction
Typically, 5 mL of dichloromethane solution of NOBF4 (0.01M) was added to
hexane solvent containing capped-TiO2 nanorods at room temperature. The mixture was
then gently shaken until the precipitation of the TiO2 nanorods. These nanorods quickly
become insoluble and are collected through centrifugation. Then, they were re-dispersed in
DMF hydrophilic solvent. To purify the TiO2 nanorods, DMF solutions were washed
through the addition of a mixture of toluene and ethanol 95% until precipitation occurs then
followed by centrifugation. This process was repeated few times. Finally, the collected
TiO2 nanorods were dried overnight in oven at 65oC to remove residual solvent molecules.
4.2.4. Synthesis of Colloidal Hybrid TiO2/CdS nanocomposite
A mixture of 4.5 mmol of NOBF4-capped-TiO2 nanorods dispersed in 10 ml of
DMF, and 9 mmol of Cadmium acetate dihydrate was stirred under room temperature for 2
h. Subsequently, 9 mmol of thioamide were added to the mixture and let under stirring for
three more hours in order to ensure a complete reaction. The precipitated TiO2/CdS
nanocrystals were washed few times using toluene and ethanol 95%, and then collected by
centrifugation.
4.2.5. Synthesis of Ni-TiO2/CdS by a Photodeposition method
Typically, Ni(NO3)2 was added to the solution containing TiO2/CdS. Because the
surface of TiO2 is negative, positive charge Ni2+ is selectively absorbed on the TiO2
surface, leading to the formation of TiO2/CdS-Ni2+. This solution is then illuminated with
visible light for 1.5h. As the potential of Ni2+/Ni is lower than the conduction band level of
75
TiO2, the electrons from the latter can effectively reduce Ni2+ species adsorbed on their
surface, then forming metallic Ni cluster18.
4.2.6. Characterization
Transmission electron microscopy (TEM) images of TiO2 nanorods, and hybrid
TiO2/CdS NCs were obtained on a JOEL JEM 1230 operated at 120 kV. Samples were
prepared as follows: a drop of a dilute toluene dispersion of nanocrystals were deposited
onto a 200 mesh carbon-coated copper grid then evaporated immediately at ambient
temperature. Elemental dispersive spectrum (SEM-EDX) analysis was obtained from a
JEOL 6360 instrument working at 3 kV. Powder X-ray diffraction (XRD) patterns of the
samples were obtained on a Bruker SMART APEXII X-ray diffractometer equipped with a
Cu Kα radiation source (λ = 1.5418 Å) in the 2θ range of 5–20° at a scan rate of 1.0°/min.
All samples were dried at 65°C overnight to eliminate guest solvent molecules on the
surface of particles before the XRD scan. Fourier transform infrared absorption spectra
(FTIR) were measured with a FTS 45 infrared spectrophotometer in the spectral range of
4000–400 cm–1. The thermal analyses of the as-made TiO2 nanorods, CdS NPs and hybrid
TiO2/CdS NCs were carried out at a heating rate of 10°C/min up to 900°C under an oxygen
flow using a Perkin-Elmer TGA thermogravimetric analyzer. The UV−visible spectra of
the nanostructures were recorded for the powder sample on a Cary 300 Bio UV−visible
spectrophotometer, and pure magnesium oxide (MgO) was used as a blank. ζ-Potential
measurements were performed with a Zetasizer Nano ZS in water at 25°C. Nitrogen
adsorption/desorption isotherms of the samples were obtained using with a Quantachrome
Autosorb-1 system, after degassing at 100 °C and 10-5 mmHg for at least 5 h. The specific
surface areas (SBET) of the samples were calculated from adsorption isotherm data using
the standard Brunauer−Emmett−Teller (BET) method. XPS characterization was carried
out in order to analyze the chemical composition of composite, as well as the electronic
state of Ni in the sample. XPS measurement was done in an evacuated ion-pumped
chamber at 1 × 10–9 Torr of Kratos Axis-Ultra instrument (UK). The X-ray source is a
monochromatic Al source (Al Kα, hv = 1486.6 eV) operated at 300 watts. The binding
energy of samples was measured by fixing an internal reference C1s peak at 285.0 eV. For
76
the separate constituents after background subtraction, all the peaks were deconvoluted by
means of standard CasaXPS software.
4.2.7. Photocatalysis characterization (Photocatalytic H2 evolution)
Before photocatalytic characterization, the surfactants adsorbed on samples were
eliminated. These samples were dried overnight at 65oC and used as such for photocatalytic
measurement. Visible light-induced H2 evolution was carried out in 80mL septum-sealed
glass vials. A mixture of 20 mg of sample and 3% Ni2+ were dispersed well in 27 mL of
aqueous solution containing ethanol (25 % wt.). The vial was deoxygenated using nitrogen,
and then placed in front of 300 W Xe-lamp with a 420 nm cut-off filter (FSQ-GG420) for
catalytic reaction. Gaseous products were then identified by collecting 0.5 ml of the gas in
the headspace of the vials. This gas was then analyzed by gas chromatography (GC) using a
thermal conductivity detector (TCD) for the quantification of H2 with N2 as the carrier.
4.3. Results and Discussions
Scheme 4.1 shows the procedure adopted for the synthesis of surfactant capped-
TiO2 nanorods by the hydrolysis of a titania precursor followed by a solvothermal reaction
in autoclave. First, an ethanol solution of titanium (IV) butoxide (TB) was modified by
hydrolysis with OA and 6AHA as surfactants. The hydrolysis process helped to yield three-
dimensional polymeric titania skeletons, which acted as the seeds for titania growth. In
order to obtain the desired TiO2 uniform sizes of particles, the subsequent solvothermal
process was carefully controlled with pre-setting the reaction time (18 hrs) and temperature
(140oC). It was observed that TiO2 nanorods were always achieved with the use of OA and
6AHA as surfactants.
77
Scheme 4.1. Sketch for the preparation of TiO2/CdS nanocomposites.
4.3.1. TEM, FTIR and BET characterization
Figure 4.1 shows TEM image of the obtained TiO2 nanorods before sonication. As
seen in the TEM image, these nanorods were attached together in a parallel configuration to
form big aggregation. This is different from the results obtained by Dinh et al.19 who
showed well dispersed TiO2 nanorods by using OA and oleylamine as surfactants. The
aggregation obtained in our approach may be due to the replacement of oleylamine by
6AHA surfactant.
Figure 4.1. TEM image of the synthesized TiO2 nanorods before sonication.
Figure 4.2 also shows TEM images of TiO2 nanorod samples obtained with
different molar TB:OA:6AHA ratios after few minutes of sonication. As seen in Figure 2,
78
by varying the molar ratio between TB, OA and 6AHA, different sizes of TiO2 nanorods
were observed. For a molar TB:OA: 6AHA ratio of 1:7:3, TiO2 nanorods of 3x40 nm were
achieved (Figure 2a). When the concentration of 6AHA was increased from 3 to 10 (e.g.,
from1:7:3 to 1:7:10), while the TB and OA concentrations kept the same, the shape of TiO2
nanorods did not change, however the length of the nanorod was decreased from 40 to 10
nm (Figure 2b). Hence, it could be assumed that the length of TiO2 nanorods is controlled
by the molar ratio OA:6AHA. Also, it should be mentioned that OA and 6AHA surfactants
have selective bindings to the different faces of TiO2. Joo et al.20 reported that OA binds
strongly to the TiO2 {001} faces, while 6AHA binding is more favored on {101} faces.
When the concentration of 6AHA is high (molar ratio OA:6AHA = 7:10), the strong
adhesion of 6AHA to the low surface energy {101} face, compared to the adhesion of OA
to {001} face, leads to a less progressive TiO2 growth along {001} direction to form TiO2
nanorods with short length. By decreasing the molar concentration of 6AHA, the adhesion
of 6AHA to the low surface energy {101} decreases while the adhesion of OA to {001} is
kept the same. The growth along {001} is then preserved, leading to longer TiO2 nanorod
shape21.
Figure 4.2. TEM images of synthesized TiO2 nanorods after sonication a) 3x40 nm nanorods
for TB:OA:6AHA molar concentration of 1:7:3, and b) 3x10nm nanorods for TB:OA:6AHA
molar concentration of 1:7:10.
79
Since OA and 6AHA were used as capping agents, the hydrophobic surfactant
capped- TiO2 nanorods were soluble in nonpolar hydrophobic solvents, such as toluene and
hexane. However, after being treated with dichloromethane solution of NOBF4, TiO2
nanorods precipitated immediately in hexane solvent after gentle shaking indicating that
NOBF4 has replaced the original hydrophobic surfactant capped to the nanorod surface.
This also indicates a dramatic change in surface properties of these NPs, from hydrophobic
to hydrophilic. As seen in Figure 4.3, it was observed that NOBF4 capped-TiO2 nanorods
were easily re-dissolved in DMF solvent as well as in water. This is considered as an
advantage during the deposit process of CdS on the surface of TiO2 nanorods since both
cadmium acetate and thioamide are well dissolved in DMF. A higher dispersion of the
initial precursors in the media (TiO2, Cd2+, S2-) increases the chance to achieve uniform
TiO2 nanorods with a higher dispersion of CdS on their surface.
Figure 4.3. (a) Surfactant-capped TiO2 nanorods dissolved in toluene; (b) TiO2 nanorods after
NOBF4 treatment dissolved in DMF.
In order to analyze the surface properties of TiO2 nanorods, FTIR characterization
was done for the samples before and after surfactant treatment. The corresponding results
are shown in Figure 4.4. FTIR spectra of the capped TiO2 nanorods before surface
treatment with NOBF4 and those of OA and 6AHA surfactants are shown in Figure 4.4a.
The small peaks at 3004 cm-1 were observed in the both FTIR spectra of OA and 6AHA
corresponding to the stretching of =C-H bond. The sharp vibrations bands at 2916 and 2857
cm-1 are attributed to the asymmetric and symmetric C-H bonds in methylene groups
80
(CH2), respectively.22The peaks at 1714 and 1282 cm-1 in the spectrum of OA are assigned
to C=O and C-O stretching and those appearing at 1463 and 936 cm-1 are due to in-plane
and out-of plane O-H. Compared to the commercial P25-TiO2, our synthesized TiO2
nanorods are identified by the additional peaks at 3004, 2922, 2853, and 1465 cm-1 due to
the presence of capping ligand on the surface. In addition, the peak appearing at 1608 cm-1
indicates the existence of carboxylic acid salt on the surface of surfactant capped-TiO2
nanorods, which is the result of the reaction between the OA surfactant and TiO2 during the
solvothermal process. Furthermore, a weak peak at 1041 cm-1 in the sample of surfactant
capped-TiO2 nanorods, which corresponds to that of C-N bonds in the amine groups,
proves the existence of amine on their surface (resulting from 6AHA surfactant).
Figure 4.4(b), shows the FTIR spectrum of the TiO2/CdS NCs after surface
treatment. No essential peak characteristic of -C-H stretching vibration at 2800-3000 cm-1
was observed after CdS deposition, as compared to that of the sample before deposition.
This could be due to NOBF4 treatment process where CdS deposition was able to remove
some residues of OA and 6AHA molecules attached to TiO2 nanorods surface (see Figure
4). As will be presented later (TGA characterization), this could explain the difference of
weight loss between TiO2 nanorods and TiO2/CdS nanocomposite. Furthermore, in
comparison with the FTIR spectrum of TiO2 nanorods before NOBF4 treatment, there is a
small peak at around 1000 cm-1, which is assigned to BF4- anions. Furthermore, no peak is
observed around 2100-2200 cm-1, which is normally ascribed to NO+. This is an indication
that surfactant exchange was between the organic ligands and inorganic BF4-, not with
NO+. The big peak at around 3050 cm-1 on the FTIR spectrum of TiO2/CdS, which is
similar to the peak observed for commercial TiO2 nanorods, is attributed to the water
absorbed on the surface of TiO2/CdS nanocomposite.
81
Figure 4.4. FTIR of (a) capped-TiO2 nanorod synthesized using OA and 6AHA as surfactants;
and (b) TiO2/CdS nanoparticles.
Figure 4.5 shows TEM image and Brunauer–Emmett–Teller (BET)
adsorption/desorption isotherm curves for the sample of TiO2/CdS nanocomposites. As
seen in Figure 5(a), TiO2/CdS nanoparticles were aggregated to form hollow nanospheres
82
with a uniform diameter of around 150 nm. When water sonication was performed, hollow
nanospheres were separated from each other. However, single nanospheres were not
separated into single NPs by sonication at low frequency ultrasound. Because the TiO2/CdS
hollow nanospheres are composed of a large number of nanoparticles, a high surface area
can be expected, as shown in Figure 4.5b. The BET specific surface area is 146 m2/g,
which is much higher than that of TiO2 nanorod (27.5 m2/g) and of CdS cubic (34.7 m2/g).
The surface area results are in agreement with the observation from the isotherm figure,
which shows that the isotherms of TiO2/CdS shift up compared to those of TiO2 and CdS.
The pore size distribution curves (see inset, Figure 4.5(b)) calculated from the
desorption branch of the nitrogen isotherms by the BJH method show a wide range of pore
diameters (from 5 to 237 nm) with a peak at a pore diameter of about 166 nm. Meanwhile,
a distinct hysteresis loop can be observed between adsorption and desorption branches, in
the range of 0.8 to 1 nm, which is an indication of mesostructured the TiO2/CdS
nanospheres23,24.
83
Figure 4.5. (a) TEM image of TiO2/CdS nanocomposite, and (b) BET characterization of TiO2,
CdS, and TiO2/CdS nanocomposite with the inset is their corresponding pore size distribution
84
4.3.2. XRD characterization
Figure 4.6 shows XRD patterns of TiO2 nanorods, CdS NPs and TiO2/CdS
nanocomposite. XRD patterns of TiO2 nanorods exhibit strong diffraction peaks at 25° and
48°, indicating TiO2 anatase phase. All peaks were in good agreement with the standard
spectrum for TiO2 (JCPDS no: 88-1175 and 84-1286). Meanwhile, it is known that CdS
NPs possess the hexagonal phase with (002) as the preferential crystalline plane with two
main peaks at 28.3o (101 planes) and 48.1o (103 planes)25, while the cubic phase has three
main peaks at 26.5o (111 planes), 43.9o (220 planes) and 51.9o (311 planes)13. Hence, with
those peaks shown in the XRD pattern of CdS NPs, we can conclude that CdS NPs are in
cubic phase.
XRD patterns of the TiO2/CdS nanocomposites confirm the presence of CdS and
TiO2. However, when mixed with high concentration of CdS NPs, the intensity of the
diffraction peaks at 48° was very low, which could be due to the attachment of CdS on the
surface of TiO2 nanorods. In the XRD spectrum of our TiO2/CdS nanocomposite, three
broad and symmetric peaks were observed at 2Ө = 26.5o (111 planes), 43.9o (220 planes),
and 51.9o (331 planes) corresponding to the cubic phase of CdS. The absence of planes
referring to hexagonal structured CdS indicates the presence of only cubic CdS
nanoparticles in the sample. Furthermore, the broadening of the peaks is due to the CdS
nanosize in the TiO2/CdS nanocomposite.
85
Figure 4.6. XRD characterization of a) TiO2 nanorod b) TiO2/CdS nanocomposite.
86
4.3.3. XPS and SEM-EDX characterization
Figure 4.7. (a) XPS characterization of Ni-TiO2/CdS nanocomposite (b) High-resolution XPS
of Ni
The XPS survey spectrum (Figure 4.7(a)) shows the existence of Ti, O, Cd, S, Ni
and C elements in the sample. Also, the high-resolution XPS spectrum of Ni 2p3/2 peak at
856.4 eV confirms the presence of Ni in the sample (Figure 4.7b), mainly from NiO.26,27
The formation of NiO could be due to the photo-induced electrons in the conduction band
87
of TiO2 transferred to Ni2+ clusters causing the reduction of a part of Ni2+ clusters to NiO
atoms due to their instability in the air.27 In addition, the Ti2p and O1s peaks are
respectively found at 458.6 and 530.95 eV, which are compatible with the assignment to
TiO2. Cd3d (405.1eV) and S2p (161.95 eV) peaks are reported to be compatible with CdS.
The observation of C1s element is due to the surfactant capped on the surface of the
sample, and also from the adventitious hydrocarbon in the XPS instrument itself. The XPS
peak at 686.91eV is ascribed to F- ions coming from NOBF4 during surfactant treatment
process.
Figure 4.8. SEM-EDX characterization of Ni-TiO2/CdS nanocomposite
The presence of Ni in the sample was also confirmed from the SEM-EDX elemental
analytical spectrum (Figure 4.8). This spectrum shows that the intensity of Ni peak is small
compared to the other elements. This is due to the small amount of Ni cluster deposited on
the TiO2/CdS composite, which is only 3% wt.
4.3.4. UV/Vis and Photoluminescence (PL) characterizations
The optical properties of TiO2 nanorods before surface treatment and TiO2/CdS
nanocomposites were investigated by UV/Vis absorption and photoluminescence (PL)
88
characterization techniques. The UV-visible absorption spectrum (Figure 4.9(a)) has been
performed to measure the photo-response of TiO2 nanorods after their loading with CdS.
The absorption edge for anatase TiO2 nanorod is approximately 380 nm (3.12 eV), which
has no significant absorption in visible-light region. However, the spectrum of CdS exhibits
a broad absorption band around 530 nm (2.32 eV), indicating the effective photo-
absorption property in the visible region. Basically, the spectrum of TiO2/CdS
nanocomposite is a combination of TiO2 and CdS spectra. The absorption edges of
TiO2/CdS nanocomposite is approximately 547 nm (2.23 eV), which is around 15 nm red-
shift than that of CdS. This probably results from the coupling between CdS and TiO2.
Figure 4.9. (a) UV-Vis spectra of TiO2, CdS and TiO2/CdS (b) Photoluminescence (PL)
emission spectra under excitation at a wavelength of 380 nm for CdS and TiO2/CdS
nanocomposite.
89
Figure 4.9(b) shows the PL emission spectra for CdS and TiO2/CdS nanocomposites
at room temperature under light excitation at a wavelength of 380 nm. According to the PL
of both CdS and TiO2/CdS sample, PL peak of TiO2/CdS exhibited much weaker intensity
than of that of CdS. The decrease in PL intensity indicates a better PL quenching, which
also indicates a decrease in light emission of the material or a coupling between CdS and
TiO2 with a better charge transfer between these two nanoparticles. As discussion above,
the efficient charge transfer from CdS to TiO2 conduction band could effectively separate
the photo-induced electrons from holes in the CdS semiconductor. Thus, the decrease in PL
intensity also could be ascribed to the lower recombination probability of photo-induced
electrons and holes in TiO2/CdS nanocomposites.28
4.3.5. Thermal Gravimetric (TGA) and ζ-potential characterization
Figure 4.10. TGA characterization of (black) TiO2 nanorods (blue) CdS NPs (red) TiO2/CdS
nanocomposite.
90
Thermal gravimetric characterization of synthesized capped TiO2 nanorods, CdS
NPs, and TiO2/CdS nanocomposites are summarized in TGA curves of Figure 4.10, which
were obtained at a heating rate of 10 ºC/min under O2 atmosphere. All the three curves
show an initial weight loss starting at around 50oC, which could be attributed to the water
absorbed on the surface of the nanoparticles. For TiO2 nanorods, the most significant
weight loss obviously occurred between 200 and 480oC and corresponds to OA surfactants.
For higher temperatures (>480oC), the very small weight loss could be attributed to the
decomposition of residual product traces that forms a sheath over the TiO2 nanorods. For
CdS NPs, the TGA spectrum shows that the main mass decrease occurred below 400oC,
which could be mainly due to the evaporation of residual solvent. A non-negligible gain in
mass was also observed between 400 and 750°C, which is an indication of the formation of
cadmium sulphate (CdSO4) through the following reaction (Equation 1).29 The
decomposition of CdSO4 starts at 750oC leading to a further decrease in mass.
CdS (g) + 2O2 → CdSO4 (4-1)
he TGA behavior of TiO2/CdS nanocomposite is basically a combination of TiO2
and CdS behaviors. The weight loss below 200oC could be attributed to the water absorbed
on the surface of particles, while weight loss from 200 to 400oC could be due to the loss of
the rest of surfactant on the surface of TiO2. The mass increase observed at the same
temperature level corresponding to the increase in CdS mass, is due, as mentioned above, to
the formation of the intermediate product, CdSO4.
The ζ-potential curves of TiO2 nanorods before and after NOBF4 treatment, CdS
NPs, and TiO2/CdS nanocomposites are shown in Figure 4.11. According to these curves,
the charge surface potential of TiO2 nanorods before surfactant exchange was zero at pH=5.
However, when treated with NOBF4, the surface of TiO2 nanorods was negatively charged,
which is in agreement with the results reported by Dong et al.30 Since the surface of
TiO2/CdS is negatively charged, Ni clusters were selectively deposited (by using
photodeposition technique) as co-catalysts on the surface of TiO2/CdS composite. In this
case, Ni2+ is selectively adsorbed on the surface of TiO2 nanorods, not on the surface of
CdS (because the ζ-potential of CdS is zero), due to the electro kinetic potential preferable
91
in colloidal systems. Under visible light illumination, the generated electrons from the
conduction band of CdS are transferred to the conduction band of TiO2. Because the
conduction band level of Ni2+/Ni is lower than that of TiO2, the electrons from the
conduction band of TiO2 are able to reduce Ni2+ to form metallic Ni clusters on the surface
of TiO2 nanorods (Scheme 4.1).
Figure 4.11. ζ-Potential distributions in aqueous solution at pH~5 of TiO2 nanorods before
and after treatment with NOBF4 surfactant; CdS NPs, and TiO2/CdS nanocomposite.
4.3.6. Photocatalytic activity
The photocatalytic activity of TiO2, CdS, and TiO2/CdS nanocomposite with Ni co-
catalyst for H2 generation were carried out under visible light irradiation (λ > 420 nm)
using ethanol as sacrificial reagent. As seen from Figure 12a, TiO2 nanorods are not able to
generate H2 because TiO2 nanorods do not absorb visible light and consequently could not
generate electron-hole to support the H2 evolution. Beside, CdS alone shows very low H2
generation rates, only 0.77 µmol•h-1•g-1 after 4.5 h of reaction. The low rate could be due to
92
the rapid recombination of photogenerated electrons and holes, which resulted in the lack of
H2 evolution sites.31,32 The coupling of CdS with TiO2 nanorods shows a big improvement
in H2 production; around 33.63 µmol•h-1•g-1 of H2 were evolved, which is around 44 times
higher than the production for Ni-CdS system. The rate of Ni-TiO2/CdS photocatalytic
activity is also reported to be faster compared to that of Ni-CdS, which could be due to a
better charge transfer between CdS and TiO2, as shown and discussed above (Figure 4.9).
The photocatalytic performance of TiO2/CdS without Ni-cocatalyst using ethanol as
sacrificial reagent was also carried out, however, the H2 production evolution maybe was
too low, and so we would not be able to detect the signal of activity. In another word,
without using Ni as cocatalyst, the composite TiO2/CdS is not active for photocatalytic H2
production using visible light.
To investigate the stability of Ni-TiO2/CdS samples, a series of tests composed of 4
cycles with intermittent deoxygenation were carried out without catalyst regeneration.
Between each cycle, the reaction system was bubbled with N2 in order to remove H2. As
shown in Figure 4.12(b), the results show good stability for the photocatalyst up to 15 h of
irradiation without noticeable catalytic de-activation; however, after 15 hours of reaction,
the activity is decreased by about 50%. Even though the photocatalyst was decreased after
15 h of irradiation, this achievement is still considered as a good improvement for the
photocatalytic activity of metal sulfides, which are often unstable for conventional CdS
photocatalysts, due to the reduction of metal cations in metal sulfides by generated
electrons, and the oxidation of S2- by generated holes.31–33
In Ni-TiO2/CdS nanocomposite, with the support of TiO2 nanorods, the photo-
oxidation is avoided due to the electrons transfer from the conduction band of CdS to that
of TiO2 and then to the metallic cocatalyst (Ni), therefore it would prevent Cd2+ from
reduction. In addition, under visible light illumination, only CdS with small bandgap
energy of 2.4 eV can generate holes in the valence band (VB). However, because the VB of
CdS (+1.5 V vs. SHE) is smaller than the VB of TiO2 (+ 3.2 V vs. SHE),34,35 these holes in
the VB of CdS cannot be transferred to the VB of TiO2. Thus, Ni clusters, which are only
located on the surface of TiO2, cannot be oxidized by holes in the VB of CdS NPs.
Therefore, with those mentioned special features above, it is not surprising to see that Ni-
93
TiO2/CdS nanocomposite exhibits not only high activity, but also good stability in the
photocatalyst production of H2 up to 15 h of irradiation.
Figure 4.12. (a) Comparison of the activity of H2 evolution using different photocatalysts; (b)
H2 production from TiO2/CdS-Ni photocatalyst monitored over 18 h. Each 4.5 h, the reaction
system is bubbled with N2 to remove the H2 inside.
The mechanism of H2 production activity of Ni-TiO2/CdS under visible light is
illustrated in Figure 4.13. The full mechanism could be similar to the mechanism of Pt-
TiO2/CdS, which was reported in literature.36 When the coupled TiO2 and CdS
semiconductors are activated under the visible light, electrons and holes are generated in
94
the conduction and valance bands of CdS. Furthermore, due to the different bandgap
positions, the generated electrons from the conduction band of CdS are transferred towards
TiO2 conduction band. As the Ni clusters are preferentially attached on TiO2 nanorod
instead of CdS NPs, they would be able to cap the electron from the conduction band of
TiO2, and act as H2 evolution.
Figure 4.13. Mechanism illustration of the activity of Ni-TiO2/CdS under visible light for the
production of H2, inset is the potential redox energy corresponding to CdS, TiO2 and H+/H2
Meanwhile, the holes at the valance band of CdS are responsible for oxidizing
ethanol and may also anodically auto-corrode the CdS particles37,38. If the photogenerated
holes do not react quickly with Cd–OH groups or ethanol, the photo-corrosion of CdS
occurs and induces a release of cadmium ion in solution leading to the formation of
cadmium hydroxide layer on the surface of the CdS particles, as shown by the following
equation:
CdS + H2O → H2 + Cd2+ + S + 2OH- (4-2)
Also, if the surface of CdS NPs is covered with cadmium hydroxide, this causes
sulfide vacancies saturation; hence the holes can no longer be trapped. The recombination
95
of generated electrons and holes will then be faster39. With those two reasons, it would be
expected to see a decrease in the photocatalysis activity of the Ni-TiO2/CdS system after
15h of reaction.
4.4. Conclusions
In conclusion, we developed a new hybrid photocatalytic system for the production
of H2 under visible light illumination using ethanol as a sacrificial agent, which was based
on TiO2 nanorods, CdS nanoparticles and Ni cluster cocatalyst. In a first time and due to
the fact that OA and 6AHA surfactants have selective bindings to the different faces of
TiO2 , different sizes of TiO2 nanorods were obtained varying the molar ratio between TB,
OA and 6AHA. A three step synthesis process was then used to develop the TiO2/CdS-Ni
nanocomposite photocatalyst. Under visible-light illumination and due to the fact that
electron-hole separation was highly enhanced, the developed TiO2/CdS-Ni photocatalyst
showed a high photocatalytic performance for the H2 production, which was around 44
times higher than that of Ni-CdS. In addition, this hybrid composite photocatalyst appeared
to exhibit a high photocatalytic performance for the production of hydrogen (H2).
Acknowledgements
The authors would like to thank the Natural Sciences and Engineering Research
Council of Canada (NSERC) for financial support of this work.
4.5. References
(1) Liao, C.-H.; Huang, C.-W.; Wu, J. C. S. Catalysts 2012, 2, 490–516.
(2) Wang, D.; Li, D.; Guo, L.; Fu, F.; Zhang, Z.; Wei, Q. J. Phys. Chem. C 2009, 113,
5984–5990.
(3) Rao, B. S.; Kumar, B. R.; Reddy, V. R.; Rao, T. S. Chalcogenide Lett. 2011, 8, 177–
185.
(4) Matsumura, M.; Furukawa, S.; Saho, Y.; Tsubomura, H. J. Phys. Chem. 1985, 89,
1327–1329.
96
(5) Chen, X.; Shangguan, W. Front. Energy 2013, 7, 111–118.
(6) Rajendran, V.; Lehnig, M.; Niemeyer, C. M. J. Mater. Chem. 2009, 19, 6348.
(7) Daghrir, R.; Drogui, P.; Robert, D. Ind. Eng. Chem. Res. 2013, 130226090752004.
(8) Xiang, Q.; Yu, J.; Jaroniec, M. J. Am. Chem. Soc. 2012, 134, 6575–6578.
(9) Wang, X.; Liu, G.; Chen, Z.-G.; Li, F.; Wang, L.; Lu, G. Q.; Cheng, H.-M. Chem.
Commun. (Camb). 2009, 3452–3454.
(10) Li, X.-H.; Zhang, J.; Chen, X.; Fischer, A.; Thomas, A.; Antonietti, M.; Wang, X.
Chem. Mater. 2011, 23, 4344–4348.
(11) Xiang, Q.; Yu, J. J. Phys. Chem. Lett. 2013, 4, 753–759.
(12) Chen, X.; Mao, S. S. Chem. Rev. 2007, 107, 2891–2959.
(13) Li, G.-S.; Zhang, D.-Q.; Yu, J. C. Environ. Sci. Technol. 2009, 43, 7079–7085.
(14) Štengl, V.; Králová, D. Int. J. Photoenergy 2011, 2011, 1–14.
(15) Bai, S.; Li, H.; Guan, Y.; Jiang, S. Appl. Surf. Sci. 2011, 257, 6406–6409.
(16) Shangguan, W. Sci. Technol. Adv. Mater. 2007, 8, 76–81.
(17) Gopidas, K. R.; Bohorquez, M.; Kamat, P. V. J. Phys. Chem. 1990, 94, 6435–6440.
(18) Yu, J.; Hai, Y.; Cheng, B. J. Phys. Chem. C 2011, 115, 4953–4958.
(19) Dinh, C.-T.; Nguyen, T.-D.; Kleitz, F.; Do, T.-O. ACS Appl. Mater. Interfaces 2011,
3, 2228–2234.
(20) Joo, J.; Kwon, S. G.; Yu, T.; Cho, M.; Lee, J.; Yoon, J.; Hyeon, T. J. Phys. Chem. B
2005, 109, 15297–15302.
(21) Li, X.-L.; Peng, Q.; Yi, J.-X.; Wang, X.; Li, Y. Chemistry 2006, 12, 2383–2391.
(22) Limaye, M. V; Singh, S. B.; Date, S. K.; Kothari, D.; Reddy, V. R.; Gupta, A.;
Sathe, V.; Choudhary, R. J.; Kulkarni, S. K. J. Phys. Chem. B 2009, 113, 9070–
9076.
(23) Sing, K. S. W.; Everett, D. H. W.; Haul, R. A.; Moscou, L.; Pierotti, J.; Rouquerol,
J.; Siemieniewska, T. Pure Appl. Chem. 1985, 57, 603–619.
(24) Qian, S.; Wang, C.; Liu, W.; Zhu, Y.; Yao, W.; Lu, X. J. Mater. Chem. 2011, 21,
4945.
(25) Hu, H.; Kung, S.-C.; Yang, L.-M.; Nicho, M. E.; Penner, R. M. Sol. Energy Mater.
Sol. Cells 2009, 93, 51–54.
(26) Hotový, I.; Huran, J.; Janík, J.; Kobzev, A. Vacuum 1998, 51, 157–160.
97
(27) Hotovy, I.; Huran, J.; Spiess, L.; Hascik, S.; Rehacek, V. Sensors Actuators B Chem.
1999, 57, 147–152.
(28) Zhu, H.; Yang, B.; Xu, J.; Fu, Z.; Wen, M.; Guo, T.; Fu, S.; Zuo, J.; Zhang, S. Appl.
Catal. B Environ. 2009, 90, 463–469.
(29) Sabah, A.; Siddiqi, S. A.; Ali, S. World Acad. Sci. Eng. Tech. 2010, 45, 82–89.
(30) Dong, A.; Ye, X.; Chen, J.; Kang, Y.; Gordon, T.; Kikkawa, J. M.; Murray, C. B. J.
Am. Chem. Soc. 2011, 133, 998–1006.
(31) Bao, N.; Shen, L.; Takata, T.; Domen, K. Chem. Mater. 2008, 20, 110–117.
(32) MA, G.; YAN, H.; SHI, J.; ZONG, X.; LEI, Z.; LI, C. J. Catal. 2008, 260, 134–140.
(33) Amirav, L.; Alivisatos, a. P. J. Phys. Chem. Lett. 2010, 1, 1051–1054.
(34) Ran, J.; Yu, J.; Jaroniec, M. Green Chem. 2011, 13, 2708.
(35) Sakai, N.; Ebina, Y.; Takada, K.; Sasaki, T. J. Am. Chem. Soc. 2004, 126, 5851–
5858.
(36) Qi, L.; Yu, J.; Jaroniec, M. Phys. Chem. Chem. Phys. 2011, 13, 8915.
(37) Meissner, D.; Memming, R.; Shuben, L.; Yesodharan, S.; Grätzel, M. Berichte der
Bunsengesellschaft für Phys. Chemie 1985, 89, 121–124.
(38) Davis, A. P.; Huang, C. P. Water Res. 1990, 24, 543–550.
(39) Weller, H.; Haase, M.; Spanhel, L.; Henglein, A. Prog. Colloid Polym. Sci. 1988, 76,
24–26.
99
Chapter 5. Synthesis of capped TiO2 nanocrystals of controlled
shape and their use with MEH-PPV to develop nanocomposite
films for Photovoltaic applications
Thi Thuy Duong Vu a,b, Frej Mighri a,b,*, Trong-On Doa,c Abdellah Ajjib,d,
aDepartment of Chemical Engineering, Laval University, Quebec, QC, G1V 0A6 Canada;
b Center for Applied Research on Polymers and Composites (CREPEC);
cCentre in Green Chemistry and Catalysis (CGCC);
dDepartment of Chemical Engineering, École Polytechnique of Montreal, C.P. 6079, Montreal, QC,
H3C 3A7 Canada.
Published in Journal of Nanoscience and Nanotechnology, 2012, 12(3), 2815-2824.
100
Abstract
This study presents the synthesis details of titanium dioxide (TiO2) nanoparticles
(NPs) of different morphologies using oleic acid (OA) and oleyl amine (OM) as capping
agents. Different shapes of NPs, such as nanospheres, nanorods, and nanorhombics, were
achieved. In order to develop nanocomposite thin films for photovoltaic cells, these TiO2
NPs were carefully dispersed in 2-methoxy-5-(2’-ethylhexyloxy)-p-phenylene vinylene
(MEH-PPV) matrix.
The properties of synthesized TiO2 NPs and MEH-PPV/TiO2 nanocomposites were
characterized using transmission electron microscopy (TEM), thermogravimetric analysis
(TGA), UV-Visible spectroscopy, and Photoluminescence technique. Obtained results
showed promising properties for photovoltaic devices, especially solar radiation absorption
properties and charge transfer at the interface of the conjugated MEH-PPV matrix and TiO2
dispersed NPs.
101
Résumé
Cette étude présente les détails de synthèse des nanoparticules (NPs) de dioxyde de
titane (TiO2) avec différentes morphologies en utilisant l’acide oléique (OA) et
l’oléylamine (OM) comme tensioactifs. Différentes formes des NPs, comme les
nanosphères, les nanotubes et les nanorhombiques, ont été obtenues. Pour développer de
films minces de nanocomposites pour les cellules photovoltaïques, ces NPs de TiO2 ont été
soigneusement dispersées dans une matrice de 2-méthoxy-5-(2’-ethylhexyloxy)-p-
phenylène vinylène (MEH-PPV).
Les propriétés des NPs de TiO2 et les nanocomposites de MEH-PPV/ TiO2 ont été
caractérisées en utilisant la microscopie électronique à transmission (TEM), l’analyse
thermogravimétrique (TGA), la spectroscopie UV-Visible et la technique de
photoluminescence. Les résultats obtenus ont montré des propriétés prometteuses pour les
dispositifs photovoltaïques, spécialement les propriétés d’absorption de la radiation solaire
et le transfert de charge à l’interface entre la matrice MEH-PPV et les NPs dispersées de
TiO2.
102
5.1. Introduction
As the global energy consumption increases continuously (it is expected to be
double within next 50 years), a significantly larger fraction of our energy supply will need
to be sourced from renewable sources in the very near future. Photovoltaic devices (PV),
which convert the solar radiation into direct current electricity, were discovered since 1954
and are presently considered as one of the fastest growing renewable energy technologies.1
Recently, polymer organic solar cells based on an interpenetrating network of electron
donors and acceptors prepared using solutions of conjugated polymers have become
attractive for use in inexpensive large area and low weight solar devices.2–4 The newest
generation and most two efficient polymer based solar cells are bulk heterojunction solar
cells (BHSC) and dye sensitized solar cells (DSSC). Up to date, the most common
conjugated polymers used as hole transporting materials in solar cells are poly (2-methoxy,
5-(2-ethyl-hexy-loxy)-p-phenyl vinylene) (MEH-PPV),5,6 poly(3 -hexylthiophene)
(P3HT),7,8 and poly(2-methoxy-5-(3,7-dimethyloctyloxy)-1,4-phenylene- vinylene)
(MDMO-PPV).8 On the other hand, different kinds of inorganic nanocrystals, such as
cadmium selenide (CdSe),9 zinc oxide (ZnO),10 cadmium sulphide (CdS),11 and TiO2,12 are
reported as charge acceptors.
Besides many attractive advantages, charge recombination of the photogenerated
electro-hole pairs is the major disadvantage in the use of conjugated polymers as active
layers in solar cells. Also, since the diffusion length of excitons in conjugated polymers is
typically about 5−15 nm,13,14 it could decay without any charge transfer from the polymer
to the nanocrystals if the light excitation occurs far from the polymer/nanocrystals
interface. To overcome this limitation, blending between conjugated polymers and nano-
size crystal oxides (especially particle sizes in the range of 2–10 nm) has been recently
proposed.15 This could create a large interface between the polymer matrix and the
dispersed nanoparticles, and as a result, enhances the charge transfer inside the
nanocomposite.
In this study, we present a new simple route to synthesis nanocomposite materials
for photovoltaic application based on a conjugated polymer (MEH-PPV) and capped TiO2
nanoparticles. These capped NPs are reported to have easy shape-controlled and proper
103
band gap. They are also low cost materials, beside their original properties of chemical
stability and nontoxicity. The first part of the study reports a synthesis process used to
develop organic-capped TiO2 nanoparticles with controllable shapes. Then the second part
shows how these organic-capped TiO2 are used to develop MEH-PPV/TiO2
nanocomposites using a simple mixing technique. Optical properties, such UV-Visible and
photoluminescence spectra, of the developed MEH-PPV/TiO2 nanocomposites showed
enhanced solar radiation absorption in visible wavelength and improved charge transfer
between conjugated polymer MEH-PPV and TiO2 NPs.
5.2. Experimental
5.2.1. Materials
All chemicals were used as received without further purification or distillation.
Titanium (IV) butoxide (TB, 97%), oleic acid (OA, 90%), oleyl amine (OM, 70%), and 2-
methoxy-5-(2’-ethylhexyloxy)-p-phenylene vinylene (MEH-PPV) (PS-3900) with the
average molecular weight of 150,000-250,000 were purchased from Sigma - Aldrich
Chemical, Canada. Absolute (pure) ethanol (EtOH) and 95% EtOH (with 5% water) from
Brampton Canada, chloroform from Fisher Scientific Canada, and toluene from Anachemia
Canada Inc., were all of analytical grade.
5.2.2. Synthesis of TiO2 nanoparticles
The synthesis of OA-capped anatase TiO2 nanocrystals was done at low temperature
by hydrolysis of TB using OA or OM as capping agents, followed by the synthesis step
reported earlier by our research group.16 Typically, 1 mmol of TB was added to a mixture
of OA and OM and absolute EtOH. The obtained mixture was stirred for 10mins then
transferred into a Teflon-lined stainless steel autoclave containing absolute or 95% EtOH to
keep equilibrium in the mixture in order to avoid any change in EtOH concentration during
the crystallization process. The system was then heated at 150oC for 18 hrs. The obtained
TiO2 nanocrystals were washed with toluene and EtOH several times then dried at room
104
temperature. After purification, TiO2 was re-dispersed in chloroform or toluene solvent for
characterization and composite preparation.
By keeping the amount of TB constant and varying the other parameters, such as the
amount and molar ratio of OA and OM surfactants, the concentration of EtOH and the
reaction temperature, different shapes of TiO2 NPs are to be achieved.
5.2.3. Synthesis of MEH-PPV/TiO2 nanocomposites
MEH-PPV polymer and TiO2 NPs of known MEH-PPV/TiO2 ratios were first
dissolved into CHCl3 then mixed together for 24 hrs at room temperature. In order to
improve the dispersion of TiO2 NPs into the MEH-PPV matrix, low frequency ultrasound
sonication was applied for 2 mins before any characterization of the MEH-PPV/TiO2
nanocomposite.
5.2.4. Characterization
Size and morphology characterization of TiO2 NPs were done at 120 kV using a
JEOL JEM 1230 transmission electron microscope (TEM). Samples were prepared by
dispersing TiO2 NPs in toluene solvents in a first step; then one drop of the mixture was
placed onto a 200 mesh carbon-coated copper grid and evaporated immediately at room
temperature. The crystalline phases of NPs were characterized on a Bruker SMART
APEXII X-ray diffractometer operated at 1200 W power (40 kV, 30 mA) to generate Cu
Kα radiation (λ = 1.5418 Å). Thermal analyses of the as-synthesized TiO2 NPs were carried
out up to 650 oC at a heating rate of 10oC/min under air atmosphere using a Perkin-Elmer
TGA thermogravimetric analyzer. The room temperature UV-visible spectra of TiO2 and
MEH-PPV/ TiO2 in CHCl3 were recorded using a Cary 300 Bio UV-visible
spectrophotometer. Pure CHCl3 was used as a blank solvent. Room temperature
photoluminescence (PL) spectra of MEH-PPV/ TiO2 in CHCl3 were measured on a Varian
Carry Eclipse fluorescence spectrophotometer. The Fourier transform infrared absorption
spectroscopy (FTIR) spectra were measured using Nicolet 380 FT-IR with a thermo
105
scientific smart performer ATR module and one reflection ZnSe crystal. The
characterization was done at room temperature using atmosphere as background.
5.3. Results and Discussion
5.3.1. Synthesis and characterization of capped TiO2 nanoparticles
Figure 5.1. TEM of synthesized TiO2 NPs with different shapes: (a) nanosphere, (b)
nanorhombic, and (c) nanorod.
Figure 5.1 shows TEM images of TiO2 NPs synthesized by keeping the molar ratio
of TB unchanged while varying other synthesis parameters, as presented TiO2 NPs of
nanorhombic shape with an average size of 10x20 nm (Figure 5.1(b)) were achieved when
the OA:OM molar ratio was 3:7 and synthesis temperature was 150oC. With increasing the
106
molar ratio of OA:OM to 6:4, the shape of NPs was changed to nanorod with a size of
10x20 nm (Figure 1(c)). Further increase of the molar ratio of OA:OM to 8:2 and treatment
temperature to 180oC, TiO2 nanospheres were achieved with average diameter of 5-7 nm
(Figure 5.1(a)). All the TiO2 crystals were confirmed in anatase phase, in which XRD
patterns exhibited strong diffraction peaks at 25°and 48° (Figure 5.2). From Figure 5.2, it
was also shown that the diffraction pattern peak intensity of the TiO2 increases in the order
of nanorhombic, nanosphere and nanorod. These results suggested that the crystallinity of
nanorod is higher than of nanosphere and of nanorhombic.
Figure 5.2. XRD of synthesized TiO2 NPs with different shapes.
According to the fact that the shape of TiO2 NPs was changed by changing the
synthesis parameters, it could be assumed that the formation of TiO2 NPs is indeed
controlled by the ratio OA:OM, treatment temperature, and water/EtOH concentration. Two
107
consecutive processes took place: i) the hydrolysis of titanium precursors to create unstable
hydroxyalkosides, and ii) the olation or oxolation of these unstable hydroxyalkosides to
form Ti-O-Ti chains.17 By controlling the rate of these two processes, the growing of TiO2
NPs could be controlled, leading to different NPs shapes. As reported in literature,18,19 the
development of Ti-O-Ti chains is favored for low hydrolysis rates, low content of water
and excess titanium precusors.18 According to Livage et al.,19 the amount of water
presented in the reaction mostly contributed to the hydrolysis path as below:
2Ti(OBu) H O Ti(OH) BuOHn nn n (5-1)
Therefore, the hydrolysis rate is higher, the formation of Ti-OH is more favored for
higher amount of water and the development of three-dimensional polymeric-like chains is
insufficient, resulting in closely packed first-order particles.
In the synthesis system reported, when the heat treatment temperature is increased,
water vapor generated from the mixture of water and EtOH is increased, hence greatly
influences the creation of hydroxyalkosides. Together with water vapor, water excess in the
95% EtOH leads to the formation of prefered first-order particles (nanospheres), as shown
in Figure 5.1(a), with average diameter of 5–7 nm. OA and OM surfactants also play
important roles during the hydrolysis process as they generate water resulting from the
acid-base pair catalyst. Moreover, these two surfactants are selective bindings to different
faces of TiO2, therefore they restrict TiO2 NPs to grow in different directions. OA binds
strongly to the TiO2 {001} faces,20 while OM binding is more favored on {101} faces.21
When the molar ratio OA:OM is low (3:7 in our case), the adhesion of OM to the low
surface energy {101} face leads to a more progressive TiO2 growth along {001} to form
nanorhombic TiO2 NPs,22 as shown in Figure 5.1(b). Increasing the molar ratio OA:OM to
6:4, the hydrolysis rate decreases and the growth along {001} is preserved leading to TiO2
nanorod shape,18 as shown in Figure 5.1(c).
108
Figure 5.3. FTIR spectra of capped- TiO2 NPs with different shapes; inset [1] in the region
1400-1700cm-1; in set[2] in the region 2800-3200 cm-1.
109
The FTIR spectra of the developed capped TiO2 NPs with different shapes are
shown in Figure 5.3 together with those of OA, OM and commercial TiO2. OA and OM
spectra show sharp vibrations bands at 2922 and 2853 cm-1, which are attributed to
asymmetric and symmetric C-H bonds in methylene groups (CH2).23 The vibrations at 1465
and 722 cm-1 are due to the characteristic of -(CH2)n- chains with n > 3;24 the small peaks at
3004 cm-1 correspond to the stretching of =C-H bond. The peaks at 1708 and 1285 cm-1 in
the spectrum of OA are assigned to C=O and C-O stretch, respectively. Those appearing at
1463 and 936 cm-1 are respectively due to in-plane and out-of plane O-H. In the spectrum
of OM, the vibrations at 1652 and 1619 cm-1 correspond to the combined motion of NH2
scissoring and N-H bending. The peak at absorbed at 1041 cm-1 is a characteristic peak of
the C-N stretch. Compared to the spectrum of commercial TiO2, the presence of capping
ligand on the surface of the synthesized capped TiO2 NPs is identified by the peaks at 3004,
2922, 2853, and 1465 cm-1. However, the intensities of the absorption bands corresponding
to those of OA and OM are small. This is due to the fact that only small amounts of OA and
OM surfactants are expected to be left on the surface of TiO2 NPs. The vibration at 1708
cm-1, which is the characteristic band of carbonyl in carboxyl acid, doesn’t appear in the
spectra of capped TiO2 NPs. However, a new peak appears at 1608 cm-1; this indicates the
existence of carboxylic acid salt on the surface of capped TiO2 NPs. This carboxylic acid
salt is the result of the reaction between OA surfactant and TiO2 NPs during the
solvothermal process. Furthermore, the weak absorbance peak appearing at 1041 cm-1 on
the three spectra of capped TiO2 NPs (which corresponds to the peak of C-N bonds in the
amine groups) proves the existence of oleyl amine on the surface of capped TiO2 NPs.
110
Figure 5.4. TGA curves of TiO2 NPs characterized at a heating rate of 10 oC/min under O2
atmosphere.
Thermal characterisation of synthesized capped TiO2 NPs of different shapes are
summarized in TGA plots of Figure 5.4, obtained at a heating rate of 10ºC/min under O2
atmosphere. All the three curves corresponding to nanosphere, nanorod, and nanorhombic
TiO2 shapes show an initial weight loss starting from 50oC. This is attributed to the water
absorbed on the surface of TiO2 NPs. Obviously, the most significant weight loss occured
between 200 and 480oC corresponds to OA and OM surfactants. For higher temperatures
(>480oC), the three thermogravimetric curves show very small wheigh loss, which could be
attributed to the decomposition of residual product traces that forms a sheath over the TiO2
NPs.25
TGA characterization puts in evidence the presence of OA and OM surfactants on
the surface of the three capped TiO2 NPs (nanosphere, nanorod and nanorhombic) at weight
ratios of about 15, 9 and 7 wt%, respectively. The high percentage of the organic part at the
111
surface of TiO2 NPs contributes to the high solubility of these NPs in common organic
solvents, such as toluene and chloroform. However, this could affect the charge transfer at
the interface of NPs.24
Figure 5.5. UV-vis absorption spectra of the three synthesized TiO2 NPs of different shapes in
CHCl3 solvent.
The optical properties of the synthesized capped TiO2 NPs with different shapes,
disloved in Cloroform solvant, were characterized by UV-vis spectrometer. The
corresponding spectra are shown in Figure 5.5. Since TiO2 is an indirect band gap
semiconductor with a large bandgap, Eg ( 3.2gE eV for anatase), the optical bandgap Eg
can be determined from the absorption coefficient, α, which depends on the wavelength, .
When scattering effects are neglected, the absorption coefficient near the absorption edge
for indirect inter-band transition is given by the following relation:26
112
2( ) /i gB h E h (5-2)
where iB is the absorption constant for indirect transition, h is the Plank’s constant, and ν
is the frequency of radiation (Hz).
From the UV-Vis spectra of three TiO2 NPs with different shapes, the value of
and related photon energy could be obtained. By plotting the graph of 1 2( )h versus
photon energy E h , the intersection of the tangent to the curve and the X-axis gives
the bandgap of the NPs, as shown in Figure 5.6. Therefore, band gap energies for the three
shapes of capped TiO2 NPs were: 2.88 eV for TiO2 nanospheres, 2.66 eV for TiO2
nanorods and 2.48 eV for TiO2 nanorhombic.
Figure 5.6. Band gaps of the three synthesized TiO2 NPs determined from the plot of versus
photon energy: (a) nanosphere, (b) nanorod, and (c) nanorhombic.
113
It is well known that in semiconductors, when a photon is absorbed, an electron-
hole pair or an exciton is formed. The normal size of an exciton in bulk crystal (defined a
critical quantum measurement or exciton Bohr radius) gives an approximate dimension for
the onset of quantum confinement effects.27 When electrons and holes are being squeezed
into a dimension that approaches the critical quantum measurement, they lead to some
change in electronic state symmetries, in the state energy, in the overall shape and
symmetry of the wave function, in the polarization, as well as in localization. All of these
changes lead to an increase of the effective bandgap.28 Hence, any difference in shape of
nanocrystals results in a difference in the magnitude of the quantum confinement and
effective bandgap. Results shown in Figure 5.6 confirm again the relationship between
bandgap energy and the shape of TiO2 NPs.
5.3.2. Development and characterization of MEH-PPV/TiO2 nanocomposite films
The dispersion of TiO2 NPs in MEH-PPV/ TiO2 nanocomposite were characterized
by TEM (Figure 5.7). Those results show well dispersed-TiO2 NPs on the films. However,
the dispersion of TiO2 nanospheres and nanorods seem to be better than that of TiO2
nanorhombics. This could be explained by the different amount of surfactants capped on
TiO2 NPs surfaces, which leads to the different solubility of different TiO2 NPs in
chloroform solvent.
114
Figure 5.7. TEM of composite of MEH-PPV and synthesized TiO2 NPs with different shapes:
(a) nanosphere, (b) nanorhombic, and (c) nanorod
TGA characterization of pure MEH-PPV and MEH-PPV/TiO2 nanocomposites
were done to observe the stability of materials. Figure 5.8(a) shows typical TGA curves of
normalized mass and derivative thermogravimetry (DTG) data of the derivative of mass as
functions of temperature for MEH-PPV at a heating rate of 10 oC min−1 in air environment.
In air, the first stage of degradation reaction begins around 280 oC and stops around 390 oC,
with the maximum rate at 333 oC. The second and third stages of the decompositions
appear between 390 and 452 oC with a maximum rate of mass loss around 417 oC, and
between 452 and 560 oC with a maximum rate of mass loss around 506 oC, respectively.
115
Figure 5.8. TGA curves of (a) pure MEH-PPV (b) MEH-PPV/TiO2 nanocomposites
characterized at a heating rate of 10 oC/min under air atmosphere.
Figure 5.8(b) are the TGA characterization of MEH-PPV/TiO2 NPs composites. As
the composite of MEH-PPV and TiO2 NPs were prepared using blending solution, in which
there is no chemical bonding building up between two materials. The TGA characterization
of MEH-PPV/TiO2 NPs composite with 20 wt% MEH-PPV shows the curves which are
expected to be the sum of both TGA curve of pure MEH-PPV and TiO2 NPs. The weight
116
loss of three composites (MEH-PPV/-nanosphere, -nanorod, -nanorhombic) are 34, 28, 27
%, respectively. The weight loss of composite is in agreement with the assumption, where
the total weight loss of composite is equal to the weight loss of 20 wt% MEH-PPV in the
composite, and weight loss of TiO2 NPs.
Figure 5.9. UV-vis absorption spectra of MEH-PPV/TiO2 nanocomposites: (a) different TiO2
shapes, and (b) TiO2 nanospheres of different concentrations.
117
Figure 5.9(a) shows the UV-vis absorption spectra for pure MEH-PPV and three
MEH-PPV/TiO2 nanocomposites developed with the three different shapes of capped TiO2
NPs. The weight composition of these three nanocomposites was maintained constant
(weight ratio MEH-PPV/TiO2 = 20/80). As shown in Figure 5.9(a), the three spectra of the
nanocomposites show a sharp onset at absorption near 590 nm (2.1 eV) and two evident
peaks corresponding to MEH-PPV matrix and capped TiO2 NPs. The MEH-PPV spectrum
also shows a sharp onset near 590 nm (2.1 eV); however, it shows three absorption peaks at
~500, 330 and 240 nm in the region between 220 and 800 nm, which are expected for PPV-
derivatives.29 The first maximum peak I near 500nm is attributed to the π – π* transition of
MEH-PPV conjugated polymer.30 However, when TiO2 is added, this peak position is blue-
shifted to peak I’ at 496 nm (2.5 eV). The second small peak II at 330 nm (3.76 eV) is not
observed for the three nanocomposites due to the resonance of wavelength between MEH-
PPV and added TiO2 NPs. The intensity of third peak III at 240 nm (5.12 eV) was observed
to increase and red-shifted (~ 0.1 eV) with the addition of TiO2. Moreover, no additional
absorption peaks and no major shift of wavelength in the visible region (390 - 750 nm)
were observed for the three MEH-PPV/TiO2 nanocomposites. This indicates that there is no
evidence of ground-state charge-transfer between the MEH-PPV matrix and TiO2 NPs and
consequently, there is no chemical bonding between MEH-PPV and TiO2 NPs. By mixing
MEH-PPV and capped TiO2 NPs, the optical absorption of the MEH-PPV/TiO2 composite
increases due to the fact that TiO2 NPs also contribute to light harvesting, particularly in the
visible region. The increase in optical absorption can be attributed to scattering caused by
TiO2 NPs in the MEH-PPV matrix.9 As a result, the optical absorption of MEH-PPV/TiO2
nanocomposites increases with increasing TiO2 concentration, as shown in Figure 5.9(b) for
TiO2 nanospheres.
Figure 5.10 shows FTIR spectra for pure MEH-PPV and the three MEH-PPV/TiO2
(20/80) nanocomposites developed using the three TiO2 shapes. All the four spectra show
the same IR absorption peaks. In a previous study,31 the authors reported a C-O-Ti
absorption band at 1265 cm-1; however, this band is not observed in our study, which
confirms that there is no chemical bonding between the MEH-PPV matrix and the
dispersed TiO2 NPs.
118
Figure 5.10. FTIR of MEH-PPV and MEH-PPV/TiO2 nanocomposites. Bottom inset: FTIR
spectra of MEH-PPV and MEH-PPV/TiO2 nanocomposites using TiO2 nanospheres.
Figure 5.11 shows the photoluminescence (PL) emission spectra for pure MEH-PPV
and MEH-PPV/TiO2 nanocomposites at room temperature under excitation at a wavelength
of 495nm. PL emission refers to the spontaneous emission of light by a material under
optical excitation. PL quenching was used as a powerful measure of the efficiency of
charge transfer in donor-acceptor blend films.32 The decrease in PL intensity indicates a
better PL quenching. It also indicates a decrease in light emission of the material or a better
charge transfer within the materials. For all the MEH-PPV/TiO2 nanocomposites, the PL
spectra show peaks at 560 and 600 nm corresponding to those of MEH-PPV. Even though
the PL emission of pure MEH-PPV is not affected by the addition capped TiO2 NPs, there
is a significant difference in the quenching of PL emission intensity with respect to the
119
shape of TiO2 NPs (Figure 5.11(a)) and also with respect to the concentration of TiO2 NPs
(Figure 5.11(b)).
Figure 5.11. Photoluminescence (PL) emission of MEH-PPV/ TiO2 nanocomposites: (a)
different TiO2 shapes, and (b) TiO2 nanorods of different concentrations.
120
It is important to mention that PL quenching due to the presence of TiO2 NPs can be
attributed to either energy or charge transfer from the MEH-PPV matrix to the inorganic
dispersed TiO2 NPs. In general, a better PL quenching is an indication of a better dispersion
of TiO2 NPs. It is evident from Figure 11(a) that the quenching of fluorescence is
significantly higher for nanorod than nanosphere capped TiO2 NPs. This can be explained
by the higher surface-to-volume ratio, and also by the higher delocalization of carriers in
nanorods compared to nanospheres. This helps to increase the probability of charge transfer
at polymer-NPs interface, and also helps to prevent the back recombination of holes in
MEH-PPV matrix and electrons in TiO2 NPs.33 However, results show that the emission
quenching for nanorhombic TiO2 NPs was lower than that obtained for TiO2 nanospheres.
This can be due to the fact that nanorhombic TiO2 NPs have bigger size than TiO2
nanospheres, which affects their distribution in the MEH-PPV matrix. The interface
between these nanorhombic TiO2 NPs and the MEH-PPV matrix is then decreased and as a
consequence, the probability of charge transfer at the interface is also decreased. Now, let
us discuss the effect of OA and OM surfactants capped on the surface of TiO2 NPs. As
mentioned before, nanorhombic and nanorod TiO2 NPs have similar sizes and are both
expected to have higher surface-to-volume ratio and higher delocalization than TiO2
nanospheres. The only difference is the amount of capped OA and OM surfactants, which,
as presented in section 3.1 were around 9 wt% for nanorod TiO2 NPs and around 7 wt% for
nanorhombic TiO2 NPs. Surfactants have been reported to form a barrier that prevents
charge transfer at the interface with the polymer matrix but helps to improve the contact
area between NPs and polymer chains, which increases the charge transfer at the surface of
materials.34 Presently, there is no clear explanation about which factor between the shape of
NPs and the amount of surfactants on the surface of these NPs dominates the charge
transfer at the interface of the polymer matrix and the dispersed NPs. However, comparison
of PL results presented in this study for MEH-PPV/TiO2 nanocomposites developed with
nanorod and nanorhombic TiO2 NPs shows that the impact of ON and OM surfactants
seems to be greater than that of TiO2 particle size.
For TiO2 nanorod shape, Figure 5.11(b) shows PL intensity for MEH-PPV/TiO2
nanocomposites of various TiO2 concentrations. For small TiO2 concentrations (up to 40
wt%), PL intensity shows a small decrease. However, for higher TiO2 concentrations, PL
121
quenching increases with increasing TiO2 concentration. This increase reaches around 20%
when TiO2 NPs concentration is around 80wt%.
5.4. Conclusions
Simple solvothermal routine in autoclave was successfully used to synthesize TiO2
NPs of different shapes (nanospheres, nanorods and nanorhombics) using OA and OM as
capping agents. The presence of OA and OM on the surface of the synthesized TiO2 NPs
was confirmed by FTIR characterization and their quantitative characterization was done
using TGA under O2 atmosphere. It was found that the shape of NPs and the amount of OA
and OM surfactants capped on their surface have an effect on their energy bandgap and also
on the dispersion quality of MEH-PPV/TiO2 nanocomposites. Even though there was no
evidence of chemical bonding between MEH-PPV matrix and TiO2 dispersed NPs, MEH-
PPV/TiO2 nanocomposites showed very promising results for light absorption and coupled
electron/hole transport, which are two main characteristics for photovoltaic materials. Work
is presently focused on the optimization of TiO2 dispersion in the MEH-PPV matrix using
the three developed NPs shapes.
Acknowledgements
The authors would like to thank the Natural Sciences and Engineering Research
Council of Canada (NSERC) for financial support of this work.
5.5. Reference
(1) Chapin, D. M.; Fuller, C. S.; Pearson, G. L. J. Appl. Phys. 1954, 25, 676.
(2) Brabec, C. J.; Sariciftci, N. S.; Hummelen, J. C. Adv. Funct. Mater. 2001, 11, 15–26.
(3) Mihailetchi, V. D.; Koster, L. J. a.; Blom, P. W. M. Appl. Phys. Lett. 2004, 85, 970.
(4) Kim, Y.; Choulis, S. a.; Nelson, J.; Bradley, D. D. C.; Cook, S.; Durrant, J. R. Appl.
Phys. Lett. 2005, 86, 063502.
(5) Kang, Y.; Kim, D. Sol. Energy Mater. Sol. Cells 2006, 90, 166–174.
122
(6) Ram, M. K.; Sarkar, N.; Bertoncello, P.; Sarkar, A.; Narizzano, R.; Nicolini, C.
Synth. Met. 2001, 122, 369–378.
(7) Wu, M.-C.; Lo, H.-H.; Liao, H.-C.; Chen, S.; Lin, Y.-Y.; Yen, W.-C.; Zeng, T.-W.;
Chen, Y.-F.; Chen, C.-W.; Su, W.-F. Sol. Energy Mater. Sol. Cells 2009, 93, 869–
873.
(8) Al-Ibrahim, M.; Roth, H.-K.; Zhokhavets, U.; Gobsch, G.; Sensfuss, S. Sol. Energy
Mater. Sol. Cells 2004, 85, 13–20.
(9) Verma, D.; Ranga Rao, A.; Dutta, V. Sol. Energy Mater. Sol. Cells 2009, 93, 1482–
1487.
(10) Huynh, W. U.; Dittmer, J. J.; Alivisatos, a P. Science 2002, 295, 2425–2427.
(11) Wang, L.; Liu, Y.; Jiang, X.; Qin, D.; Cao, Y. J. Phys. Chem. C 2007, 111, 9538–
9542.
(12) Roberson, L. B.; Poggi, M. A.; Kowalik, J.; Smestad, G. P.; Bottomley, L. A.;
Tolbert, L. M. Coord. Chem. Rev. 2004, 248, 1491–1499.
(13) Ginger, D. S.; Greenham, N. C. Synth. Met. 2001, 124, 117–120.
(14) Salafsky, J. Phys. Rev. B 1999, 59, 10885–10894.
(15) Nelson, J. Curr. Opin. Solid State Mater. Sci. 2002, 6, 87–95.
(16) Dinh, C.; Nguyen, T.; Kleitz, F.; Do, T. ACS Nano 2009, 3, 3737–3743.
(17) Bessekhouad, Y.; Robert, D.; Weber, J. V. J. Photochem. Photobiol. A Chem. 2003,
157, 47–53.
(18) Li, X.-L.; Peng, Q.; Yi, J.-X.; Wang, X.; Li, Y. Chemistry 2006, 12, 2383–2391.
(19) Livage, J.; Henry, M.; Sanchez, C. Prog. Solid St. Chem. 1988, 18, 259–341.
(20) Joo, J.; Kwon, S. G.; Yu, T.; Cho, M.; Lee, J.; Yoon, J.; Hyeon, T. J. Phys. Chem. B
2005, 109, 15297–15302.
(21) Jun, Y.-W.; Casula, M. F.; Sim, J.-H.; Kim, S. Y.; Cheon, J.; Alivisatos, A. P. J. Am.
Chem. Soc. 2003, 125, 15981–15985.
(22) Wu, B.; Guo, C.; Zheng, N.; Xie, Z.; Stucky, G. D. J. Am. Chem. Soc. 2008, 130,
17563–17567.
(23) Limaye, M. V; Singh, S. B.; Date, S. K.; Kothari, D.; Reddy, V. R.; Gupta, A.;
Sathe, V.; Choudhary, R. J.; Kulkarni, S. K. J. Phys. Chem. B 2009, 113, 9070–
9076.
123
(24) Chen, S.; Liu, W. Mater. Chem. Phys. 2006, 98, 183–189.
(25) Tzitzios, V.; Niarchos, D.; Margariti, G.; Fidler, J.; Petridis, D. Nanotechnology
2005, 16, 287–291.
(26) Serpone, N.; Lawless, D.; Khairutdinov, R. J. Phys. Chem. 1995, 99, 16646–16654.
(27) Buhro, W. E.; Colvin, V. L. Nat. Mater. 2003, 2, 138–139.
(28) Li, J.; Wang, L.-W. Nano Lett. 2003, 3, 1357–1363.
(29) Miller, E.; Yoshida, D.; Yang, C.; Heeger, A. Phys. Rev. B 1999, 59, 4661–4664.
(30) Kim, S.-S.; Jo, J.; Chun, C.; Hong, J.; Kim, D. J. Photochem. Photobiol. A Chem.
2007, 188, 364–370.
(31) Su, B.; Ma, Z.; Min, S.; She, S.; Wang, Z. Mater. Sci. Eng. A 2007, 458, 44–47.
(32) Greenham, N.; Peng, X.; Alivisatos, A. Phys. Rev. B 1996, 54, 17628–17637.
(33) Petrella, A.; Tamborra, M.; Curri, M. L.; Cosma, P.; Striccoli, M.; Cozzoli, P. D.;
Agostiano, A. J. Phys. Chem. B 2005, 109, 1554–1562.
(34) Liu, J.; Wang, W.; Yu, H.; Wu, Z.; Peng, J.; Cao, Y. Sol. Energy Mater. Sol. Cells
2008, 92, 1403–1409.
125
Chapter 6. The effect of surfactants on the photovoltaic
properties of hybrid bulk heterojunction solar cells based on
MEH-PPV and TiO2-based materials
Thi Thuy Duong Vu a,b, Frej Mighri a,b,*, Trong-On Doa,c Abdellah Ajjib,d,
aDepartment of Chemical Engineering, Laval University, Quebec, QC, G1V 0A6 Canada;
b Center for Applied Research on Polymers and Composites (CREPEC);
cCentre in Green Chemistry and Catalysis (CGCC);
dDepartment of Chemical Engineering, École Polytechnique of Montreal, C.P. 6079, Montreal, QC,
H3C 3A7 Canada.
Will be appeared in Green Processing and Synthesis Journal, 2015 March.
126
Abstract
In this work, we present the synthesis details of uniform shape and size-controlled
titanium dioxide (TiO2) nanorods followed by the deposition of cadmium sulfide (CdS)
quantum dots on their surface. The achieved surfactant-capped-TiO2 nanorods as well as
CdS/TiO2 nanocomposites were dispersed in nonpolar solvents, which enabled an easy
solution blending with MEH-PPV conjugated polymer to prepare the active layer of bulk
heterojunction solar cells (BHJSCs). The properties of the synthesized capped-TiO2
nanorods, CdS/TiO2 nanocomposites, as well as those of their corresponding blends with
MEH-PPV were characterized using transmission electron microscopy (TEM),
thermogravimetric analysis (TGA), UV-Visible spectroscopy, and Photoluminescence (PL)
technique. The characterization of the effect of the surfactants (oleic acid, OA, olyamine,
OM, and 6-aminohexanoic acid, 6AHA) left on TiO2 surface and CdS surface modification
on BHJSC photovoltaic power conversion efficiency (PEC) showed that: i) for the same
surfactants, when CdS was added on the surface of TiO2 nanorods, the PEC increased due
to the higher efficiency of CdS compared to MEH-PPV; and ii) the best PEC was obtained
with CdS/OA-6AHA-capped-TiO2 nanocomposite due to the shortest length of the carbon-
chain of 6AHA, leading higher charge carrier mobility.
127
Résumé
Dans ce travail, on démontre la synthèse des nanotiges de dioxide de titanium avec une
forme uniforme et une taille contrôlée suivie d’une déposition des points quantiques des
CdS sur leurs surfaces. Les capsules de nanotiges du surfactant TiO2 et les nanocomposites
CdS/TiO2 peuvent être dispersé dans un solvant non polaire ce qui facilite la préparation
d’une couche active pour une hétérojonction du mélange (BHJ) des piles solaires. Ces
dernières comprennent MEH-PPV et des capsules de nanotiges du surfactant TiO2 et les
nanocomposites CdS/TiO2. Les propriétés des nanotiges TiO2 recouvert en surfactants et
composites de CdS/TiO2 ainsi que les propriétés de leurs mélanges avec MEH-PPV ont été
caractérisées par la microscopie électronique à transmission (TEM), l’analyse
thermogravimétrique (TGA), la spectroscopie UV-Visible et la technique de
photoluminescence (PL). La caractérisation du l’effet de la déposition des CdS et des
surfactants (acide oléique, olyamine et acide 6-aminohexanoïque) sur la surface de TiO2 sur
l’efficacité de conversion de BHJSC photovoltaïque a montré que : i) Pour la même
surfactant, l’addition de CdS sur la surface des nanotiges de TiO2 a augmenté la PEC due à
la grande efficacité de CdS par rapport à celle de MEH-PPV; et ii) le meilleur PEC a été
obtenu avec les nanocomposites de TiO2 recouverts par CdS/OA-6AHA due à la longueur
le plus court de chaine de carbone de 6AHA entrainant la plus haute mobilité des particules
chargées.
128
6.1. Introduction
Over the last decade, hybrid bulk heterojunction solar cells (BHJSC) using soluble
conjugated polymers and inorganic nanoparticles have become attractive for use as large
area, physically flexible and low-cost solar cells.1-3 Conjugated polymers can exhibit both
electronic and optical properties, and they are well known for their mechanical properties
and intrinsic processing advantages, which makes the fabrication of BHJSCs easy leading
to a low manufacturing cost. However, conjugated polymers have low mobility of charge
carriers, particularly electrons, which results in insufficient charge generation and
transportation. Therefore, the introduction of inorganic nanomaterials into polymer matrix
would further increase the performance and the efficiency of the organic solar cell by using
the high electron mobility of the inorganic phase. Currently, the most common conjugated
polymers used as hole transporting materials in solar cells are poly (2-methoxy, 5-(2-ethyl-
hexy-loxy)-p-phenyl vinylene) (MEH-PPV),4-7 poly(3-hexylthiophene) (P3HT),8,9 and
poly(2-methoxy-5-(3,7-dimethyloctyloxy)-1,4-phenylene-vinylene) (MDMO-PPV);9 while
different types of inorganic nanoparticles and nanoparticle system are an example of
materials used such as CdS,4 CdTe,7,10 CdSe,11,12 ZnO,13-15 TiO2,16-18 CdSe-CdTe,19
PbS/TiO2 have been reported.20
Due to the difference of electrical and optical properties of the two materials, the
photogenerated excitons can diffuse to the interface between the conjugated polymer and
the inorganic nanomaterials, but the charge transportation through the nanocrystalline
network can be strongly affected by surface defects and adsorbed species on nanocrystals
surface, which acts as surface traps. Therefore, these traps can affect the mobility of charge
carriers and probably can cause their recombination, thus they affect the power efficiency
conversion (PEC) of the solar cells. Therefore, by controlling particle crystallinity as well
as their surface properties, we could help to enhance the PEC of BHJSCs.
In this work, we present a simple low temperature solvothermal synthesis process to
develop surfactant-capped-TiO2 nanorods using different capping surfactants. The shape
and size of these TiO2 nanorods were controlled and optimized in nano-scale range for
BHJSC application. We also present a facile method for developing CdS/TiO2 collide
systems using the above developed capped-TiO2 nanorods. As a final step, both capped-
129
TiO2 nanorods and CdS/TiO2 nanocomposites were separately blended with MEH-PPV
polymer. Blend solution was used to develop the active layer of BHJSC prototypes by
using the spin-coating method at room temperature. The photovoltaic properties of these
prototypes were then characterized and the effect of TiO2 surface modification on these
properties was studied.
6.2. Experimental
6.2.1. Materials
Commercially titanium (IV) butoxide (TB, 97%), oleic acid (OA, 90%), oleyl amine
(OM, 70%), 6-aminohexanoic acid (6AHA), cadmium acetate dehydrate, thioamide,
nitrosonium tetrafluoroborate solution (NOBF4) and 2-methoxy-5-(2’-ethylhexyloxy)-p-
phenylene vinylene (MEH-PPV) (PS-3900) with the average molecular weight of 150,000-
250,000 were purchased from Sigma - Aldrich Chemical, Canada. Poly(3,4-
ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT:PSS) and Indium-tin-oxide
(ITO) glass substrate 25x25 mm were respectively purchased from Nova Chemical and
Merck, Canada. Absolute (pure) ethanol (EtOH), from Brampton, Canada, N,N-
dimethylformamide (DMF), dichloromethane, dichlorobenzyl, and hexane, from Fisher
Scientific, Canada, and toluene, from Anachemia Canada Inc., were all of analytical grade.
All chemicals were used as received without further purification or distillation.
6.2.2. Synthesis of OA and OM or 6-AHA Capped TiO2 nanorods
OA, OM or 6AHA capped anatase TiO2 nanorods were synthesized at low
temperature by the hydrolysis of TB using OA and OM or 6AHA as surfactants followed
by the steps that we already described in our previous work.21, 22 Typically, 1 mmol of TB
was added to a mixture of 7 mmol of OA, 3 mmol of OM or 6AHA, and ethanol. The
system was then heated at 140oC for 18 h. The resulting TiO2 nanocrystals were washed
with toluene and ethanol 95% several times then dried at room temperature.
130
6.2.3. Synthesis of CdS modified TiO2 nanocomposite
It is important to mention that the use of surfactants during the synthesis of TiO2
nanorods results in the presence of capping molecules on the surface of these nanorods.
These capping molecules create an insulating barrier around each nanorod and block the
access to its surface. Therefore, in order to deposit CdS nanoparticles on the surface of
TiO2 nanorods, the latter must undergo a surface treatment, typically by replacing the
original ligands with specifically designed species through a ligand-exchange process. For
this, NOBF4 was used in our study. This strategy helps to enable sequential surface
modification of TiO2 nanorods without affecting their size and shape, and also to increase
the access to their surface.
Basically, surfactants-capped-TiO2 nanorods dispersed in hexane solvent were
added into 5 mL of dichloromethane solution of NOBF4 (0.01M). The mixture was well
shaken; the precipitation of TiO2 nanorods was then observed just after few minutes.
Precipitated nanorods were collected then washed by using a mixture of toluene and
ethanol 95%.
As a final step, 4.5 mmol of the above treated TiO2 nanorods were dispersed in 10
ml of DMF and 9 mmol of cadmium acetate dihydrate then stirred under room temperature
for 2 h. Subsequently, 9 mmol of thioamide were added to the mixture and let under stirring
overnight. The precipitated CdS/TiO2 nanocrystals were washed few times using toluene
and ethanol 95%, and then collected by centrifugation.
6.2.4. Preparation of MEH-PPV/capped-TiO2 and MEH-PPV/CdS/TiO2 blend
solutions
Mixtures of MEH-PPV and TiO2 or CdS/TiO2 were prepared by solution blending
in dichlorobenzyl solvent with a mole ratio MEH-PPV:TiO2 or MEH-PPV:CdS/TiO2 of
1:3. Blend solutions were continuously stirred overnight under dark. Ultrasonic agitation
was applied to disperse nanoparticle agglomerates in the solvent and to ensure dispersion
homogeneity.
131
6.2.5. Fabrication of BHJ solar cell devices
The fabrication process consists to develop a sandwich structure, as shown in
scheme 1, composed of i) ITO/PEDOT:PSS anode, ii) MEH-PPV/CdS/TiO2 nanocomposite
active layer, and iii) Al cathode.23 A 25x25 mm ITO glass substrate with a sheet resistance
of 15Ω/square was ultrasonically cleaned with soap, followed by acetone then isopropanol
solvents. A thin layer of PEDOT:PSS was then spin-coated on the cleaned ITO glass
substrate, then baked at 80oC for 5 min to remove any possible contamination. Being a
highly hole-conducting metal-like polymer, PEDOT:PSS eases the conduction of holes
from the active layer to the ITO layer by lowering the intrinsic energetic barrier between
the ITO work function and the HOMO of the active material. So this layer acts as a semi-
permeable membrane for holes only and blocks the extraction of electrons on the ITO side.
Therefore, PEDOT:PSS layer helps to reduce electron-hole recombination phenomena. In
addition, it also allows ITO surface smoothing, which is usually quite rough, and hence
reduces the risks of shortcuts within the BHJSC. The composite CdS/TiO2/MEH-PPV
solution already prepared, which constitutes the active layer, was then deposited over the
PEDOT:PSS layer by the same spin-coating technique. Finally, four aluminum cathode
layers (700 Å in thickness) were then thermally evaporated in vacuum chamber at a
pressure below 3 x 10-4 Pa. The final device is then composed of four heterojunction cells
with individual active area of 0.24 cm2.
Scheme 6.1. Architecture scheme of a BHJSC device
132
6.2.6. Characterization
Powder X-ray diffraction (XRD) characterization was done using a Bruker SMART
APEXII X-ray diffractometer equipped with a Cu Kα radiation source (λ = 1.5418 Å) in the
2θ range of 5–20° at a scan rate of 1.0°/min. All samples were dried at 65°C overnight to
eliminate guest solvent molecules on the surface of particles before the XRD scan. Fourier
transform infrared (FTIR) characterization was done using a FTS 45 infrared
spectrophotometer in the spectral range of 4000–400 cm–1. The characterization was done
at room temperature using atmosphere as background. The thermal characterization of the
developed TiO2 nanorods, CdS nanoparticles and hybrid CdS/TiO2 nanocomposites was
carried out at a heating rate of 10°C/min up to 900°C under an oxygen flow using a Perkin-
Elmer TGA thermogravimetric analyzer. The UV−visible spectra of the developed
nanostructures were recorded on powder samples using a Cary 300 Bio UV−visible
spectrophotometer; pure toluene was used as a blank. Room temperature
photoluminescence (PL) characterization was done by using an optical spectrum analyzer
(ANDO AQ6317, Japan). Electronic transport in the BHJSC device was studied via current
density-voltage (I-V) characterization using a 2400 Keithley source meter. The voltage was
varied from -0.5 to 2.0 V. I-V curves were taken both in the dark and under white
illumination provided by a halogen source through the ITO electrode.
6.3. Results and Discussions
6.3.1. Analysis of synthesized capped-TiO2 nanorods
6.3.1.1. TEM Characterization
Figures 6.1(a) and (b) show representative TEM images of the as-synthesized TiO2
nanoparticles using respectively OA/OM and OA/6AHA as surfactant combinations. In
both cases, the shape of TiO2 nanoparticles was in the form of nanorods of 4–5 nm in
diameter and 20–40 nm in length. The morphologies of TiO2 nanoparticles were closely
controlled by the presence of OA/OM and OA/6AHA surfactants. According to literature,
133
these surfactants play important roles during the hydrolysis process as they generate water
resulting from the acid-base pair catalyst. More they generate water, faster is the hydrolysis
process, leading to larger nanoparticle dimensions. Besides, the presence of surfactant
lowers the surface tension, which allows particles further grow in surfactant direction. For
the case of addition of TB precursors into the reaction solution, the early formation of TiO2
truncated octahedral bi-pyramid seeds were expected. These seeds terminated by {001}
faces, which have high surface energy, and {101} faces with relative low energy. When the
hydrolysis process increases, the final shape of TiO2 nanoparticles is more controlled by the
competition between the relative surface energies of the {001} and {101} faces and,
therefore, the growth rate ratio between [001] and [101] directions.21, 24-26 Hence, with the
utilization of OA surfactant, which is more favored to bind strongly to TiO2 {001} faces,
and amines like OM or 6AHA that weakly bind on {101} faces, this leads to a more
progressive TiO2 growth along [001] direction than along [101] direction. Therefore, this
oriented growth leads to TiO2 nanorods instead of TiO2 nanospheres or nanorhombics, as
presented in our previous work.21
Figure 6.1. TEM of TiO2 nanorods synthesized using (a) OA/OM, and (b) OA/6AHA
surfactants combinations
6.3.1.2. XRD and FTIR characterization
Figure 6.2 shows XRD spectra of OA/OM-capped-TiO2 and OA/6AHA-capped-
TiO2 nanoparticles. Both spectra show strong diffraction peaks at 25° and 48°, indicating
134
TiO2 anatase phase with typical anisotropic growth pattern along the [001] direction
(JCPDS no: 88-1175 and 84-1286). In order to understand the surface properties of TiO2
nanorods, FTIR characterization was carried out. The FTIR of pure OA, OM, and 6AHA
surfactants were also analyzed for comparison. The corresponding results are shown in
Figure 6.3. The FTIR bands at 2850-2920 cm-1 are attributed to the asymmetric and
symmetric C–H stretching vibrations of methylene groups, the vibration at 1465 cm-1 is a
characteristic of -(CH2)n- chains with n > 3, the small peaks at 3004 cm-1 correspond to the
stretching of =C-H bond.27-29 Furthermore, it was observed that the vibration bands at 1710
and 1285 cm-1, which are the characteristic bands of carbonyl -C=O- and -C-O- stretch in
the carboxyl acid of OA, appear weak on the spectrum (A2) of OA-6AHA-capped-TiO2
nanoparticles and that of OA-OM-capped-TiO2 nanoparticles (A1). This indicates in both
cases chemisorptions of oleic acid onto the surface of TiO2 nanorods.30 However, the
intensities of those absorption bands are all reduced compared to the absorption bands of
pure OA, OM and 6AHA. This is due to the fact that only small amounts of surfactants are
expected to be left on the surface of TiO2 nanoparticles. Also, the FTIR spectrum (A1) of
OA-OM-capped-TiO2 nanoparticles shows a vibration band at 1379 cm-1, which is identical
to that observed on the FTIR spectrum of pure OM.31 This indicates the presence of OM on
the surface of OA-OM-capped-TiO2 nanoparticles. A close look to the spectrum shows two
peaks at 1536 and 1560 cm-1 corresponding to the symmetric and asymmetric stretching
vibrations of uncoordinated –COO- group.32 These two peaks are observed to be absent on
the spectrum (A2) of OA-6AHA-capped-TiO2 nanoparticles. The latter spectrum also
shows a band at 1386 cm-1 corresponding to C–N stretching mode of AHA molecules, and
two vibration peaks at 1622 and 1506 cm-1, which are anti-symmetric and symmetric
deformation peaks of NH3+. These results prove the binding of amino groups on the surface
of TiO2 nanorods and show that i) only the amino (–NH2) group of AHA molecules capped
on the surface of TiO2 nanorods and ii) the free carboxylic (–COOH) terminus was oriented
outward.33 Finally, the broad band at around 3000 to 3600 cm-1 is due to the presence of
adsorbed water on surface of the sample. This band also appears on spectrum (A1) of OA-
OM-capped- TiO2 nanoparticles but with a lower intensity, which is an indication that the
amount of water on OA-OM-capped- TiO2 was less than that on OA-6AHA-capped- TiO2.
This will also be confirmed in by TGA characterization presented in the following section.
135
Figure 6.2. Powder XRD patterns of OA-OM-capped-TiO2 and OA-6AHA-capped-TiO2
nanoparticles. The diffraction pattern of TiO2 anatase is also reported as a reference
Figure 6.3. FTIR spectra of OA-OM-capped-TiO2 (A1), OA-6AHA-capped-TiO2 (A2)
nanoparticles, pure OA, OM, and 6AHA
136
6.3.1.3. TGA characterization
TGA curves of capped-TiO2 nanorods, obtained at a heating rate of 10ºC/min under
O2 atmosphere, are shown in Figure 6.4. For both surfactants combinations, an initial
weight loss starting from 50oC was observed. The most significant weight loss occured
between 200 and 480oC, which is a clear indication of the presence of surfactants OA/OM
and OA/6AHA surfactants on the surface of TiO2 nanorods. For higher temperatures (>
480oC), the small weigh loss is attributed to the decomposition of residual product traces
that forms a sheath over the TiO2 nanoparticles.34 By calculating the weight loss different
from 200 to 480oC, the weight proportions of OA/OM and OA/6AHA surfactants were
around 9% and 16%, respectively. These results are all in agreement with those already
obtained in our previous work.21 In addition, according to the TGA spectrum, the amount of
water absorbed on the surface of TiO2 nanoparticles, which was already observed above by
FTIR characterization, was around 1.4% for OA-OM-capped-TiO2 and 3.5% OA-6AHA-
capped-TiO2.
Figure 6.4. TGA spectra of OA-OM-capped-TiO2 nanoparticles (A1) and OA-6AHA-capped-
TiO2 (A2) nanoparticles (heating rate: 10ºC/min, O2 atmosphere)
137
6.3.1.4. UV-vis characterization
The UV–vis absorption spectra of both OA/OM and OA/6AHA surfactant-capped-
TiO2 nanorods are shown in Figure 6.5. As shown, their corresponding absorption band
edges were around 370 and 380 nm, which are approximately similar to the gap energy of
bulk anatase TiO2 (385 nm). It is important to mention that, for small TiO2 nanoparticles
(size < 20 nm), Ti atoms at the surface of the nanoparticles adjust their coordination
environment (compression of the Ti-O bond) in order to accommodate the curvature of the
nanoparticles.35 Then, the hybrid localized defect sites could enhance the selective
reactivity of TiO2 nanoparticles towards bidentate ligands binding. As a consequence, the
chelation of Ti atoms on the surface with electron donating bidentate ligands changes the
electronic properties of TiO2 nanoparticles. Then, the absorption of light by the charge–
transfer complex yields to the excitation of electrons from the chelating ligand directly into
the conduction band of these nanoparticles.36 By using different surfactants during the
synthesis process, different capping ligands were observed to bind on the surface of TiO2
nanoparticles, which results in different charge-transfer processes from the chelating ligand
into the conduction band for the different types of TiO2 nanoparticles. Therefore, a red shift
of the UV-vis absorption spectra is expected for both OA/OM and OA/6AHA surfactant-
capped- TiO2 nanoparticles. Hence, UV-vis spectra show that the charge transportation is
better for OA-6AHA-capped-TiO2 nanoparticles than for OA-OM-capped-TiO2
nanoparticles. The energy band gap was evaluated from the following Tauc relation:
1 2( ) gh E h (6-1)
where α (cm-1) is the absorption coefficient, h (J.s) is the Plank constant, ν (Hz) is the
frequency of radiation, and Eg (eV) is the energy band gap for direct band gap
semiconductor. By drawing the tangent line on the linear part of the curve 1 2( )h versus
photon energy, h , the intercept of this line with the photon energy axis gives the value of
Eg (see the insets of Figure 6.5).37 Based on that, the energy band gaps for OA-OM-capped-
TiO2 and OA-6AHA-capped-TiO2 nanoparticles were respectively 3.53 and 3.31 eV.
138
Figure 6.5. UV-vis characterization of capped-TiO2 nanoparticles (a) OA-OM-capped-TiO2,
(b) OA-6AHA-capped-TiO2. The insets show their respective band gap energy plots.
6.3.2. Analysis of the synthesized CdS modified TiO2 nanorods
6.3.2.1. FTIR characterization
FTIR characterization was done for the CdS/TiO2 nanocomposite samples in order
to analyze their surface properties. Figure 6.6 shows the corresponding spectra, together
with those of pure OM, OM, 6AHA and NOBF4. As shown, after doing the surface
treatment with NOBF4 and depositing CdS on the surface of TiO2 nanorods, the essential
peak characteristics of -C-H, =C-H, -C=O-, -C-O-, and –NH2 stretching vibration of the
different surfactants either disappeared or appeared very weak. This is an indication that
NOBF4 treatment process was able to remove a big part of OA, OM and 6AHA molecules
attached to TiO2 nanorod surface.
139
Figure 6.6. FTIR curves of the two developed CdS/TiO2 nanocomposites, together with those
of pure OA, OM, 6AHA and NOBF4.
6.3.2.2. TGA characterization
Figure 6.7 shows the TGA curves of both CdS/TiO2 nanocomposites and that of
bulk CdS. The weight losses below 200oC were attributed to water absorbed on the surface
of the nanocomposites, while the weight losses from 200-480oC were attributed to the loss
of OA/OM and OA/6AHA surfactants from the surface of capped-TiO2 nanoparticles. As
shown, a non-negligible gain in mass was observed between 480 and 750°C for both CdS
and CdS/TiO2 nanocomposites. This surprising gain in mass is due to the formation of
cadmium sulphate (CdSO4).38 The latter began to decompose at around 750oC, which
explains the decrease in mass observed at higher temperatures.
140
Figure 6.7. TGA spectra of CdS/OA-OM-capped-TiO2 (A1) and CdS/OA-6AHA-capped-TiO2
(A2) (heating rate: 10ºC/min, O2 atmosphere).
6.3.2.3. UV-vis characterization
As reported before, TiO2 nanoparticles with their characteristic band gap of around
3.2 eV had no absorption band in the visible region. They only show a characteristic
absorption spectrum (absorption of Ti–O bond) in ultraviolet light range from 320 to 400
nm. However, as clearly shown in Figure 6.8, the addition of CdS on the surface of TiO2
nanorods can effectively shift the absorption range of TiO2 into visible light region of 400–
550 nm due to the narrow band gap of CdS (2.4 eV). Compared to the spectra of CdS and
both capped-TiO2 nanoparticles, the spectra of CdS/TiO2 nanocomposites are basically a
combination of those of CdS and capped-TiO2, where the absorption bands around 330 nm
were from TiO2, and the broad absorption bands around 530 nm were from CdS. The red-
shift of absorptions bands of the CdS/TiO2 nanocomposites were probably due to the
coupling between TiO2 and CdS, leading to the decrease of surface defects.39 With these
shifts in absorption bands, the light-harvesting efficiencies of CdS/TiO2 nanocomposites
141
were larger than those of TiO2 nanoparticles in the visible light region, which is benefic for
the photovoltaic activity. As shown, the highest was observed for CdS/OA-6AHA-capped-
TiO2 nanocomposite.
Figure 6.8. UV-vis characterization of capped-TiO2 nanoparticles, CdS, and CdS/ TiO2
nanocomposites.
6.3.3. Characterization of BHJSCs with active layers based on MEH-PPV/capped-
TiO2 or MEH-PPV/CdS/TiO2
6.3.3.1. SEM characterization
Figure 6.9 shows the typical SEM cross thickness images of BHJSC devices
prepared using different active layer materials (blends A1, A2, S1, and S2). BHJSC whole
thickness was around 0.5µm, which is composed of 70 nm Aluminium cathode, 100 nm
ITO glass, and around 330 nm of PEDOT:PSS and photoactive layers. The thickness of the
photoactive layer plays an important role in the final device efficiency. It should be thick
enough to absorb all the maximum incident-light but would not burden the charge
142
transportation. Therefore, a careful optimization of layer thicknesses is necessary to place
the maximum of the optical field in the photoactive material and maximize the absorption
of incident photons.40,41
Figure 6.9. SEM pictures of BHJSC active layer blends: A1 (a), A2 (b), S1 (c), and S2 (d)
6.3.3.2. TGA characterization
The TGA characterization of MEH-PPV/capped-TiO2 (active layer blends A1, A2)
and MEH-PPV/CdS/TiO2 (active layer blends S1 and S2) are presented in Figure 6.10. As
all the blends of MEH-PPV with TiO2 nanoparticles or with CdS/TiO2 were prepared using
the blending solution method, in which there is no chemical bonding up between two
materials, their corresponding TGA curves are expected to be the combination of those of
pure MEH-PPV and TiO2 nanoparticles and CdS/TiO2 nanocomposites. As the surfactant-
143
capped-TiO2 nanorods used in blends A1 and A2 had respectively OA/OM and OA/6AHA
surfactants on their surfaces, consequently the weight loss difference between the blend A1
(around 18%) and the blend A2 (around 24%) was not far from the weight loss difference
between the two surfactant-capped-TiO2 nanorods themselves (~7%). However, as shown
in the figure, the TGA curves corresponding to the blends S1 and S2 show no big loss
difference, due to the effective reduced residues of OA, OM and 6AHA molecules on TiO2
nanorod surface.
Figure 6.10. TGA spectra of BHJSC active layer blends A1, A2, and S2 (heating rate:
10ºC/min, O2 atmosphere).
6.3.3.3. UV-vis characterization
Figure 6.11 shows the UV-vis spectra of MEH-PPV/capped-TiO2 (active layer
blends A1, A2) and MEH-PPV/CdS/TiO2 (active layer blends S1 and S2) together with the
spectrum of pure MEH-PPV conjugated polymeric matrix. As shown, the UV-vis spectrum
of pure MEH-PPV consists of three absorption peaks at ~500, 330 and 240 nm in the region
between 220 and 800 nm. These peaks correspond to PPV-derivatives; also, the spectrum
shows a sharp onset near 590 nm (2.1 eV). The absorption peak observed near 500 nm (in
144
the visible region) is related to the transition π – π* of the MEH-PPV conjugated
polymer.6,42 The figure shows that this peak remained at approximately the same position
when MEH-PPV was blended with capped-TiO2 nanoparticles (blends A1 and A2).
However, when MEH-PPV was blended with CdS/TiO2 nanocomposites (blends S1 and
S2), this peak was broaden, red-shifted and extended to 525-530 nm range. This is due to
the fact that MEH-PPV and CdS/TiO2 nanocomposites have complementary absorption
spectra and light harvesting, while the absorption spectra in the visible region of MEH-
PPV/capped- TiO2 mostly correspond to the absorption of the MEH-PPV matrix.43
Furthermore, in the region between 300 and 750 nm, all the blends show no new peaks; the
absorption is almost the overlap of their pure components, indicating no chemical bonding
between the MEH-PPV matrix and capped-TiO2 nanoparticles or CdS/TiO2
nanocomposites.
Figure 6.11. UV-vis of polymer composite of MEH-PPV and two different CdS/TiO2
nanocomposites. MEH-PPV/OA-OM-capped-TiO2 NPs (blend A1), MEH-PPV/OA-6AHA-
capped-TiO2 NPs (blend A2), MEH-PPV/CdS/OA-OM-capped-TiO2 NPs (blend S1) and
MEH-PPV/CdS/OA-6AHA-capped-TiO2 (blend S2).
145
6.3.3.4. PL characterization
The efficiency of charge trapping and recombination of photo-induced electrons and
holes in the developed composites could be verified by the PL characterization. Figure 6.12
shows the PL emission spectra for pure MEH-PPV, MEH-PPV/TiO2 (active layer blends
A1 and A2), and MEH-PPV/CdS/TiO2 (active layer blends S1 and S2) in dichlorobenzyl at
room temperature under light excitation at a wavelength of 495 nm. It can be observed that
a significant quenching of the emission intensity of MEH-PPV occurs with the addition of
TiO2 nanorods or CdS/TiO2 nanocomposites. According to PL spectra of all the samples,
MEH-PPV/CdS/TiO2 blends exhibited much weaker intensity of peaks than MEH-
PPV/capped TiO2 blends. Though the emission features of the polymer are not affected by
the presence of CdS/TiO2 nanocomposites, the significant quenching in the emission
intensity could be ascribed to the effectively charge transfer from MEH-PPV to CdS then to
TiO2 surface, and also to the lower recombination probability of photo-induced electrons
and holes in MEH-PPV. In addition, this degree of PL quenching is an indication of how
well the nanoparticles are mixed in the polymer and the quality of the interface between the
MEH-PPV matrix and the dispersed nanoparticles.44
146
Figure 6.12. Photoluminescence (PL) of pure MEH-PPV and BHJSC active layer blends A1,
A2, S1, and S2.
It is well known that surfactants are very important to stabilize and control the shape
and size of TiO2 nanorods during the synthesis step. Also, they are very useful to increase
the dispersion of particles inside the polymer matrix. However, these useful surfactants
could be a major hurdle to the charge transfer between the MEH-PPV matrix and TiO2
nanorods or CdS/TiO2 nanocomposites. As shown in Figure 6.12, the presence of different
surfactants capped on the surface of TiO2 nanorods led to a significant difference in the PL
quenching of their corresponding blends. The PL quenching of the blends A2 and S2 were
respectively higher than those of the blends A1 and S1. This could be due to the nature of
the capping agent itself that affects the efficiency of charge transfer between TiO2
nanoparticles and also within the nanoparticle itself. In fact, both A1 and S1 blends were
based on OA-OM-capped-TiO2 nanorods, while A2 and S2 blends S2 were based on OA-
6AHA-capped-TiO2 nanorods. Only the surfactant OM (used for A1 and S1 blends) that
was replaced by 6AHA for A2 and S2 blends. The main difference between these two
surfactants is that the length of the carbon-chain of 6AHA is much shorter than of that OM.
It was reported in literature that carriers’ mobility decreases exponentially with increasing
147
ligand length.45 Hence, the probability for electrons and holes recombination is higher with
using OM as surfactant rather than 6AHA due the lower carrier mobility, which explains
the lower PL quenching of A1 and S1 blends compared to A2 and S2 blends.
6.3.3.5. Current density- voltage (J-V) characterization of BHJSCs
Figure 6.13 show the J-V curves under dark and under 100 mW/cm2 (AM 1.5 G)
illumination for four different BHJSC devices with the active layer composed of A1, A2,
S1 and S2 blends. The corresponding photovoltaic parameters [(open-circuit voltage (VOC),
short-circuit current density (JSC), fill factor (FF) and the PCE)] are summarized in Table
6.1. As shown by the insets of both figures, the J–V curves in the dark of all the devices
pass through the origin, which is currently reported for heterojunction solar cells.
As mentioned in literature46 for an active layer composed of MEH-PPV/TiO2 blend,
due to the fact that the work function of TiO2 anatase is 5.1 eV, which is closed to the work
function of ITO (4.8 eV) and the HOMO of MEH-PPV, the transportation of electron-holes
is not effective, leading to a poor PEC. In our case, the first BHJSC device has an active
layer composed of MEH-PPV blended with OA-OM-capped-TiO2 nanorods (blend A1).
This device showed a PCE of 0.003%, which is very low. When OM is replaced by 6AHA
as surfactant (blend A2), the PCE was tripled. As expected from UV-vis and PL
characterizations of the four blends A1, A2, S1 and S2, an increase in VOC, JSC, and PCE
could be obtained by using an active layer composed of MEH-PPV blended with the
CdS/capped-TiO2 composite (blends S1 and S2). This increase is due to the higher
efficiency of CdS compared to MEH-PPV; so more electrons from MEH-PPV would be
able to travel faster to CdS.47 Therefore, more electrons from the MEH-PPV matrix would
be transported to the photoanode via CdS, TiO2 and ITO. Moreover, the better results were
observed with the S2 blend.
148
Figure 6.13. J-V Characterization under light illumination (1.5 AM) of BHJSC devices with
the active layer blends (a) A1 and A2, and (b) S1 and S2. The insets are their corresponding
log J–V properties in the dark and under light.
Voltage (V)
-0.5 0.0 0.5 1.0 1.5
Curr
ent density
(mA
cm
-2)
0.000
0.005
0.010
0.015
0.020
0.025
0.030
A1 under light
A2 under light
Voltage (V)
-0.5 0.0 0.5 1.0 1.5
Lo
g
-7
-6
-5
-4
-3
-2
-1
under light
under dark
A2
A2
A1
A1
a)
Voltage (V)
-0.5 0.0 0.5 1.0 1.5
Cu
rre
nt
de
nsity (
mA
cm
-2)
-0.0005
0.0000
0.0005
0.0010
0.0015
0.0020
0.0025
0.0030
S2 under light
S1 under light
Voltage (V)
-0.5 0.0 0.5 1.0 1.5
Lo
g
-6
-5
-4
-3
under darkunder light
S1
S2
S1 S2
b)
149
Table 6.1. Summary of the photovoltaic parameters of BHJSC devices with active layer
blends A1, A2, S1, and S2
Sample ID Active layer Voltage (V) Current
(mA) FF (%)
PEC
(%)
A1 MEH-PPV blend with OA-
OM-capped-TiO2 (A1) 0.06 0.19 24.41 0.0028
A2 MEH-PPV blend with OA-
6AHA-capped-TiO2 (A2) 0.32 0.1 27.96 0.0090
S1
MEH-PPV blend with
CdS/OA-OM-capped-TiO2
(S1)
0.89 0.16 33.82 0.0467
S2
MEH-PPV blend with
CdS/OA-6AHA-capped-TiO2
(S2)
0.95 0.36 31.48 0.1062
6.4. Conclusion
In summary, the synthesis of TiO2 nanorods has been proposed using OA, which is
more favored to bind strongly to TiO2 {001} faces, and OM or 6AHA that weakly bind on
{101} faces. OA/OM and OA/6AHA combinations led to an oriented TiO2 growth to form
TiO2 nanorods. The UV-vis characterization of the developed caped-TiO2 nanorods showed
that the charge carrier was better for OA-6AHA-capped-TiO2 nanorods than for OA-OM-
capped-TiO2 nanorods due to shorter 6AHA chain length, compared to that of OM. Further
optimization of the developed TiO2 nanorods was done by doping CdS nanoparticles on
TiO2 surface. UV-vis characterization of the developed CdS/TiO2 nanocomposites showed
red-shift of absorptions bands due to the coupling between TiO2 and CdS, which is benefic
for their photovoltaic activity. When blended with MEH-PPV conjugated polymer,
photoluminescence characterization showed that the developed MEH-PPV/CdS/TiO2
blends presented improved charge transfer from MEH-PPV to TiO2 surface.
Finally, the same trend was also observed when the MEH-PPV/capped-TiO2 blends
and MEH-PPV/CdS/TiO2 blends were used as BHJSC active layers. The BHJSC devices
made from the latter blends showed greater improvement in their PEC. This improvement
150
is due to the increase of electron mobility thanks to the presence of CdS quantum dots on
TiO2 surface, as mentioned above. Also, the obtained PEC results showed that the
combination of OA/6AHA surfactants was better than that of OA/OM.
Acknowledgements
The authors would like to thank the Natural Sciences and Engineering Research
Council of Canada (NSERC) for financial support of this work.
6.5. Reference
(1) Brabec, C. J.; Sariciftci, N. S.; Hummelen, J. C. Adv. Funct. Mater. 2001, 11, 15–26.
(2) Mihailetchi, V. D.; Koster, L. J. a.; Blom, P. W. M. Appl. Phys. Lett. 2004, 85, 970.
(3) Kim, Y.; Choulis, S. a.; Nelson, J.; Bradley, D. D. C.; Cook, S.; Durrant, J. R. Appl.
Phys. Lett. 2005, 86, 063502.
(4) Wang, L.; Liu, Y.; Jiang, X.; Qin, D.; Cao, Y. J. Phys. Chem. C 2007, 111, 9538–
9542.
(5) Kang, Y.; Kim, D. Sol. Energy Mater. Sol. Cells 2006, 90, 166–174.
(6) Ram, M. K.; Sarkar, N.; Bertoncello, P.; Sarkar, A.; Narizzano, R.; Nicolini, C.
Synth. Met. 2001, 122, 369–378.
(7) Verma, D.; Ranga Rao, A.; Dutta, V. Sol. Energy Mater. Sol. Cells 2009, 93, 1482–
1487.
(8) Wu, M.-C.; Lo, H.-H.; Liao, H.-C.; Chen, S.; Lin, Y.-Y.; Yen, W.-C.; Zeng, T.-W.;
Chen, Y.-F.; Chen, C.-W.; Su, W.-F. Sol. Energy Mater. Sol. Cells 2009, 93, 869–
873.
(9) Al-Ibrahim, M.; Roth, H.-K.; Zhokhavets, U.; Gobsch, G.; Sensfuss, S. Sol. Energy
Mater. Sol. Cells 2004, 85, 13–20.
(10) Shiga, T.; Takechi, K.; Motohiro, T. Sol. Energy Mater. Sol. Cells 2006, 90, 1849–
1858.
(11) Zhou, Y.; Riehle, F. S.; Yuan, Y.; Schleiermacher, H.; Niggemann, M.; Urban, G. a.;
Kruger, M. Appl. Phys. Lett. 2010, 96, 013304.
151
(12) Zarazúa, I.; De la Rosa, E.; López-Luke, T.; Reyes-Gomez, J.; Ruiz, S.; Ángeles
Chavez, C.; Zhang, J. Z. J. Phys. Chem. C 2011, 115, 23209–23220.
(13) Huynh, W. U.; Dittmer, J. J.; Alivisatos, a P. Science 2002, 295, 2425–2427.
(14) Olson, D. C.; Piris, J.; Collins, R. T.; Shaheen, S. E.; Ginley, D. S. Thin Solid Films
2006, 496, 26–29.
(15) Wang, M.; Wang, X. Sol. Energy Mater. Sol. Cells 2008, 92, 766–771.
(16) Roberson, L. B.; Poggi, M. A.; Kowalik, J.; Smestad, G. P.; Bottomley, L. A.;
Tolbert, L. M. Coord. Chem. Rev. 2004, 248, 1491–1499.
(17) Balis, N.; Dracopoulos, V.; Stathatos, E.; Boukos, N.; Lianos, P. J. Phys. Chem. C
2011, 115, 10911–10916.
(18) Slooff, L. H.; Wienk, M. M.; Kroon, J. M. Thin Solid Films 2004, 451-452, 634–638.
(19) Zhou, Y.; Li, Y.; Zhong, H.; Hou, J.; Ding, Y.; Yang, C.; Li, Y. Nanotechnology
2006, 17, 4041–4047.
(20) Acharya, K. P.; Hewa-Kasakarage, N. N.; Alabi, T. R.; Nemitz, I.; Khon, E.; Ullrich,
B.; Anzenbacher, P.; Zamkov, M. J. Phys. Chem. C 2010, 114, 12496–12504.
(21) Vu, T. T. D.; Mighri, F.; Do, T.-O.; Ajji, A. J. Nanosci. Nanotechnol. 2012, 12,
2815–2824.
(22) Vu, T. T. D.; Mighri, F.; Ajji, A.; Do, T. Ind. Eng. Chem. Res. 2014, 53, 3888–3897.
(23) Liu, J.; Wang, W.; Yu, H.; Wu, Z.; Peng, J.; Cao, Y. Sol. Energy Mater. Sol. Cells
2008, 92, 1403–1409.
(24) Joo, J.; Kwon, S. G.; Yu, T.; Cho, M.; Lee, J.; Yoon, J.; Hyeon, T. J. Phys. Chem. B
2005, 109, 15297–15302.
(25) Li, X.-L.; Peng, Q.; Yi, J.-X.; Wang, X.; Li, Y. Chemistry 2006, 12, 2383–2391.
(26) Jun, Y.-W.; Casula, M. F.; Sim, J.-H.; Kim, S. Y.; Cheon, J.; Alivisatos, A. P. J. Am.
Chem. Soc. 2003, 125, 15981–15985.
(27) Chen, S.; Liu, W. Mater. Chem. Phys. 2006, 98, 183–189.
(28) Chen, M.; Feng, Y.-G.; Wang, X.; Li, T.-C.; Zhang, J.-Y.; Qian, D.-J. Langmuir
2007, 23, 5296–5304.
(29) Limaye, M. V; Singh, S. B.; Date, S. K.; Kothari, D.; Reddy, V. R.; Gupta, A.; Sathe,
V.; Choudhary, R. J.; Kulkarni, S. K. J. Phys. Chem. B 2009, 113, 9070–9076.
152
(30) He, J.; Kanjanaboos, P.; Frazer, N. L.; Weis, A.; Lin, X.-M.; Jaeger, H. M. Small
2010, 6, 1449–1456.
(31) Si, H.; Wang, H.; Shen, H.; Zhou, C.; Li, S.; Lou, S.; Xu, W.; Du, Z.; Li, L. S.
CrystEngComm 2009, 11, 1128.
(32) Raupach, M. Clays Clay Miner. 1976, 24, 127–133.
(33) Nguyen, T.; Mrabet, D.; Vu, T.-T.-D.; Dinh, C.-T.; Do, T.-O. CrystEngComm 2011,
13, 1450.
(34) Tzitzios, V.; Niarchos, D.; Margariti, G.; Fidler, J.; Petridis, D. Nanotechnology
2005, 16, 287–291.
(35) Chen, L. X.; Rajh, T.; Jäger, W.; Nedeljkovic, J.; Thurnauer, M. C. J. Synchrotron
Radiat. 1999, 6, 445–447.
(36) Persson, P.; Bergström, R.; Lunell, S. J. Phys. Chem. B 2000, 104, 10348–10351.
(37) Murphy, A. Sol. Energy Mater. Sol. Cells 2007, 91, 1326–1337.
(38) Sabah, A.; Siddiqi, S. A.; Ali, S. World Acad. Sci. Eng. Tech. 2010, 45, 82–89.
(39) Wang, C. L.; Zhang, H.; Zhang, J. H.; Li, M. J.; Sun, H. Z.; Yang, B. J. Phys. Chem.
C 2007, 111, 2465–2469.
(40) Günes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. Rev. 2007, 107, 1324–1338.
(41) Adikaari, A. A. D. T.; Dissanayake, D. M. N. M.; Silva, S. R. P. IEEE J. Sel. Top.
Quantum Electron. 2010, 16, 1595–1606.
(42) Kim, S.-S.; Jo, J.; Chun, C.; Hong, J.; Kim, D. J. Photochem. Photobiol. A Chem.
2007, 188, 364–370.
(43) Yang, B. D.; Yoon, K. H.; Chung, K. W. Mater. Chem. Phys. 2004, 83, 334–339.
(44) Salafsky, J. Phys. Rev. B 1999, 59, 10885–10894.
(45) Liu, Y.; Gibbs, M.; Puthussery, J.; Gaik, S.; Ihly, R.; Hillhouse, H. W.; Law, M.
Nano Lett. 2010, 10, 1960–1969.
(46) Xiong, G.; Shao, R.; Droubay, T. C.; Joly, a. G.; Beck, K. M.; Chambers, S. a.; Hess,
W. P. Adv. Funct. Mater. 2007, 17, 2133–2138.
(47) Greenham, N. C.; Peng, X.; Alivisatos, A. P. Phys. Rev. B 1996, 54, 628–637.
.
153
Chapter 7. Conclusion
7.1. General conclusions
During the course of this research work, a number of conclusions have been
reached, regarding the preparation, characterization and utilization of TiO2 based
photocatalysts in the H2 production via water splitting and in BHJ solar cells. In addition,
recommendations and suggestions for future work in the area of photocatalytic hydrogen
production and photovoltaic are also evaluated.
Firstly, we recommended a new simple synthesis routine using solvothermal method
in autoclave at low temperature to synthesize TiO2 nanoparticles. The shape and size of
those TiO2 NPs were able to be controlled just by adjusting the reaction condition,
especially the capping agents. In details, by varying the presence of different capping
agents, which include oleic acid, oleylamine, and amino 6-aminohexanoic acid, we
achieved anatase TiO2 NPs with shape of sphere, rod, and rhombic; and with size of
nanorod from 3 x 40 nm to 3 x 20 nm. Based on that, the effects of capping agent on the
TiO2 particles morphologies as well as optical properties were studied, analyzed and
understood.
Secondly, following the first step of the study, in order to enhance the optical
properties of TiO2 NPs in visible light region, we proposed a two-step-synthesis process to
deposit CdS nanoparticles on the surface of TiO2 nanorods: steps (i) the surface properties
of the achieved TiO2 nanorods were modified by surfactant exchange with nitrosonium
tetrafluoroborate (NOBF4) in the direction to promote the hydrophilic properties of
nanorods. (ii) CdS NPs were then deposited on surface of NOBF4-TiO2 nanorods. Several
optical characterizations was done for CdS/TiO2 and it showed that the nanocomposite has
a great potential for the application both in photocatalyst H2 production via water splitting
and in BHJ solar cells, especially in visible light region.
In chapter 4, we presented the application potential of achieved CdS/TiO2
nanocomposite in photocatalyst H2 production via water splitting. Based on CdS/ OA-
6AHA-capped-TiO2, Ni clusters were selectively deposited on surface of nanocomposite
surface as cocatalyst. Under visible-light illumination and due to the fact that electron-hole
154
separation was highly enhanced, the developed CdS/TiO2 -Ni photocatalyst showed a high
photocatalytic performance for the H2 production via water splitting using ethanol as a
sacrificial agent, which was around 44 times higher than that of Ni-CdS. With the coupling
with TiO2, the system appeared to be resistant to photo-corrosion, which usually was a
concern when using photocatalyst that contains CdS particles. The photocatalytic activity
of the system can be expected to run up to 15h of reaction.
Potential application of TiO2 NPs and their CdS/TiO2 nanocomposite in BHJ solar
cells was demonstrated in chapter 6 and showed very promising results. The power
efficiency conversion (PCE) of devices using active layer of MEH-PPV and CdS/OA-
6AHA-capped-TiO2 or CdS/OA-OM-capped-TiO2 nanocomposites were reported to
increase 11.9 and 16.7 times compare to device using active layer combine of MEH-PPV
and only OA-6AHA-capped-TiO2 NPs or OA-OM-capped-TiO2, respectively. In this
demonstration, the BHJ device architecture was designed, optimized and fabricated using
the available and simplest method in the laboratory, include solution blending,
photolithography, spin-coating, and thermal evaporation. The advantage of the device
architecture is that photogenerated charges can be collected by the electrodes easily due to
direct pathways. Where else the active layer, which is the most important layer of BHJ
solar cells, was combined of MEH-PPV conjugated polymer and TiO2 nanorods or TiO2-
based nanocomposites.
Finally, the effects of different capping agents (OA/OM and OA/6AHA) on
properties of BHJ solar cells based on TiO2 NPs and their CdS/TiO2 nanocomposite were
carefully studied in chapter 5 and 6. The work presented in chapter 5 found that the amount
of OA and OM capping agents used in synthesis process have effects on the shape and
surface properties of TiO2 NPs. As resulted, the capping agent posed many effects on the
energy bandgap and also on the dispersion quality of MEH-PPV/TiO2 nanocomposites.
Among those three different morphologies of OA-OM-capped-TiO2 achieved, TiO2
nanorods were found to have higher surface-to-volume ratio, higher delocalization of
carriers, and sufficient amount of surfactant capped on surface of particles compared to
TiO2 nanosphere and TiO2 nanorhombic. Though they helped to increase the probability of
charge transfer at polymer-NPs interface, and helped to prevent the back recombination of
155
holes in MEH-PPV matrix and electrons in TiO2 NPs. Therefore, TiO2 nanorods were
reported to be the most efficient for utilizing in BHJ solar cells.
In chapter 6, further studies were focused on the evaluation of effects of different
surfactants on photovoltaic PCE of BHJ solar cells. The results showed that the PCE
increase 3.2 times when using OA-6AHA-capped-TiO2 nanorods (PCE = 0.009%) instead
of using OA-OM-capped TiO2 nanorods (PCE = 0.0028%). The better in PCE was
explained due to the capping agent factor, in which the length of carbon-chain of 6AHA is
much shorter than of OM. The carrier mobilities were reported to be decrease exponentially
with increasing ligand length, hence with the usage of OM, the electron and holes would be
very fast recombination before effectively transferred from MEH-PPV to TiO2, as resulted,
the lower PCE was expected. However, the poor photovoltaic response of MEH-PPV and
TiO2 NPs could be ascribed due to the surfactant NPs, which may prevent such kind of loss
in charge separation in these particular types of devices. In agreement, PCE of BHJ using
CdS/OA-6AHA-capped-TiO2 was 2.3 times higher compares to the device using when
using CdS/OA-OM-capped-TiO2 nanocomposites.
7.2. Prospects
There are several directions that can be further pursued in the future:
1. Further study the effects of many other different capping agents (for example n-
octyl-phosphonic acid, thiophenol, pyridyl, etc) on the performance of TiO2 and
its derivatives in BHJ solar cells and in H2 production via water splitting.
2. Further investigate the electron-hole separation, recombination and
transportation mechanism between TiO2 and another smaller band gap energy
semiconductor, in order to optimize their application in photocatalyst water
splitting as well as in BHJ solar cells.
3. Further study the effects of TiO2-based nanocomposite which compounds of
surfactant-capped-TiO2 and other metal chalcogenides on PEC of BHJ solar
cells.
4. Combine surfactant-capped-TiO2 with different chemical compositions such as
carbon nanotube, N-doped GaN-Zn, etc. to enhance their photocatalytic activity.
157
Annex A - Aminoacid-asisted Synthesis of TiO2 Nanocrystals
with Controllable Shape and Size: A Novel Agent for the
Fabrication of Polymer/TiO2 Photovoltaic Materials
Thi Thuy Duong Vu a,b, Frej Mighri a,b,*, Trong-On Doa,c Abdellah Ajjib,d, Jayesh D. Pate a,b
aDepartment of Chemical Engineering, Laval University, Quebec, QC, G1V 0A6 Canada;
b Center for Applied Research on Polymers and Composites (CREPEC);
cCentre in Green Chemistry and Catalysis (CGCC);
dDepartment of Chemical Engineering, École Polytechnique of Montreal, C.P. 6079, Montreal, QC,
H3C 3A7 Canada.
Published in Proceedings of the Polymer Processing Society 26th Annual Meeting, PPS-26
~ July 4-8, 2010 Banff (Canada).
158
Abstract
This paper reports the synthesis of TiO2 nanoparticles (NPs) with different
morphologies using oleic acid (OA) and oleylamine (OM) as surfactants. These TiO2
nanoparticles were then dispersed in polystyrene (PS) matrix using the drop-coating
technique. Two different NPs shapes (nanosphere and nanobar) were achieved by varying
the ratio of surfactants. Transmission electron microscopy (TEM) clearly shows these two
shapes: The nanosphere shape had an average size of around 5nm and the the nanobar
shape had an average size of 10 nm x 20 nm. The dispersion of TiO2 NPs in PS matrix was
characterized by scanning electron microscopy (SEM) technique on TiO2/PS
nanocomposite films. SEM results showed that the best dispersion of TiO2 was achieved
with the nanosphere shape. Optical absorption spectrum of TiO2/PS nanocomposite films
showed a strong absorption in visible region. It also showed that nanobar TiO2/PS films
absorb more in visible region compared to nanosphere TiO2/PS films. This may be due to a
combined effect of NPs size and shape and also to NPs dispersion. Fourier transform
infrared spectroscopy of TiO2/PS films showed a weak interaction between NPs and the
host PS matrix.
159
Résumé
Cet article élucide la synthèse des nanoparticules TiO2 (NPs) avec des différentes
morphologies en utilisant les acides oléiques (OA) et oleylamine (OM) comme surfactant.
Ces nanoparticules TiO2 ont été dispersées dans la matrice polystyrène (PS) en utilisant la
technique de drop-coating. Deux formes différentes de NPs (Nanosphère ou nanobare) ont
été obtenues en variant le rapport des surfactants. La microscopie électronique en
transmission (TEM) montre clairement deux formes : la nanosphère a une taille moyenne
aux alentours de 5 nm et la taille des nanobares est entre 10 nm et 20 nm. La dispersion des
TiO2 NPs dans la matrice de PS a été caractérisée par une microscopie électronique à
balayage (SEM) sur des films de nanocomposite TiO2/PS. Les résultats du SEM montrent
que la meilleure dispersion des TiO2 a été obtenue avec des nanosphères. Les spectres
d’absorption des films nanocomposites montrent une forte absorption dans la région du
visible. Les résultats montrent aussi que l’absorption des films de nanobars TiO2/PS dans la
région visible est plus forte que celle des films nanosphère TiO2/PS. Ceci peut être dû à
l’effet de la taille des NPs combiné à l’effet de la forme et de la dispersion des NPs. La
spectroscopie infrarouge à transformée de fourrier des films de TiO2/PS montrent des
interactions faibles entre NPs et la matrice de PS.
160
A1. Introduction
During the past decade, there have been considerable efforts in design and
controllable preparation of organic-inorganic nanocomposites with different morphologies
due to their potential application in various fields, such as catalysis, microelectronics,
optics and photovoltaic devices. PS is considered as a conventional low cost polymer with
good physical properties. However, PS does not absorb light in visible range and its optical
properties are poor. Zan et al.1 found that by adding TiO2 into PS polymer matrix, a
significant enhancement of PS optical properties was achieved because of its proper
bandgap (3.2 eV for anatase phase, 3 eV for rutile phase). These materials have promising
applications for efficient electron transportation.
In this paper, we report the synthesis and characterization of TiO2 nanoparticles
with different shapes and their use for the preparation of TiO2/PS composite thin films. The
optical properties of PS-TiO2 composites as well as the dispersion of TiO2 NPs in the
composite have been also evaluated.
A2. Experimental
All chemicals were used as received without further purification. Titanium (IV)
butoxide (TB, 97%), oleic acid (OA, 90%), oleylamine (OM, 70%); absolute ethanol and
toluene (analytical grade) were purchased from Aldrich. Polystyrene (PS-3900) was
purchased from Nova Chemical.
The synthesis of OA capped anatase TiO2 nanocrystals with various shapes
(nanosphere and nanobar) were synthesized by hydrolysis of TB using OA and OM as
surfactants at low temperatures followed by the method reported in the literature.2
Typically, 1 mmol of TB was added to a mixture of OA, OM and ethanol. The system was
then heated at 150oC for 18 h. The resulting TiO2 nanocrystals were washed with toluene
161
and ethanol several times then dried at room temperature. After purification, TiO2 was re-
dispersed in toluene for thin film fabrication.
A mixture of PS and TiO2 was prepared by solution blending process with a mole
ratio of PS:TiO2 is 1:1 in toluene solvent. Ultrasonic agitation was used to break up
nanoparticle agglomerates in the solvent and to ensure dispersion homogeneity. The
PS/TiO2 thin film was fabricated by drop-coating method. 5 drops of mixture were dropped
on surface of the glass then heated at 50oC to dry the solvent.
TiO2 size and morphology characterization were done at 120 kV using a JEOL JEM
1230 transmission electron microscope (TEM). Scanning electron microscopy (SEM)
analyses were carried out to observe the overall morphology of the composite on the JEOL
6360 instrument with an accelerating voltage of 15 kV. The UV−visible spectrum of the
nanocrystals was recorded for the thin films on a Hitachi U-3010 spectrometer. Fourier
transform infrared absorption spectroscopy (FTIR) spectra were measured with a FTS 45
infrared spectrophotometer using KBr pellet technique.
A3. Results and Discussions
Figure A.1 shows TEM images of TiO2 nanoparticles capped by OA and OM,
which were synthesized using different molar ratios of OA/OM. TiO2 uniform nanospheres
with diameter of 5 nm were obtained using a molar ratio of 6:4 (Figure A.1a) and TiO2
nanobars (with size of 10 nm in width and 20 nm in length) were achieved using a molar
ratio of 8:2 (Figure A.1b).
162
Figure A.1. TEM images of TiO2 nanoparticles (a) TiO2 nanospheres with an average size of 5
nm (b) TiO2 nanobars with an average size of 10 nm x 20 nm.
The dispersion quality of TiO2 nanoparticles into the PS matrix was evaluated by
SEM. It could be obviously seen from Figure A.2 that TiO2 nanospheres show better
dispersion in PS matrix (Figure A.2b) than TiO2 nanobars (Figure A.2a). This could be due
to the difference in shape and size between nanospheres and nanobars. Nanospheres with
smaller size showed better dispersion into PS matrix.
163
Figure A.2. SEM images of (a) nanobars TiO2/PS and (b) nanosphere TiO2/PS films
Figure A.3 shows the UV-vis absorption spectra for the samples of glass, pure PS
film, and nanobars TiO2/PS and nanospheres TiO2/PS films. With adding TiO2 NPs into PS
matrix, the absorption band of the composite was shifted to visible light range, which is
promising for photovoltaic devices. However, the absorption of nanobar TiO2/PS films was
164
significantly higher than that of nanosphere TiO2/PS. This could be due to the difference in
shape and size of TiO2 nanoparticles, which can affect their electronic state (energy
bandgap).3
Figure A.3. UV-vis absorption spectra for the samples of glass, pure PS film, TiO2
nanobars/PS and TiO2 nanosphere/PS films.
FTIR spectra corresponding to nanospheres TiO2/PS, nanobars TiO2/PS, and pure
PS films are shown in Figure A.4. As seen in the spectrum of TiO2/PS films, the peaks of
phenyl ring are at 1496, 1450, 756 and 701 cm-1, and they are same as in pure PS film.
However, there is a significant decrease in intensity at 756 and 701 cm-1. This could be due
to the phenyl ring opening reaction in PS. The spectra also show a growth of carbonyl
group (C=O) in the band region 1700-1710 cm-1.1 The existence of C=O proves the
oxidation of some –CH2– to –CO– on the chain of PS during the heat treatment, even at
low temperature. In a previous study done by Su et al.,4 the authors obtained a C-O-Ti band
at 1265.78 cm-1; composite films were prepared using ionic polymerization. However, this
band was not clearly observed in our case by using the solution blending technique, which
has more advantages in controlling the shape and size of doped-nanoparticles.
165
Unfortunately, the latter technique has disadvantage in creating strong coupling between
nanoparticles and polymer matrix.
Wavenumber (cm-1
)
1000200030004000
Tra
nsm
itta
nce (
%)
20
40
60
80
100
(a)
(b)
(c)
Figure A.4. FTIR spectra of (a) nanosphere TiO2/PS film (b)nanobar TiO2/PS film, and (c) pure PS
film.
A4. Conclusion
In summary, various shapes of TiO2 nanoparticles (nanospheres, nanobars) can be
controlled by changing the reaction parameters. TiO2/PS thin films were successfully
prepared by solution blending and drop-coating techniques. The dispersion quality of TiO2
NPs in PS matrix films as well as optical properties of these films were characterized.
Results showed a promising application in solar cell devices. The next step of our study is
to replace PS by MEH-PPV (2-methoxy-5-(2’-ethylhexyloxy)-p-phenylene vinylene)
conjugated polymer.
166
Acknowledgements
The authors would like to thank the Natural Sciences and Engineering Research
Council of Canada (NSERC) for financial support of this work.
A5. Reference
(1) Zan, L.; Tian, L.; Liu, Z.; Peng, Z. Appl. Catal. A Gen. 2004, 264, 237–242.
(2) Dinh, C.; Nguyen, T.; Kleitz, F.; Do, T. ACS Nano 2009, 3, 3737–3743.
(3) Li, J. Nano Lett. 2003, 3, 1357–1363.
(4) Su, B.; Ma, Z.; Min, S.; She, S.; Wang, Z. Mater. Sci. Eng. A 2007, 458, 44–47.
