xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003312
Fabrication and characterization of surfactant-free pbse quantum dot films and pbse-polymer hybrid structures
h [electronic resource] /
by Gayan Dedigamuwa.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains X pages.
Dissertation (Ph.D.)--University of South Florida, 2010.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
ABSTRACT: ABSTRACT This work describes an experimental investigation of methods of synthesis, determination of structural and physical properties, and analysis and correlation of the properties to the structures of semiconductor quantum dots and quantum dot-polymer hybrid structures. These structures are investigated for applications in flexible solar cell devices. The main synthesis process used in the work was a Laser-Assisted Spray (LAS) process that was developed in our laboratory to deposit surfactant-free PbSe quantum dot (QD) films directly on a substrate. The QD films formed by this technique are in close contact with each other forming a percolation path for charge transport. Analytical instruments that include Atomic Force Microscopy (AFM) and Transmission Electron Microscopy (TEM) were used for structural characterization while optical absorption spectroscopy and photoluminescence were used for determining the quantum confinement of charge carriers in PbSe QDs. In addition, charge transport across lithographically patterned paths was used to determine the transport characteristics and generation of photocurrent in the fabricated structures. Absorption spectroscopy confirmed the quantum confinement of PbSe QDs deposited by LAS deposition. Room temperature current-voltage measurements across a 2m tunnel junction formed by the QDs produced a power-law dependence of the form that describes a percolation path of dimensionality slightly above two-dimensional. Absence of surfactants in LAS deposited films improved the conductivity by more than three orders of magnitude. Temperature dependent conductance studies showed thermally activated transport at high temperatures and temperature independent tunneling followed by previously unobserved metallic conduction at low temperatures. The LAS system was successfully modified by incorporating two spray nozzles to transport aerosols of two different precursors, one containing the QDs and the other containing the polymer. This new co-deposition system was successfully used to deposit QDs/Polymer hybrid structures. The TEM and XRD studies of LAS co-deposited films were shown to be uniformly distributed and crystalline. The photo-current experiments of QD/polymer hybrid composites showed clear evidence of enhanced carrier generation and transport as a result of intimate contact between quantum dots (QDs).
Advisor: Sarath Witanachchi, Ph.D.
Fabrication and Characterization of Surfactant-Free PbSe Quantum Dot Films and PbSe-Polymer Hybrid Structures
t USF Electronic Theses and Dissertations.
Fabrication and Characterization of Surfactant Free PbSe Quantum Dot Films and PbSe Polymer Hybrid Structure s by Gayan S. Dedigamuwa A dissertation submitted in partial fulfillment of the requirement for the degree of Doctor of Philosophy Depar tment of Physics College of Arts and Sciences University of South Florida Major Professor: Sarath Witanachchi, Ph.D. Pritish Mukherjee, Ph.D. Hariharan Srikanth, Ph.D. Xiaomei Jiang, Ph.D. Date of Approval : March 22 2010 Keywords: thin films, lead selenide quantum dots nanoparticles, multiple excit on generation, ex citon dissociation, solar cells Copyright 2010 Gayan S. Dedigamuwa
ACKNOWLEDGMENTS I would like to thank Dr. Sarath Witanachchi, my advisor for his suggestions and unwavering support. Everything I know about research has come from him. His vast experience and creativity helped me greatly along the way, and were esse nt ial to completion of this dissertation I would also like to thank Dr. Pritish Mukherjee Dr. Hariharan Srikanth and Dr. Xiaomei Jiang for agreeing to serve on my Committee and p roviding advice Further, I would also like to thank Dr. Xiaomei Jiang and D r. Zhang for assisting me with colloidal PbSe nanoparticle growth and conducting optical measurements of my samples Also I would like to thank all my lab mates specially Robert Hyde and Postdoctoral Associate Dr. Tara Dhakal for t he assistance they gave m e throughout this project. And last, but not least, I would like to thank my wife Nilusha and my parents Premadase and Jayanthi who encourag ed me along and helped me through the difficult times.
i TABLE OF CONTENTS LIST OF FIGURES i ii ABSTRACT viii CHAPTER 1. INTRODUCTION AND BACKGROUND 1 1.1. Advantage of Nanocrystalline Materials 1 1.2. Semiconduct or Nanocrystals (Quantum Dots) 3 1.3. Semiconductor base Traditional Solar Cell 11 1.4. Absorption of Solar Radiation by Semiconductors 13 1.5. PbSe Quantum Dots (QDs) 15 1.5.1 Exciton Dissociation 19 CHAPTER 2. NANOCRYSTALLINE GROW TH TECHNIQUES & PROPERTY CHARACTERIZ ATION TECHNIQUES 23 2.1. Nanopar ticle Synthesis 23 2.1. 1. Chem ical Synthesis of Nanoparticles 23 2. 2. Spray Pyrolysis 25 2.3 Laser A ssisted Spray Process 26 2.4 Structural Characterization of Nanoparticles 29 2.4 .1. Tra nsmission Electron Microscopy 29 2. 5. Optical Characterization of Nanoparticles 30 2. 5. 1 Photoluminescence 32 CHAPTER 3. EXPERIMENTAL PROCEDU RE 3 5 3.1. Synthesis of S urfactant free PbSe Nano cry s talline F ilms 3 5 3.1.1. Preparation of PbSe Colloidal Precursor 3 5 3.1.2. Laser Assisted Spray (LAS) Deposition of PbSe Q uantum D ots 37 3.2. Sample Preparation for T EM Studies 38 3.3. Sampl e Preparation for Conductivity M easurements 39 3.3.1. Deposition of PbSe QD layer on Electrodes 40 3.4. Synthesis of P3HT Polymer Coating 42
ii 3.5. Co Deposition of QD/Polymer Composite F ilms 43 3.6. Fabricati on of PbSe QDs/P3HT Solar Cell S tructures 45 CHA PTE R 4. EXPERIMENTAL RESULTS 47 4.1. Charact eristics of Self A ssembled PbSe nanoparticles 47 4.1.1. Op tical Characterization of PbSe N anoparticles 53 4.1.2. Structure, Morphology, and Optical Properties of PbSe Nanoparticle Coatings Deposited by LAS P rocess 57 4.1.3. Elec trical Cond uctivity of PbSe QD Coatings D eposited by the LAS P rocess 61 4.1.4. Temperature Dependent Conductivity M easurement 64 4.1.5. Photocurrent Measurement 67 220.127.116.11. Direct Current (DC) Measurements 67 18.104.22.168. Photo Generated Current by a Pulse Laser 69 4.2. Characteri za tion of P3HT Polymer C oat ings 7 1 4.2.1. Crystallinity 7 1 4.2.2. Optical C haracterization 7 3 4.2.3. Photo Ge nerated Current by Pulse Laser E xcitation 74 4.3. Characterization of PbSe QDs/P3HT Hybrid Structures 75 4.3.1. Photo Generated Current Measurements 77 4.3.2. Frequency Dependent Photo Generated Current Measurement s 79 4.4. Photo Generated Current Measurement of the Solar Cell Structure 84 CHAPTER 5. DISCUSSION 86 5. 1. S ynthesis of PbSe Nanocrystals 86 5. 2. LAS D eposition of Surf actant Free PbSe Nanocrystals 86 5.2.1 Charge Transport in QD Films 87 5.2.2. Pho toconducti vity 89 5.2 3 Temperature De pendent Conductivity 89 5.3 Fabrication of PbSe QDs/P3HT Hybrid Structure by LA Co Deposition 91 REFERENCES 94 ABOUT THE AUTHOR End Page
iii LIST OF FIGURES Figure 1 .1. The image above represents the fluo rescent emission colors of cadmium sulfide nanocrystals, or "quantum dots," excited with a near ultraviolet lamp 2 Figure 1.2 The progression of confinement and effects on the density of states 7 Figure 1.3 Change in band gap energy E g with particle size. 7 Figure 1.4 Linear absorption spectra at 10K of CdS crystalline deposited on small crystal sizes in the quantum co nfinement regime. The bulk sample has large crystal sizes, exhibiting bulk CdS properties. 8 Figure 1.5. Schematic representation of the electron hole pair transition in a semiconductor quantum dot. 9 Figure 1.6 An example of size related absorbance of the CdSe QDs. The peaks determine the 1S transition of each QDs size distribution. (Image taken from Sigma Aldrich Material Sciences website.) 11 Figure 1.7. Schematic diagram of a semiconductor based traditional pn junction solar cell (Fig taken from J. Nozik, NREL report). 12 Figure 1.8. Air Mass (AM) 1.5 solar spectrum. The dashed lines in green and red represent limits of organic and silicon based solar cells, respectively. 13 Figure 1.9. Room temperature optical absorption spectra for a series of Pb Se NC samples measuring (a) 3.0 nm, (b) 3.5 nm, (c) 4.5 nm, (d) 5 nm, (e) 5.5 nm, (f) 7 nm, (g) 8 nm, and (h) 9 nm in diameter. (C. B. Murray et al .) 14 Figure.1.10. Bandgap increment in various semiconductor QDs due to quantum confinement (from bulk to 3. 9nm). (Taken from H. Weller, Pure Appl. Chem. 72, 295 (2000).) 16
iv Figure 1.11 Multiple exciton generation of quantum dot PV solar cells ( Taken from J. Nozik, NERL) 17 Figure 1.12 Comparison of energy levels and threshold energy for MEG in QDs for (a) m e ~ m h 18 10 Figure 1. 13 The schematic energy level diagram for PbSe nanodots and P3HT(polymer) showing the charge transfer of electrons to PbSe and holes to P3HT. 20 Figure 1. 14 Possible paths for exciton dissociation in polymer/QD composites. (a) Generation of an exciton in polymer followed by electron transfer to QD by Dexter mechanism (b) Generation of an exciton in polymer followed by exciton transfer to QD by Forster mechanism. Exciton in QD can dissociate by transferring the hole to the polymer. (c) Generation of an exciton in QD and transfer of the hole to the polymer by Dex ter mechanism. 22 Figure 2.1. The figure shows the Surfactant coated nanoparticle. Head groups of the surfactant attaches to the particle while tails are extended to the surrounding liquid (Taken from http://www.freepatentsonline .com/) 25 Figure 2.2 Particle size difference between Regular spray pyrolysis vs Laser assisted spray pyrolysis (a) AFM image of film deposited by regular spray pyrolysis (b) AFM ima ge of the film deposited by LASP technique. 2 8 Figure 2.3 TEM image of PbSe quantum dot, which produced by sol gel technique. The image shows the spacing d for 200 planes. 30 Figure 2 .4 Absorption spectra of chemically deposited lead selenide. (Taken The abso rption spectra of solid lead sulfide selenide, telluride 31 Figure 2.5 Photoluminescence spectra from various size Si nanoparticles The unit.aist.go.jp/.../lanproc/en/contents.html 33 Figure 2 .6 Schematic diagram of the Photoluminescence se t up. 34 Figure 3.1. Set up for synthesis PbSe NC colloidal solutions Taken from C. B. Murray et al. ) 36 Figure 3.2. Schematic diagram of the laser assisted spray pyrolysis film growth system. 37
v Figure 3 .3. The graph of temperature of the plume Vs flow rate of SF 6 gas. The line shows the opt imum flow rate of the system. 38 Figure 3.4. SEM images of Au/Ti electrodes fabricated for conductivity measurements. (a) ga p sizes. (b) and 40 Figure 3.5. Method of using a Shadow mask for QDs deposition 4 1 Figure 3.6. Schematic diagram of Laser Assisted Co deposition system 44 Figure 3.7 Schematic layout of the PbSe P3HT based hybrid solar ce ll structure. 46 Figure 4.1. PbSe particles separated from the solut ion after (a) 20s (b) 60s (c) 12 0s. 48 Figure 4.2. High resolution TEM image of individual PbSe se miconductor nanocrystal. The single crystal s tructure is clearly apparent. 49 Figure 4. 3. Electron diffractogram of the PbSe nanocrystals 50 Figure 4.4. X ray diffraction pattern of the samples made by three different particle distributions (a) 6 7nm (b) 9 10nm (c) 10 12nm. 51 Figure 4.5 Table of Miller indexes and Lattice parameters of P bSe nanoc r ystals calculated from XRD peaks obtain ed from Fig. 4.4. 5 2 Figure 4.6. Near infrared absorption spectra of as prepared PbSe semiconductor nanocrystals immersed in PCE. (Provided by Dr. Jiang's lab) 53 Figure 4.7. Change in bandgap energy with particle size; (red line) is experimental, (blue line) is calculated. 54 Figure 4.8. Absorbance and Ph otoluminescence spectra of 11 nm PbSe semiconductor nanocrystals. (Provided by Dr. Jiang's lab) 55 Figure 4.9. Absorbance and Ph otoluminescence spectra o f 9.5 nm PbSe semiconductor nanocrystals. (Provided by Dr. Jiang's lab) 5 6
vi Figure 4.10. Absorption spectra of 11 nm quantum dots in solution and after LAS deposition. There is a 29.1 Blue shift between them due to reduce QD sizes after burning the surf ac tant. In side: Zoom in absorption spectrum of LAS deposited sample (Provided by Dr. Jiang's lab) 57 Figure 4.11. TEM images of PbSe QDs deposited on carbon coated TEM grid for 1min with different plume temperatures. (a) drop casted (b) 80 100 o C (c) 150 20 0 o C (d) 200 230 o C. 59 Figure 4.12. High resolution TEM images of (a) drop casted and (b) LAS deposited PbSe NCs. LAS shows the intimate contact between the NCs while drop casted shows a 1 2nm surfactant coating between them. 60 Figure 4.13. TEM image of a coating deposited for 2mins at the optimum plume temperature 150 o C 200 o C. 61 Figure 4.14 Conductivity of PbSe QD films deposited by LAS technique across With Laser OFF. (c) Au/Ti Electrodes on glass substrate. (Provided by Dr. Jiang's lab) 62 Figure 4.1 5 Current Vs Voltage graph for different gap sizes. W/L: Laser is ON and W O/L: Laser OFF. (Provided by Dr. Jiang's lab) 63 Figure 4.16 Change in conductivity with the gap size. The current values were obtained at 5V bias volt age 64 Figure 4.17 Sample made for Temperature depende nt conductivity measurements. 65 Figure 4.18 versus the inverse of T for three PbSe QDs samples deposited by LASP technique. 66 Figure 4.19 Sample made for photocurrent measurement by Laser Assist ed Spray technique. 68 Figure 4.20 The graph of app lied voltage Vs current density through surfactant free PbSe quantum dots film. 69 Figure 4.21 Circuit diagram for detecting photocurrent generated by a pulse laser 70
vii Figure 4.22 The graph of photo generated current from the QD device at various lase r power levels. 71 Figure 4.23 XRD pattern of P3HT polymer coating deposited by LASP technique on glass substrate heat up to 60 o C 72 Figure 4.24 A bsorption spectrum of P3HT films deposited by Spray Pyrolysis at substrate temperature 200 o C, 100 o C an d 60 o C compared to Spin coated film (Provided by Dr. Jiang's lab) 73 Figure 4.25 The graph of pho to generated current in the P3HT layer sandwich between ITO and Al electrodes at various laser power levels. 75 Figure 4.26 XRD pattern o f PbSe QDs/P3HT hybrid composite deposited by laser Assisted Co Deposition. 76 Figure 4 .2 7 TEM image shows the initial formation of P3HT polymer and the PbSe NC on the film. 77 Figure 4.28 The graph of pho to generated current in the PbSe QDs/P3HT layer sandwich betwee n ITO and Al electrodes at various laser power levels compare d to PbSe QDs and P3HT polymer separately 78 Figure 4.29 The graph of calculated and experimental values for frequency dependent photo generated current through PbSe QDs/P3HT polymer composite 80 Figure 4.3 0 Graph of 1/I 2 Vs 1/ 2 82 Figure 4.3 1 Graph of 1/C Vs bias Voltage applied to the device structure 83 Figure 4.32. The graph of photo generated current in the PbSe QDs/P3HT Solar Device at various laser power levels when Forward and Reverse bias condition. 84 Figure 5.1 The versus the inverse of T for PbSe QDs samples d eposited by LAS technique compariso n with the calculated curves. 91 Figure 5.2 Energy level diagram of the PbSe QDs/P3HT based hybrid composite 92
viii FABRICATION AND CHARACTERIZATION OF SURFACTANT FREE PbSe QUANTUM DOTS FILMS AND PbSe POLYMER HYBRID STRUCTURES Gayan S. Dedigamuwa ABSTRACT This work describes an experimental investigation of methods of synthesis, determination of structural and physical properties, and analysis and correlation of the properties to the structures of semiconductor quantum dots and quantum dot polymer hybrid structures. These structures are investigated for applications in flexible solar cell devices. The main sy nthesis process used in the work was a Laser Assisted S pray (LAS) process that was developed in our laboratory to deposit surfactant free PbSe quantum dot (QD) films directly on a substrate. The QD films formed by this technique are in close contact with each other forming a percolation path for charge transport Analytical instruments that include Atomic Force Microscopy (AFM) and Transmission Electron Microscopy (TEM) were used for structural characterization while optical absorption spectroscopy and ph otoluminescence were used for determ ining the quantum confinement of charge carriers in PbSe QDs. In addition, charge transport across
ix lithographically patter n ed paths was used to determine the transport characteristics and generation of photocurrent in the fabricated structures. Absorption spectroscopy confirmed the quan tum confinement of PbSe QDs deposited by LAS deposition Room temperature current voltage measurements across a 2 m tunnel junction formed by the QDs produced a power law dependence of the form that describes a percolation path of dimensionality slightly above two dimensional Absence of surfactants in LAS deposited films improved the conducti vity by more than three orders of magnitude. Temperature depende nt conductance studies showed thermally activated transport at high temperatures and temperature independent tunneling followed by previously unobserved metallic conduction at low temperatures The LAS system was successfully modified by incorporating two spray nozzles to transport aerosols of two different precursors, one containing the QDs and the other containing the polymer. This new co deposition system was successfully used to deposit QD s/Polymer hybrid structures. The TEM and XRD studies of LA S c o deposited films were shown to be uniform ly distributed and crystalline The photo current experiments of QD/ polymer hybrid composites showed clear evidence of enhanced carrier generation and tr ansport as a result of intimate contact between quantum dots (QDs)
1 CHAPTER 1 I NTRODUCTION & BACKGROUND 1.1. A dvantage of Nanocrystalline Materials N anocrystalline materials have attracted tremendous interest in the chemistry and physics communities because of their novel properties Nanocrystalline materials exhibit exceptional mechanical properties, representing an exciting new class of structural materials for technological applications. The advancement of this important field depends on the development of new fabrication methods, n ew characterization methods and a n appreciation of the underlying nano scale and interface effects. The first major identifiable activity in the field of nanocrystalline (or nanostructured, or nanophase) materials was reported in the early 1980s by Gleiter and co workers 1 who synthesized ultraf ine grained materials by the in situ consolidation of nanoscale atomic clusters. The ultra small size (< 100 nm) of the grains in these nanocrystalline materials can result in dramatically improved or different properties from conventional grain size (> 1 m) polycrystalline or single crystal materials of the same chemical composition The difference s were attributed to two main physical phenomena that become prominent as the material dimension change s from three dimensions (3D) to zero dimension (0D). These include classi cal mechanical effects 2 as well as quantum mechanical effects, for example the
2 and optical properties of solids are altered due to changes in the band structures Additionally, the enhanced surface/volume ratio in nano dimensions forces more than 33 % of the atoms to be on the surface (for 10nm dot 35 ) drastically altering the physical properties such as having lower melting temperature and lower sintering temperature and higher diffusion force at elevated temperatures Electronic devices incorporating nanocrystalline materials are being extensively st udied to exploit the unique electrical and optical propert ies that arise from confinem ent of charge carriers A good example of the manipulation of optical properties is variation of emission spectra obtained fro m different size Cadmium Selenide (CdS e ) quantum dots 3 Fig. 1.1: The image above represents the fluo rescent em ission colors of cadmium selenide nanocrystals, or "quantum dots," excited with a near ultraviolet lamp 3
3 The excitation fluorescence depends on the nanocrystal size which is characteristic of the particle s band gap energy. As the physical dimensions of the particle become smaller, the band gap energy becomes higher. Therefore different size particles of the same material have different band gaps and therefore emit different colors. Since particle size determine s the optical and electrical properties, as demonstrated for CdS e ability to produce monodispersed pure nano particles with a narrow size distribution is very important The average size and the size distribution of the nanoparticles are largely determined by the technique that is used to produ ce them. Out of many techniques developed to fabricate nanoparticles chemical precipitation techniques are the most suitable for producing narrow particle distributions. 1.2. Semico nductor Nanocrystal s (Quantum Dots) In a bulk semiconductor material the e nergy levels are closely spaced within bands The number of available energy levels given in an interval of energy is describe d as the density of state In the case of a bulk semiconductor the to p occupied band, known as the vale nce band, is mostly fil led. The conduction band of a semiconductor is mostly empty, and it is separated from the valance band by a bad gap where the density of states within the gap for a pure semiconductor is zero. Electron s in the vale nce band are not free electrons because they are bonded by Coulomb forces formed by the n ucleus and the covalent bonds formed by the neighboring electrons. If an electron in the vale nce band gains energy that is greater than or e q ual to band gap energy, the electr on can breaks these bonds to dis sociate from the atom and become associated with the conduction band
4 yet subjected to the periodic pote ntials of the lattice. This process therefore, forms an electron and a hole in the conduction and valance bands, respectively. T his vacancy can be filled easily by anoth er electron in the vale nce band. I n other word s the vacancy can move free ly in the layers due to continuous density of state in the bulk material Since t his generated electron and hole pair can move freely in the lattice of the bul k semiconductor the pairs are unbounded When a semiconductor ha ving bulk properties is reduced in size to a few hundred atoms ( zero dimension) the density of s t at es in vale nce and conduction band is change d drastically. The continuous density of state in the bands is replaced by a set of discrete energy levels i.e S, P and D levels which may have energy level spacing e.g. E p E s much larger than the thermal energy k B T ( at room temperature ) T his new arrangement of density of s t at es in quantum l evel s change s the optical and electronic properties of the ma terial 4 In semiconductor quantum dot s excitation of an electron by optical radiation into the unfilled conduction band to create an electron hole pair re quires more energy that that required by a bulk semiconductor of the same material. 4 I n quantum dot the electron hole pairs generated by absorption of photons are not free as in bulk material because of the discrete nature of the energy levels. These bound electron hole pairs are known as excitons. Also, when th e electron and hole are confined within a particle dimension that is closer to or less than the Bohr exciton size they interact strong ly with one another through Coulomb forces As the size of a semiconductor nanocrystal decreases below the bulk semiconductor Bohr exciton size, the exciton experiences increasingly strong confinement effects which significantly alter the allowed energy levels and result in increasingly discrete electroni c structure s 4
5 In theory the separation of energy levels is mainly governed by the dimensionality of the semiconductor dot As expl ained in the above section, when the size of the material is reduced from bulk to nano, the density of state is changed from being continuous to discrete by replacing continuous energy band with discrete energy levels T hen t hese discrete levels can be found by solving the Schrodinger equation for different particle dimensions. In the case of a bulk semiconductor material electron wavefunction is delocalized and allow s electrons to move freely throughout the entire medium. Then, one can use free particle wavefuction without any confinement condition to solve the Schrodinger equatio n for find ing possible energy values According to this analysis th e energy of the electron in three dimensional (3D) free space can be expresse d as (1) w here k is a wave vector and m e is the effective mass of an electron. A s the dimensions of the semiconductor is reduced, the e nergy of the discrete states for 2D (thin films), 1D (nano wires) and 0D (quantum dots) for spatial dimensions of Lx, Ly, and Lz can be given by, For 2D (2) For 1 D (3) And for 0D (4)
6 Where , and for a bound quantum dot. Similarly, the energy eigen values for spherical quantum dot s are given by, 5 (5) (6) (7) I n these equations is the n th root of the th order Bessel functio n for the electron while is the n th root of the th order Bessel function for the hole. Fig. 1. 2 : change in band gap energy E g with particle size.
7 Additionally, E g is the band gap energy for the bulk semiconductor material. According to Eq 7 the effective band gap, the energy required by an e lectron to transfer from the highest val e nce band energy to the lowest conduction band energy, increases with decreasing particle size. T he Fig. 1.2 shows the increasing bandgap energy and intra band spacing with quantum confinement The quantum confinement effect is what allows the tuning of ab sorption and emission wavelengths over the entire visible spectrum as a function of quantum d ot size. As is shown in Fig. 1.2 and 1.3 the density of states changes from being continuous at all energy levels for a bulk material to becoming discrete transit ions at certain energy levels for the quantum dot. Fig.1.3 : The progression of confinement and effects on the density of states
8 Furthermore, the quantum mechanical wave functions of electrons and holes are confined within the limits o f the material. Hence, when the confinement increases, the wave functions also become more localized 5 Quantum confinement leads to altered emission lifetimes as well as altered luminescence quantum efficiency in quantum dots 6 Additionally, quantum confined structures exhibit a sh ifted band edge that allows the production of varied emission peak wavelengths as dictated by the strength of the confinem ent (i.e. the size of the quantum dot) (Fig1.4 ) Fig. 1.4 : Linear absorption spectra at 10K of CdS crystalline in glass. The two confinement regime. The bulk sample has large crystal sizes, exhibiting bulk CdS properties. 6
9 Fig. 1.4 display s the absorption spectra for bulk and quantum dot CdS sample at a temperature of T=10K. The lower temperature reduces the phonon broadening which w as not discussed w ithin the previous theoretical discussion of energy le vels and increasing confinement. S must be mentioned here. The two peaks observed in C dS quatum dot samples show the simultaneous electron hole transition between the inter bands during the light absorption. The first peak shows the 1S e 1S h 1st intraband transition, while second peak shows the 1S e 1P h /1S h 1P e 2 nd intraband transition (Fig. 1.5 ) The transition li nes in Fig 1.4 are much broader than transitions in bulk materials. The width of the transition results from Fig. 1.5: Schematic representation of the electron hole pair transition in a semiconductor quantum dot.
10 homogeneous and inhomogeneous broadening effects that cause the theoretically discrete transitions to become Gaussian distributions of transitions Homogeneous broadening is the result of phonon and other scattering effects that becaus e when looking at a set of quantum dots there is a distribution of sizes and aspect ratios that translate into a small distribution of emission wavelengths rather than performing as a single discrete transition. The electrical and optical properties of s emiconductor nanoparticles are mainly governed by the quantum size effect s T he size of the structure limits the exciton Bohr radius of the bound electron hole pairs, leading to altered electronic and optical properties, and causes high surface energy, whi ch alters the physical properties. 7 The physical material is actually smaller than the natural exciton Bohr radius (i.e. radius of lowest energy Bohr orbital). For Si this is reached at 4.9 nm, for PbSe at 6.1 nm defined to be: (8) w here e o is the dielectric constant of the Quantum Dot ( QD ) (at low and m r is the reduced electron hole mass. 7
11 1.3. Semiconductor b ased Conventional Solar Cell s Bulk semiconductors absorb optical photons to generate unbound e h pairs. Bandgap ( E g ) of the s emiconductor material plays an important role in the process of photo generation. In an ideal case, photons with an energy hv< E g will not contribute to the photo gener a tion, whereas all photons with an energy hv >E g will ea ch contribute to the photo ge nera ted electron hole pair followed by spontaneous recombination. If these elect r on hole pair s can be separated before they spontaneously recombine, a current can be generated. However, several energy loss mechanisms limit the current extraction from such a device. When an electron is excited to a level above the edge of the conduction band, the energy in excess of the band gap will be lost as heat as a result of e Fig. 1.6: An example of size related absorbance of the CdS e QDs The peaks determine t he 1S transition of each QDs size distribution (Image taken from Sigma Aldrich Material Sciences website.)
12 phonon coupling, and the electron is relaxed to the edge of the conduction band. Finally, the electron at the band edge will spontaneously recombine with a hole, radia tively or non radiatively Schematic diagram in Fig. 1.7 shows a semiconductor based traditional p n junction solar cell. As a result of above discussed energy loss mechanism s, and considering the fact that one photon generates only one e h pair, the maximum achievable efficienc y of a solar device has been computed to be 33%. Traditional single junction silicon sola r cells today exhibit efficiencies between 11 20% as a result of light loss due to reflection from the front surface of the cell, shadowing by the electrical contacts, and ohmic losses at the semiconductor/electrode junctions. Fig. 1.7 : Schematic diagram of a semiconductor based traditional pn junction solar cell (Fig taken from J. Nozik, NREL report).
13 1.4. Absorption of Solar Radiation by S emiconductors Examination of the solar spectrum (Fig. 1.8 ) reveals that the optimum coupling of sunlight should take place in the wavelength region of 300 1100 nm (4 eV 1.1 eV). Typical silicon solar cells (band gap 1.1eV) absorb all the radiation left of the red line (90 % of sun light). The Air mass 1.5 (AM1.5), the conventional spectrum used in the photovoltaic industry, corresponds to the sun being at an angle of elevation of 42 0 and at an integrated power density of 100 mW cm 2 Referring to the AM1.5 spectrum shown in Fig 1.8 close to 50% energy lies in the infrared. Fig. 1.8 : Air Mass (AM) 1.5 solar spectrum The dashed lines in green and red represent limits of organic and silicon based solar cells, respectively. (Taken from : www.solems.com/Radiometry basics )
14 As a result, the optimal bandgaps for solar cells in both the single junction and even the tandem architectures lie beyond 850 nm. High efficiency multijunction solar cells offer the prospect of exceeding 40% efficiency 8 through the inclusion of infrared bandgap materials. For double and triple junction solar cells, the smallest bandgap junction optimally lies at 1320 nm and 1750 nm respectively 8 When the absorber band gap is optimized around 1.1 1.4 eV, the excess energy of high energy photons are waste d as heat. This is one of the main limiting factors of the conventional p n junction solar devices. Fig.1.9 : Room temperature optical absorption spectra for a series of PbSe NC samples measuring (a) 3.0 nm, (b) 3.5 nm, (c) 4.5 nm, (d) 5 nm, (e) 5.5 nm, (f) 7 nm, (g) 8 nm, and (h) 9 nm in diameter. 53
15 1.5. P b S e Quantum Dots ( QDs ) PbSe is a direct band gap semiconductor with a band gap of 0.26eV that corresponds to absorption wavelength of 4770nm. As a result of the long wavelength PbSe nanoparticles found application in IR detectors. However, reducing crystal size to a few nanometers can bring the band gap into visible region A PbSe QD diameter of 5.7 nm will have a band gap of 0.72 eV corresponding to absorption wavelength of 1 720nm, compared to bulk of 0.26 eV. 6 While size tunability of QDs enable manipulation of the band gap, shown in Fig. 1.9 large Bohr radius (23nm) allow s PbSe to provide an alternate way to change the band gap and quantum levels without having to decrea se the QD size very much. This allows l arger QD sizes to have higher photon absorption cross sections Also the equal effective electron and hole mass in PbSe allows strong quantum confinement due to phonon bottle neck in both valance and conduction band, compared to other semiconductor materials such as Cadmium selenide, Indium phosphate, Gallium Arsenide etc. (Fig. 1.10 ). This strong quantum confinement in t he PbSe QDs provides a unique opportu nity for electro optical application s Also researchers have shown that due to phonon bottle neck in both valance and conduction band, the Multiple Exciton Generation (MEG) in PbSe NCs is highly efficient, extremely fast, and occurs in wavelength range that has a potential to provide significantly increased solar ce ll power conversion efficiency. 6
16 1.2.2. Multiple Exciton Generation (MEG) in QDs In bulk semiconductors, the absorption of one photon of energy greater than the bandgap leads to one electron hole pair by losing the excess energy as heat through electron phonon scattering and subsequent phonon emission, as the hot photogenerated c arriers (hot carrier) relax to their respective band edges ( Fig. 1.7 ) The main approach to reduce this loss in efficiency has been to use a stack of cascaded multiple p n junctions with bandgaps better matched to the solar spectrum (Fig. 1.8 ). In the limit of an infinite stack of bandgaps perfectly matched to the solar spectrum, the ultimate conversion efficiency at one sun intensity can increase to about 66%. 6 Another approach is to use the hot carriers before they relax to the band edge via phonon em ission. 6 There are two fundamental ways to use the hot carriers for enhancing the efficiency of photon conversion. One way produces an enhanced photovoltage, and Fig.1.10 : Bandgap increment in various semiconductor QDs due to quantum confinement (from bulk to 3.9nm). (Taken from H. Weller, Pure Appl. Chem. 72, 295 (2000).)
17 the other way produces an enhanced photocurrent. The former requires that the carriers be ex tracted from the photoconverter before they cool 6 To achieve the enhanced photovoltage the rates of photogenerated carrier separation, transport, and interfacial transfer across the contacts to the semiconductor must all be fast compared to the rate of carrier cooling T he latter requires the energetic hot carriers to produce a second (or more) electron hole pair (Fig 1.11 ). The formation of multiple electron hole pairs per absorbed photon is a process explained by impact ionization (I.I.). In this pro cess, an electron or hole with kinetic energy greater than the semiconductor bandgap produces one or more additional electron hole pairs. The kinetic energy can be created either by applying an electric field or by absorbing a photon with energy above the semiconductor bandgap energy In order to have efficient multiple exciton generation 6 the rate of impact ionization has to be greater than the rate of carrier cooling and other relaxation processes for hot carriers. Fig.1.11 : Multiple excitons generation of quantum dot PV solar cells (Taken from J. Nozik, NERL)
18 In quantum dots the relaxation dynamics of photogenerated carr iers are largely affected by the quantization effects. The reason for that is, when the carriers in the semiconductor are confined by a potential barrier to a re gion of space that is smaller than or comparable to their deBroglie wavelength or to Bohr radius of excitons in the semiconcuctor bulk, the cooling rate of hot carriers is dramatically reduced, a nd the rate of impact ionization could become competitive with the rate of carrier cooling. 9 The threshold photon energy for multiexciton generation is given by w here, m e and m h are electron and hole effective mass es respectively and E g is the bandgap of the quantum dot If wh ich is the case for PbSe the excess energy of the initial exciton will be equally shared between the electron and the hole in generating the second e h pair, and thus, the threshold energy for carrier 3Eg 9 For example 9.5nm QD where band gap is .65eV can harvest maximum solar power according to the AM 1.5 solar spectrum shown in Fig 1.8 Fig.1.1 2 : Comparison of energy levels and threshold energy for MEG in QDs for (a) m e ~ m h
19 1.5.1 Exciton Dissociation The success of electronic devices based on semiconductor quantum dots hinges on the ability to efficiently extract charge carriers from QD and to transport the carriers between electrodes For example, in quantum dot photovoltaic devices bound electron hole pairs ( excitons ) generated by light absorption must be dissociated and the resulting carriers must be collected by electrodes. 10 This mechanism depend s on several properties of the int erface between the QD and the transporting medium. (1) The coupling between QD and transporting medium must be strong enough to separate the electron hole pairs either by charge transfer across the boundary or diffusing one exciton to the next medium prior to non radiative Auger recombination. (2) Intimate contact between the QD and the transporting medium, which lowers the interface traps. (3) Large cross section area between the QD and transporting medium. (4) Built in electric f ield between electrodes to drive the charges to electrodes. Polymer/QD composite s have been recognized to posse both the attributes where charge separation is enhanced at interfaces and the electron and hole transport takes place in two different materials 11 The most important para meters that control exciton dissociation are; ionization potential, electron affinity, band gaps of polymer and QD, density of QDs, distribution of QDs within the polymer, and the nature of contact between polymer and QD. The effect of these parameters on exciton dissociation is discussed below Choi ce of the polymer is crucial for effective coupling between the QD and the polymer. In order for the holes in the QD to transfer into the HOMO level of the polymer, the ionization potential of the polymer must be less than that for the QD. Similarly,
20 trans fer of an electron in the LUMO level of the polymer to the QD is possible only if electron affinity of the QD is higher than that of the polymer. The range of diffusion f or excitons in polymers is 5 15 nm 1 1 O nly the excitons generated close to an interface can be dissociated. For this reason, density of QDs in the polymer plays a crucial role in dissociation of excitons in the polymer. There for e, it is important to maintain an o ptimum density of QD for a given system fo r efficient dissociation process. Uniform distribution of QDs within the polymer is important to avoid formation of isolated QD aggregates surrounded by polymer. Electrons accumulat ed in QD aggregates become trapped and eventually lead to carrier recombination. Such aggregation is readily observed in polymer/QD composites when surfactants are removed prior to mixing 1 2 Therefore, the removal of surfactant has to happen just before t he QDs are mixed with the polymer. Several chemical treatments including hydrazine (N 2 H 4 ) treatment have been developed to remove the surfactant coating from QDs that are Fig.1.13 : The schem atic energy level diagram for PbSe nanodots and P3HT (polymer) showing the c harge transfer of electrons to Pb Se and holes to P3HT.
21 deposited on substrates 13 However, to fabricate QD/host composite structures che mical washing to remove surfactants has to be done prior to embedding them in the host, which inevitably leads to agglomeration and non uniform distribution of the QDs in the host material. Therefore, washing off of the organic surfactant using a hydrazin e solution is not suitable to make a composite of quantum dots and polymers. In addition, toxicity of hydrazine is also a concern. The nature of the contact between the polymer and the quantum do t is important for fast dissociation If QD makes an intimate contact with the polymer such that there is significant overlap of the wave functions between the two materials, tunneling of one charge carrier type of the exciton into the other material is energeticall y favorable 1 4 This mechanism that is dominant in QDs without surfactants is kn own as the Dexter mechanism 1 4 If there is a significant spectral overlap between the emission and absorption spectrum of the two materials, a Coulomb dipole dipole interaction promotes energy transfer bet ween the two materials. This mechanism that is not hindered by the presence of an interfacial surfactant layer is kno wn as the Forster mechanism 1 4 E x citon dissociation through Dexter and Forster mec hanisms are shown in Fig. 1.14 1 5 ,1 6 Acco rding to the d iagram in Fig 1.14 i n order to have an efficient Forster mechanism the band gap of the QD and the polymer ha ve to be in the same range. For example, single exciton dissociation in surfactant coated CdSe nanoparticles (band gap 2.0 2.6 eV) and the polymer MEH PPV ( poly[2 methoxy 5 ethyl hexyloxy) p phenylenevinylene ), band gap ~2.1 eV, have been observed while such a transition is absent in MEH PPV and surfactant coated QDs o f CdS (band gap 2.6 3.1 eV) 14
22 In order to have an efficient Dexter mechanism the surfactant has to be removed from the quantum dot during the interaction. E fficient exciton dissociation between uncoated CdS and MEH PPV has been observed due to the Dexter mechanism. 14 Finally, t hese results have led to the conclusion that for a p olymer/QD composite where the band gaps of the two materials are far apart, presence of surfactants at the interface suppresses the exciton dissociation process Therefore, one of the primary focuses of th is project is to develop a method to deposit surfac tant free quantum dots on a substrate and to produce QDs/Polymer hybrid structure with direct contact between quantum dots and the polymer. Fig.1.1 4 : Possible paths for exciton dissoc iation in polymer/QD composites. (a) Generation of an exciton in polymer followed by elect ron transfer to QD by Dexter me ch a nism. (b) Generation of an exciton in polymer followed by exciton transfer to QD by Forster mechanism. Exciton in QD can dissociate by transferring the hole to the polymer. (c) Generation of an exciton in QD and transfer of the hole to the polymer by Dexter mechanism.
23 CHAPTER 2 NANOCRYSTALLINE GROW TH & PROPERTY CHARAC TERIZATION TECHNIQUES 2.1. Nanoparticle Synthesis Currently there are several different techniques available for nanoparticle growth that include sol gel processing 17 chemical synthesis 18 chemi cal vapour deposition (CVD) 19 plasma or flame spray synthesis 20 laser pyrolysis 21 and atomic or molecular condensation 22 Chemical synthesis permits the manipulation of matter at the molecular level. Because of mixing at the molecular level, good c hemical homogeneity can be achieved. Also, by understand ing the relationship between how matter is assembled on an atomic and molecular level and the resulting macroscopic properties, molecular synthetic chemistry can be tailored to prepare novel compounds 2.1 1 Chemical Synthesis of Nanoparticles C hemical synthesis is a well established industrial process for the generation of colloidal nanoparticles from liquid phase Nanoparticles with diameters in the range of 1 to 15 nm with narrow distributions have been chemically synthesized Typical size
24 variances are about 20%; however, for measurable enhancement of the quantum effect, this must be reduced to less than 5 % 53 Size distribution of semiconductor, metal and metal oxide nanoparticles can be manip ulated by either dopant introduction 2 3 or heat treatment 2 4 Better size and stability control of quantum confined semiconductor nanoparticles can be achieved through the use of inverted micelles 2 5 polymer matrix architecture based on block copolymers 2 6 or polymer blends 2 7 porous glasses 2 8 and ex situ particle capping techniques 2 9 However, these techniques involve elaborate steps of solid liquid separation, washing, and drying to separate the nanopa r ticles for applications. As a result of the la rge surface area to volume ratio of nanoparticles the Van der Waa l attractio n between adjacent particles is considerably high. These forces tend to cause agglomeration of particles as they are fo rmed. Surfactants are used t o prevent agglomeration in a lmost all the chemical processes The surfactants coat the particles and prevent them from aggregating. When the nanoparticle s are close to each other they encounter a repulsive potential stemming from surface bound organic molecules The organic surfa ctant s consist of he ad group s surface via dative bonds, actual covalent bonds or electrostatic attraction. The surfactant extending into the surrounding liquid medium as shown in Fig. 2.1 This tail is important because its polar/nonpolar nature dictates the nanoparticle solubility within the surrounding organic or aqueous media. For many chemically synthesized nanoparticle s, their primary solu bility will be within organic solvents.
25 2 .2 Spray Pyrolysis Spray Pyrolysis technique has been used for the growth of coatings of a variety of materials, including semiconductors and oxides. 30 The main component of a spray pyrolysis system is an atomizer that generates microdroplets of a precursor solution dissolved in a relatively volatile solvent. The droplets, in the form of a fine spray, are carried out of a nozzle onto a heated substrate by a carrier gas that can be inert or reactive. The constituents of the droplets decompose and react on the hot substrate to form the chemical compound. The substrate temperature should be high enough to evaporate the volatile solvents. The spray nozzle is usually scanned continuously during Fig. 2.1: The figure shows the Surfactant coated nanoparticle. Head groups of the surfactant attaches to the particle while tails are extended to the surrounding liquid (Taken from http://www .freepatentsonline.com/ )
26 the growth to coat a large area of the substrate. Even though this technique produces low quality films compared to Chemical Vapor Depo si tion (CVD) or Sputtering, it has been used over the years due to numerous advantages such as simplicity, low cost, non toxic prec ursors, good reproducibility and no need for high vacuum. 2.3 Laser Assisted Spray Process Laser Assisted Spray (LAS) techniques have been used to generate nanaocrystalline powders and coatings. 31 In contrast to the s pray pyrolysis, the droplets inte ract with a continuous wave laser beam as they come out of the nozzle. If the molecules of the precursor have a strong absorption band at the wavelength of laser beam, the molecule is dissociated. This method has been used to form amorphous nanoparticles o f Si/N/C. 32 However, this method is restricted to compounds that have resonance absorption bands at an available laser wavelength. The sizes of the produced droplets depend on the technique used to atomize the solution. The simplest way to generate an aero sol spray is by a pneumatic process. In this method, the pressure drop at the orifice of a nozzle from a high flow rate of gas causes the dispersion of the solution into microdroplets. However, control of the particle size distribution produced by this met hod is very difficult. On the other hand, ultrasonic nebulizers (model 241CST; Sonaer Ultrasonic, Farmingdale, NY) are known to produce a fairly uniform distribution of micrometer size droplets. Generally, the nebulizer is operated at a frequency of 2.4MHz where the precursor solution is converted into a n aerosol of particles in a range of 1 2 m in diameter. These particles have lack sufficient
27 inertia and thus have to be transported by a carrier gas. Because the aerosol generating rate is independent of t he gas flow rate, the transport of the droplets to the substrate can be controlled without affecting the function of the nebulizer. W hen the droplets impinge upon the heated substrate, the precursor is decomposed to form the compound. The subsequent film f ormation and the morphology of the film are dependent on the velocity of the drop, the rate of reaction, and the rate of evaporation of the solvent. At high velocities, the droplet s will flatten on the substrate leading to large particle size. 33 If most of the solvent is evaporated by the laser heating, the salt in the drop condenses into a solid particle and thus forms a smaller grain on the substrate. Therefore, by controlling the concentration of the s olvent, the size of the grain s deposited on the subst rate can be controlled. (a)
28 The Fig.2.2 shows the difference b etween Regular spray pyrolysis v s Laser assisted spray pyrolysis. The average particle size in regular spray pyrolysis is about 250nm. Also irregular particle shape and s ize is visible due to flattening of the drops on t he substrate. In comparison, film s deposited b y laser assisted spray technique show a distribution of well defined particles with an average size of about 80nm. 54 Fig.2.2 : Pa rticle size difference between r egular spray pyrolysis V s Laser assisted spr ay process (a) AFM image of film deposited by regular spray pyrolysis (b) AFM image of the film deposited by LAS technique. (b)
29 2 .4 Structural Characterization of N ano particles The primary tool used for the structural analysis of nanoparticles is the Trans mission Electron Microscope (TEM). Both high resolution ( ~1 nm) and low resolution (~10 nm) TEM have been used t o study the nanoparticles small er than 10nm In a ddition X ray diffraction ( XRD) of coatings also provides evidence of crystallinity in coatin gs of nanoparticles. Unlike bulk materials the diffraction peaks of nanomaterials tend to be much broader Fig. 2.3 is a TEM image which shows the crystalline planes of a nanocrystal l ine particle. 2.4.1. Analysis by Transmission Electron Microscope TEM is u sed to study nanoparticles, which synthesis in this project High resolution TEM enables near atomic resolution and thus allows the observation of atomic planes in single crystalline nanoparticles. In order to obtain the best result in TEM, the thickness of the film needs to be comparable with the mean free path of the electrons. Much thinner films exhibit too little scattering to provide better images, and in thick films multiple scattering events dominate, making the images blurred and difficult to interpret. Therefore, it is very important to maintain a thickness of the sample less than 100nm when preparing samples for TEM studies
30 2.5 Optical Characterization of Nano particles As the semiconductor particle radius approaches the value of exciton Bohr radius there are interesting new effects manifest due to quantum confinement. The most striking pr operty of semiconductor nanopar t i cles is the pr onounced changes in their optical properties indicating a change in the band structure compared to that of the bulk semiconductor material. Fig 2.3 : TEM image of PbSe quantum dot which was produced by sol g el technique. The image shows the spacing d for 200 planes.
31 In bulk semiconductor material, as explained in Section 1.1, Chapter 1 absorption of a photon by a n electron in valance band undergo es a transition to the conduction band This can be observed in optical absorbance spectra for bulk semiconducto r materials. For example Fig 2.4 shows the optical absorption spectra of a chemically deposited Lead Selenide (PbSe) film on a glass substrate. The onset of absorption corresponds to the band gap of PbSe (0.26eV) The re are two interesting changes that can be seen in absorption spectra as the size of the nanoparticle becomes smaller than or comparable to the o rbit al radius of the electron hole pair. First, the energy levels b ecome quantized according to equation 4 g iven in section 1.2. Light induced transition between these energy levels produced a Fig.2.4 : Absorption spectra of chemically deposited lead selenide. (Taken from A. F. The abso rption spectra of solid lead sulfide selenide, telluride
32 series of optical absorption s that can be labeled by the princ ipal quantum numbers of the excitonic levels Secondly the absorption spectrum is shifted due to the confinement of the electron movement within the exciton Bohr radius 34 This can be clearly seen in Fig 1.9 change in optical absorption spectra of a PbS e nanoparticle with change in particle size One can see that lowest energy absorption region, referred to as the absorption edge, is shifting to higher energy with decreasing size of the particle Since the absorption edge is due to the band gap, this m eans that the band gap increases as decreases the size of the particle. Notice also that th e intensity of the absorption de creas es as the particle size decreases due to reduced absorption cross section. 2.5 .1. Photoluminescence Photoluminescence spectr oscopy is a contactless, nondestructive method of probing the electronic structure of the material. 34 Photo excitation of a sample causes electrons within the material to move into permissible excited states. When these electrons return to their equilibri um states, the excess energy is released and may include the emission of light (a radiative process) or may not (a nonradiative process). The energy of the emitted light (photoluminescence) relates to the difference in energy levels between the two electro n states involved in the transition between the excited state and the equilibrium state. The intensity of the emitted light is related to the relative contribution of the radiative process.
33 The photoluminescence technique involves scanning the frequency of the excitation signal, and recording the emission due to recombination within a very narrow spectral range Si nce this process is ultra fast ( of the order of 10 nanoseconds) short pulse (< nanosecond FWHM) laser is need ed to excite the el ectron into the higher levels. Fig. 2.5 shows the photoluminescence spectra of various size silicon nanoparticle samples As shown in f igure 2.5 s ince most common transition in semiconductors is between states in the conduction and valence bands, the ba ndgap of the material can be easily determined by using this technique Since radiative transition also occurs from the defect levels of the semiconductor, t he photoluminescence energy associated with these Fig. 2.5 : Photoluminescenc e spectra from various size Si nanoparticles unit.aist.go.jp/.../lanproc/en/contents.html
34 levels can be used to identify specific defects, and the amount of photoluminescence can be used to determine their concentration. T his study can be further extended to determine the quality of the material by quantifying the amount of radiative recombination. 35 Measurement Principle An Ar gon pulse laser ( beam is focused on the sample which is located in the center of the sample compartment. If the energy of photons coming from the laser source is greater than the energy gap of the semiconductor, the sample emits photons. These are collected and a nalyzed with a dual flat field spectrograph. T his system allows investigation of bandgap energies from 0.75 to 2.4 eV. Fig.2 .6 : Schematic diagram of the Photoluminesence set up
35 CHAPTER 3 EXPERIMENTAL PROCEDURE 3 .1. Synthesis of S urfactant free PbSe Nanoc rys talline F ilms Synthesis of PbSe nanocrystalline films in volve s two steps. (1) Preparation of PbSe nano crystals in a colloidal precursor using chemical synthesis. (2) La ser Assisted Spray deposition of surfactant free PbSe nanocrystalline films using the colloidal precursor prepared in step (1) 3.1.1. Preparation of PbSe Colloidal Precursor Chemicals used for PbSe nanoparticle g r owth: Lead Oxide (99.99%), Oleic Acid (90%), Selenium (99.5%), Trioctylphosphine (90%), 1 Octadecene (90%), Tertrachloroethylene (99%), Chloroform, Methanol, Hexane and Toluene. 5 3 PbSe semiconductor nanocrystals were synthesized in a three neck flask equipped with an e lectro thermal magnetic stirrer, condenser and a thermocouple The procedure included the preparation of two separate solutions and subsequent mixing under controlled conditions. Solution (1) was prepared by mixing 0.64g of selenium powder in
36 5.76g of trioctylphosphine soluti on at 100 150 o C under inert gas (argon), and stirring until the selenium powder was completely dissolved. Solution (2) was prepared by mixing 0.892g of Lead Oxide, 2.825g of Oleic acid in 12.283g of 1 Octadecene (ODE) solution at 150 o C under inert gas. Th e mixture was further heated to 180 o C where it became colorless When t he temperature was at 180 o C, solution (1) was qui ckly injected into solution (2). The solution immediatel y changed color from colorless to dark brown T he reaction mixture was allowe d to cool down to 150 o C for growth of the PbSe semiconductor nano crystals. The n anocrystal s continue to grow until the solution is diluted with Chloroform The time between injecting the solution (2) and adding Chloroform to the mixture determine the siz e of the PbSe nanocrystal in the solution. For our experiment we recovere d solution after 20s, 60s and 12 0s and added equal volume of room temperature chloroform into each solution to stop the reaction. The chloroform solution was ex tr acted twice with equ al volume of methanol Addition of acetone caused the precipitation of the PbSe nanocrystals. The precipitated PbSe nanocrystals Fig.3.1: Setup for synthesis PbSe NC colloidal solutions Taken from C. B. Mur ray et al. 53 )
37 were redispersed in hexane again by excess methanol. Then the solution was centrifuge d at 3500 RPM for about 2 mins to precip itate remaining PbSe nanoparticles The clear solution was r emoved from the bottle and hexane was added to redissolve the PbSe nanocry s tals. These steps were repeated three times to remove all the excess surfactant from the PbSe nanocrystals 53 3 .1.2 La ser Assisted Spray ( LAS ) Deposition of PbSe Quantum D ots The PbSe nanoparticle (10nm) immersed in hexane was used as a precursor in the Laser Assisted Spray deposition. The aerosols of the precursor with an average drop size of 1.5 m were generated by a n ultrasonic nebulizer (Model 241T) which operated at a frequency of 2.4 MHz. Each 6 of PbSe nanoparticles. The aerosol was carried through a conical nozzle into the growth chamber by the SF 6 gas at a flow rate of 0.68slpm. Th e schematic diagram of the laser assist ed spray system is shown in Fig.3.2 The droplets and the SF 6 gas interact with a focused continuous wave Fig.3.2 : Schematic diagram of the laser assisted spray film growth system.
38 (CW) CO 2 laser beam at the nozzle. The CO 2 laser radiation at a wavelength of 10.6 m is strongly abs orbed by the SF 6 gas through vibrational excitation. The laser power of 14W used in the experiment was sufficient to increase the temperature of the gas and subsequently the droplets to about 300 o C. The temperature of the flow just after the laser interac tion was measured by using a thermocouple. The graph in F ig.3.3 shows the variation of temperature with the flow rate. The nitrogen gas was introduced to the growth chamber to maintain the pressure at about 650 Torr. The s ubstrates were placed 4cm away fro m the nozzl e to obtain a uniform coating. 3.2. Sample Preparation for TEM Studies PbSe nanoparticles that are heated by the laser were deposited on carbon coated TEM grids for about a minute to form a PbSe quantum dot mon olayer on the grid Films Fig.3.3 : The gra ph of temperature of the gas flow v s flow rate of SF 6 gas. The line shows the optimum flow rate of the system.
39 were deposited under d ifferent flow rates to obtain different plume temperatures according to the graph in Fig 3.3. 3 .3 Sample Preparation for Conductivity M easurements One of the expected outcomes of the LAS process is the formation of QD f ilms without the surfactant barrier between adjacent particles. The close contact between particles is expected to enhance the conductivity of QD films. Nanoparticle coatings with and without the laser heating were deposited on glass substrates under sim ilar growth conditions. Substrates were pre patterned with gold titanium pads to produce gaps with spacing for conductivity measurements There are several techniques available for fabricating Au/Ti electrodes on glass substrate T he aim is to produce precise patterning of the electrodes with well define d gap sizes T he photolitho graphy lift off technique followed by a thermal evaporation method was used to fabricate the Au/ Ti electrodes. First w ell cleaned glass substrate s w ere spin coated by a photo resist and a pattern to generate two electrodes of 50 m width I nter electrode spacing of 2, 5, 10, and 20 m were etched away by the photolithograph ic lift off technique. Then a 20nm Ti adhesive layer followed by a 200nm Au layer was deposited using thermal evaporation. Finally the remaining photo resist was removed to expose t he T i/Au electrodes as shown in Fig. 3.4
40 3 .2.2. Deposition of PbSe QD layer on E lectrodes The deposition of surfactant free PbSe layer for conductivity me asurement was performed according to the procedure discussed in following section w ith an add itional step of shadow mask ing the substrate A shadow mask provide d a well define d area for deposition over the gap. The shadow mask was set up on the substrate in the following manner D uring the masking thin spacers were placed b etween the substrate a nd shadow masked to protect the electrodes from scratching Fig.3.4: SEM images of Au /Ti electro des fabricated for conductivity measurements. (a) (a) ( b )
41 Prior to the deposition, the patterned glass substrates were cleaned thoroughly with ultras onic agitator to remove all organic surface contamination. The PbSe QD layer was dep os ited on electrodes by using LAS for two conditions (with and without laser) and the conductivity was measured The deposition was carried out for about 30mins to obtain a thickness of about 200nm. The thickness and the morphology of the deposi ted layer w ere measured by a Mechanical Profilometer and the Scanning Electron Microscope (SEM) Fig.3.5: Method of using a Shadow mask for QDs deposition
42 3.4 Synthesis of the P3HT P olymer Coating Typically, polymer coati ngs are formed by spin coating a substrate with a solution containing the polymer. Subsequent h eating evaporates the solvent to form a dense polymer film 36 In order to enable co deposition of QDs and the polymer, the possibility of using the LAS process for the fabrication of the polymer film was investigated. Th e precursor, which contained 5mg/m l Poly(3 hexylthiophene 2,5 diyl) (P3HT)+Tolu ene was prepared by heating the mixture for 2hrs in Nitrogen environment. (The chemical s were purchased from Sigma Al drich). Since the polymer can decompose due to high temperature of the Laser, during the pol ymer deposition the process was run without turning the laser The sub strate was heated below glass transition temperature of the polymer (<150 o C) Prior to the deposition, the system was flushed with Nitrogen for 5 mins to remove all the oxygen from t he system. The pressure of the chamber was set to 650T in order to maintain a pressure difference for adiabatic expansion though the nozzle. The process was run for about 30min to deposit a 400nm layer on t he heated glass substrate. S everal sets of sampl es were obtained by changing the substrate temperature and the thickness of the films. After the deposition, the samples were cooled down in the chamber with constant nitrogen flow and then carefully transferred for optical measurement with minimum exposur e to the atmosphere The crystallinity of the deposited films was investigated by an X ray Diffraction technique at very low angles since t he highest peak for crystalline P3HT was reported at 2 o (degree ) The mor phology of the films was studied by
43 Th e results for these studies are shown and discussed in chapter 4 under section 4.2. 3.5. Co Deposition of QD /Polymer Composite F ilms In the formation o f QD/polymer composite coatings for applications in solar devices the uniform distribution of QDs within the polymer is crucial for optimum performance. The conventional practice is the mixing of the QDs and the polymer in a common solvent and spin coati ng the mixture on a substrate followed by thermal treatment to evaporate the solvent. 37 There are two drawbacks in this method; (1) the QDs will have a surfactant barrier between the particle and the polymer surrounding it, which adversely affect ed charge transport, (2) obtaining a uniform distribution of particles within the polymer matrix is not possible through spin coating and lead s to pockets of agglomerated QDs. The LAS process enables co deposition of QDs and polymer, and thus provides a way to dep osit the composite structures with uniform coverage T he c o deposition system comprised two spray nozzles to transport aerosols of two different precursors one containing the QDs and the other containing the polymer The schematic diagram of the modified system is shown in Fig 3.6 As shown in Fig 3.6 the two nozzles are attached to the two nebulizers and therefore can be controlled independently to spray two different materials In the case of PbSe QDs/Polymer deposition, one nozzle carries the aeros ol of PbSe quantum dots/hexane precursor while the other nozzle carries the aerosol of P3HT polymer precursor. The CW CO 2 laser is focused on to the nozzle that carries the PbSe QDs to
44 burn the surfactant during the deposition. The P3HT polymer droplets a re deposited on the substrate without any laser interaction. Polymer aerosols were not laser heated to prevent solidification prior to arriving at the substrate. During the deposition, both P3HT polymer and surfactant free QDs are combined at the substr ate to form a hybrid structure. The g as flow rate thr ough the nebulizer controls the growth rate of each species at the substrate. This co deposition technique provides clean surface contacts between the QDs and the polymer for the PbSe QDs to transfer ge nerated excitons to the P3HT polymer during the dissociation process The precursor containing the PbSe QDs was made by dispersing PbSe QDs in hexane as discussed in section 3.1.1 and P3HT solution was made by mixing P3HT Fig. 3.6 : Schematic diagram of Laser Assisted Co deposition system
45 powder with Tol uene as discusse d in section 3.4 The substrate was heated below glass transition temperature of the polymer. Prior to the deposition, the system was flushed with Nitrogen for 5mins to remove all the oxygen fro m the system. The flow rate of 0 .68slpm of SF 6 and 0.68slpm of Nitrogen were maintained respectively through PbSe and P3HT nebulizers to carry the aerosol to the nozzles. The pressure of the chamber was set to 650T in order to maintain a pressure difference for adiabatic expansion though the nozzle. A 400 nm film w as deposited within about 30 min. S everal sets of samples were obtained with different film thickness es After the deposition, the samples were cooled down in the chamber with constant nitrogen flow and then carefully transferred for optical measuremen t with minimum oxygen contamination. The crystallinity of the deposited films was investigated by X ray d iffraction technique Initial stage of growth was studied by depositing a thin coating of the composite on TEM grids. 3 .6 Fabrication of PbSe QDs/ P3HT Solar Cell S tructures T ypical structure of a QDs/Polymer Solar cell consist s of an ITO coated glass substrate as the bottom electrode, spin coated transparent layer of polyethyl enedioxythiophene (PEDOT), spin coated layer of QD/Polymer (P3HT) blend and a thermally evaporated a luminum film as the top electrode. 37 In this device structure photons in the visible and infrared region of the solar spectrum are absorbed by the polymer P3HT and the PbSe QDs to generate electron hole pairs PbSe will accept electron and transport them to the Al electro de through the PbSe nanocrystal network while P3HT will accept holes and transport them through PEDOT to the ITO electrode.
46 PEDOT layer in the structure will minimize the hole extraction barr ier at the anode surface. Therefore, the fabrication of a solar cell device starts by solution casting of PEDOT on the ITO glass substrate followed by a 30min annealing at 200 o C in N 2 gas flow to create an 80nm thick hole transport layer. Then the solutio ns of PbSe QDs and P3HT dissolved in Toluene were sprayed on to the heated PEDOT layer at 150 o C by using Laser Assisted Co Deposition under N 2 environment. The thickness of the composite layer was maintained at about 500nm to form the active region of th e photovoltaic device Then the sample was further heated in N 2 gas inside the chamber at about 150 o C for 10mins in order to fully remove the solvent and to crystallize the P3HT in the composite film. Finally, an aluminum electrode was thermally evaporated through a shadow mask onto the film A simplified structural layout of the PbSe QDs P3HT s olar device is shown in Fig. 3.7 Fig. 3.7 : Schematic layout of the PbSe P3HT ba sed hybrid solar cell structure.
47 CHAPTER 4 EXPERIMENTAL RESULTS 4 .1 Characteristics of Self A ssembled PbSe nanoparticles The distribution of PbSe nanoparticles prepared by the solvothermal technique that is outlined in Chapter 3 is shown in the low resolution TEM images (Fig. 4.1 ). Figs. 4.1 (a), 4.1 (b) and 4.1 (c) correspond to particle growth times of 20, 60 and 180 seconds, respectively. The samples were prepared by drop casting the final solution onto carbon coated TEM grids purchased form Electron Microscopy Sciences. Gradual increase of the average particle size from 8nm to 12 nm with increasing reaction time can be observed. (a)
48 These low resolution TEM images also show that the PbSe NCs are pack ed in a simple hexagonal mono layer wit h the long organic ligands filling the inte rstices between NCs When the temperature of the mixture (solution 1 and 2) was at 180 o C the PbSe Fig 4.1: PbSe particles separated from the solution after (a) 20s (b) 60s (c) 12 0s. (b) (c)
49 particles started to nucleate L owering the reaction temperature immediately after injecti on terminated the forma tion of new nuclei. Low temperature permits the nuclei to grow continuously until the growth is stopped by adding chloroform. Therefore, larger part icle sizes can be obtained at 12 0s compared to 20s. The images provide clear evidence of particle size va riation over the course of reaction between lead oxide and selenium trioctylphosphine. These images also show the particle size distribution to be relatively narrow. For particles of average diameter of 10 nm, the standard deviation was about 10 15%. H igh resolution TEM microgra phs of these PbSe nanocrystals show the single crystal nature of the nanoparticles (Fig. 4.2). Fig 4.2: High resolution TEM image of individual PbSe se miconductor nanocrystal. The si n g al crystal structure is clearly apparent.
50 In addition, electron diffraction from t he nanoparticles showed circular patterns corresponding to the crystal planes of PbSe (Fig. 4.3). Since the diffraction image result s from a collection of randomly distributed single crystal nanoparticles, the patterns resemble t hat of a polycrystalline sample. However, the bright spots on the circles correspond to single crystal diffract ion from each particle. X ray diffraction was performed by using a B r u ker AXS D8 X Ray diffracto meter on samples made by three different nanoparticle size distributions. Samples for XRD were prepared by forming multiple layers with multiple drop casting steps. Fig. 4.3: Electron diffractogram of the PbSe nanocrystals
51 The XRD patterns of PbSe nanoparticle coatings prepared by drop casting show h,k,l values were all even or all odd, one can confirm that the PbSe nanocrystals have a perfect rock salt (FCC Fm 3m) structure. Also from X ray diffraction pattern one can determine the "lattice parameter of PbSe nanocrystals It can be calculated by using following equation. Fig. 4.4: X ray diffraction pattern of the samples made by three different particle distributions (a) 8 9nm (b) 9 10nm (c) 10 12nm. 4.1 (c) (b ) ( a )
52 o and h,k.l are corresponding mil ler indices for each angle Table 4.1 shows the calculated values for latt ice parameter for each XRD peak Angle (2 ) h, k, l Lattice parameter (a) A o 25.16 (111) 6.125 29.12 (200) 6.128 41.66 (220) 6.127 49.35 (311) 6.120 51.62 (222) 6.128 60.3 8 (400) 6.127 66.44 (331) 6.128 68.41 (420) 6.128 76.04 (422) 6.127 81.58 (511) 6.127 According to the table the lattice parameter for PbSe nanocrystals w as found to be 6.1265 A o The calculated values of lattice parameters are in excellent a greement with the cubic (rock salt) bulk PbSe lattice parameter of 6.128A o Fig. 4.5: Table of Miller indexes and Lattice parameters of PbSe nanocystals calculated from XRD peaks obtain ed from Fig. 4.4.
53 4.1.1. Optical Characterization of PbSe N anoparticles The visible IR absorption spectra of nanocrystalline PbSe in PCE is shown in Fig. 4.6 for different particle sizes prepared by the method outlined in Section 3.1.1. According to Fig. 4.6 one can clearly see three absorption peaks in each absorption spectrum. The first strong absorption peak corresponds to the first inter band transition (1S h to1S e ), the second absorption pea k corresponds to the second inter band Fig 4 .6 : Near infrared absorption spectra of as prepared PbSe semiconductor nanocrystals immersed in PCE. (In collaboration with Dr. Jiang's lab)
54 transitions 1S h to1P e and 1P h to 1S e and the third broad absorption peak corresponds to the 1P h to 1P e transition. These absorption peaks confirm the quantum confinement of these PbSe semiconductor nanocrystals. As the size of the semiconductor NC decreas es the absorption peaks shifted towards high energy thus confirming the increasingly strong confinement effects. The band gap of the QDs can be computed based on the 1 st absorption peak. Fig. 4. 7 shows the change in bandgap of PbSe nanocrystals with nanocrystalline size that was derived from the absorption data. These values are compared with values computed from equation 4.2 The variation between the calculated and experimentally determined bandgap s (~8 % for particle sizes of 9 + 0.9 nm) may be due to the size distribution. Fig 4.7 : Change in bandgap energy with particle size; (red line) is experimental, (blue line) is calculated.
55 E g is the bandgap energy for bulk PbSe, m e and m h are effective masses of PbSe (for PbSe m e m h ) and L is the p article size (for Quantum dot L x = L y = L z ). Photolumine scence spectroscopy was performed in order to confirm the excitation energy levels of the PbSe nanoparticles in the solution. The Photoluminescence spectra for two particle size distributions with average sizes of 11 nm and 9 nm are presented in Fig. 4.8 and Fig. 4.9 respectively. 4.2 Fig 4.8 : Absorbance and Ph otoluminescence spectra of 11 nm PbSe semiconductor nanocrystals. (In collaboration with Dr. Jiang's lab)
56 Both spectra clearly show the strong near infrared emission in as synthesized PbSe s emiconductor nanocrystals In figure 4.9 the first absorption peak is positioned at 2105nm while the emission (Photolum inescence) peak is positioned at 2130nm. There is a 25nm Stokes shift which is the energy difference betwee n absorbed and emitted photon. In figure 4.9 the first absorption peak is positioned at 1905nm while the emission peak is positioned at 2040nm. Th ere is a 135nm Stokes shift. Fig 4.9 : Absorbance and Photoluminescence spectra of 9 nm PbSe semiconductor nanocrystals. (In coll aboration with Dr. Jiang's lab)
57 4 .1.2 Structure, Morphology, and Optical P roperties of PbSe Nanoparticle Coatings Deposited by LAS P rocess After confirming the crystallin ity and the quantum confinement of the PbSe semiconductor particles, they were d ispersed in excess hexane for use as the precursor for Laser Assisted Spray deposition Fig. 4.10 shows the comparison of absorption spectra between a LAS deposited coating of PbSe on a glass substrate and as synthesized PbSe nanoparticles suspended in he xane Fig. 4.10 : A bsorption spectra of 11 + 1 nm quantum dots in solution and after LASP deposition. There is a 29.1 Blue shift between them due to reduce QD sizes a fter burning the surfactant. In side: Zoom in of LAS deposited Pb Se absorption spectrum. ( In collaboration with Dr. Jiang's lab) (c)
58 The three peaks in the absorption spectra of the PbSe coatings clearly indicate the first inter band transition (HOMO LO MO transition ) second inter band transition (1S h to1P e and 1P h to 1S e ) and the third transition confirming th e quantization of energy levels in the PbSe nanoparticle coating However, the intensities are much smaller than that for the nanoparticles suspended in solution due to the much smaller absorption depth available for the particles on the film. One of the important outcomes of this result is the confirmation that the LAS process, under the growth conditions used, does not significantly alter the optical properties of the PbSe quantum dots. The only noticeable difference in optical properties is a 40.1nm b lue shift in LAS deposited coating, most probably due to removal of surface layer along with the surfactant coating of the particles making the particles smaller A similar blue shift had been reported by Edward when long oleic ligand s were exchanged with short butylamine ligand and also with completely removing the ligand s from PbSe nanocrystals 3 8 For further investigation of this surfactant burning effect three s amples were made on TEM grid s with different plume temperatures and were characterized by Transmission Electron Microscop y (TEM). Fig. 4.11 compares the TEM images of the samples prepared at three different plume temperatures. They are also being compared to a drop casted film. Inter particle separation of 2 5nm is noticeable in drop casted f ilms due to the prese nce of the surfactant (Fig. 4.1 1 (a)). When the laser heated gas temperature is estimated to be 80 100 o C (Fig. 4.11 (b)), some of the particles become connected as their surfactants are burned off. As shown in Fig. 4.11 (c), for an estimated temperature of 150 200 o C, most of the particles are in close contact indicating surfactants have been removed
59 effectively, yet they are not coalescing into larger particles. At higher heating rates, particles appeared to coalesce together. Therefore, the optimum temperature of the laser heated gas was determined to be in the range of 150 200 o C The h igh r esolution TEM images (Fig. 4.12 ) clearly show the intimate contact between QD s while drop casted film shows 2nm gaps between them. It prov ides clear evidence of burning the surfactant during the growth process. The same effect was also observed in vacuum annealed PbSe Q D s, but at very high temperature (~ 523K) due to decomposition of ligand s 39 Fig. 4.11 TEM images of PbSe QDs deposited on carbo n coated TEM grid for 1min with different plume temperatures. (a) drop casted (b) 80 100 o C (c) 150 200 o C (d) 200 230 o C.
60 Fig. 4.12 : High resolution TEM images of (a) drop casted and (b) LAS deposited PbSe NCs. LAS shows the intimate contact between the NCs while drop casted shows a 1 2nm surfactant coating between them. Fig. 4.13 : TEM image of a coating deposited for 2mins at the optimum plume temperature 150 o C 200 o C.
61 Fig. 4.13 shows a TEM image of PbSe nanop articles deposited on a TEM grid under the optimum growth condition. The uniform distribution and the continuous connectivity between the particles are apparent in this micrograp h 4.1.3. Elec trical Conductivity of PbSe QD C oatings D eposited b y the LAS P rocess The conductivity of a film of surfactant cap p ed PbSe coating is expected to have a high resistance as the charge carriers will have to hop from particle to particle across the 2 5 nm insulating surfactant barrier. Therefore, the conductivity of a film consisting of surfactant free PbSe nanoparticles, suc h as the film shown in Fig. 4.13 should be much higher than that for a cap p ed PbSe film. We have measured the I V characteristi cs of films deposited across 2, 5, 10, and 20 m gaps prepared on Ti Au electrodes shown in Fig. 3.4. The t hickness of each film was about 300 nm. The I V characteristics of films deposited by the LAS process with and without the laser heating have been compared Fig 4.14 shows the I V measure ments across a with and without the laser on by LAS technique. The graph shows a more than three orders of magnitude enhancement in current for films deposited with laser compared to that of a film deposited without laser hea ting This is even higher than the measured conductivity of drop casted or spin coated films published in literatures. 39,40,41
62 It is clear that in absence of insulating organic ligands between adjacent particles, the hopping distance for charge carrie rs is significantly decreased resulting in higher conductivity. This similar to what was observed for a collection of PbSe nanoparticles that were washed and annealed to remove the surfactants 39 Fig. 4.14 : I V curve of PbSe Qd films deposited by LAS between Ti/Au electrodes with and without the laser on The inset (a) is the same graph as in (a), with current in pA scale to show the current in the film without the laser heating. (c) Au/Ti Electrodes on glass substrate. (In collaboration with Dr. Jiang's lab)
63 The Fig 4.15 shows the current passed t hrough the films deposited with laser ON compared to laser OFF for different gap sizes. Based on these results, summarized in Fig. 4.15, films of capped PbSe coatings showed very little variation in film resistance with the electrode gap size. That means electron conduction is dominated by the interparticle gap produced by the presence of the surfactant. On the other hand, the surfactant free nanoparticle coatings show a drastic change in resistance with changing gap size. This is expected if the length of percolation paths, which depends on the dimensionality of the path, become s shorter with decreasing gap spacing. Howeve r, Fig. 4.15 : Current Vs Voltage graph for different gap siz es. W/L: Laser is ON and WO/L: Laser OFF. (Provided by Dr. Jiang's lab)
64 above a gap dimension of 10 m, resistance change became negligible. (This will be discus sed in section 5.1.3 ). 4.1.4. Temperature D ependent Conductivity M easurement The temperature dependent conductivity of surfactant free PbSe quantum dots was measured by using a four point probe method. The samples for this measurement were prepared by depositing s urfactant free PbSe films on glass substrates by the LAS process. The film thicknesses were in the range of 500 600 nm. Subsequent to film Fig. 4.16 : Change in conductivi ty with the gap size. The current values were obtained at 5V bias vo lt age
65 growth, four aluminum contacts were deposited on the film by thermal evaporation through shadow mask as shown in Fig 4.17 The four point probe technique was used to measure the film conductivity, w h ere current was applied across the outer contacts while the voltage was measured across the inner contacts (1mm apart) at different temperatures. The sample was mounted in a close d cycle cryostat that enabled temperature variation from room temperature to 40 K. The temperature was reduced from 300K to 40K in 10K int ervals and the variation in voltage was observed Since the dimension of the conductive path was defined, th e voltages were converted to conductivity ( ) using the relation =I The graph in Fig. 4.17 shows the change in conductivity with the temperature. The temperature dependent conductivity of QD films studied by a four point technique indicated three different transport regimes (Fig. 4.18 ). In the region where the tempe rature is above 250 K the ln (T) vs. T plots showed a linear behavior describing a functional dependence of the form ln (T) = (E/k B T) + ln o where E is the activation Fig 4.17 : Sample made for t emperature dependent conductivity measurements
66 energy and o the conductivity of the material at T= 0K In the temperature range of 250K 200K gr aph show s a behavior that fits an equation of the form ln (T) = (T o /T) P + ln o which corresponds to hopping conduction where P can fall within the range of to 1. (P=1/4 is known e Hopping (M VRH) and P =1/2 known as Efros Shklovskii variable range hopping (ES VRH) ) 39 From 200 K to about 110 K the conductance became less dependent on temperature indicating a tunneling mechanism. Fig. 4.18 : The gra log scale) v ersus the inverse of T for three PbSe QDs samples deposited by LASP technique.
67 However below 110 K conductance turn s to me tallic behavior that has not been observed for cap p ed QD assemblies. This phenomenon can be explained by using theoretical models that have been developed based on the quantum mechanical coupling of energy levels of adjacent QDs, facilitated by negligible inter particle separations 4 43,44 The coupling energy between two neighboring QD is given by h where h is the Plan c k constant and is the tunneling rate between two ground state orbitals. In the weak coupling regime where h << k B T the tunneling is restricted to adjacent QDs. However, in the strong coupling regime ( h > k B T ) orbitals may extend over many QDs forming a continuous band that enhances carrier transport. This condition may be satisfied for our films at low temperatures since the inter particle spacing is negligible. 4 .1.5. Photocurrent Measurement 4.1.5 .1 Direct Current (DC) Measurements The PbSe QDs samples for studying photogenerated current were fabricated by depositing a 500nm PbSe quantum dot layer using Laser Assisted Spr ay technique on a n Indium tin oxide (ITO ) coated glass substrate. The ITO coated glass substrate is a commonly used transparent conducting film that has a surface resisti square 1 The substrates were first cleaned by ultrasonication in acetone and methanol and dried in a nitrogen flow. First, PbSe QDs film was deposited on a cleaned ITO coated glass substrate. Then, Silver ( A g ) electrode s (as show n in Fig 4.18 ) were made by t hermal evaporation using a shadow mask. Current generated by application of a bias
68 voltage between 5V to 5V was meas ured by a Keithley SMU 2400 voltage source under an illumination intensity of 80mWm 2 Fig 4.20 sh ows the current response of the surfactant free PbSe QD device under illumination. It is very clear that under illumination the current values at each voltage level are higher than the dark current values. This confirms the photo induced carrier generatio n in the surfactant free PbSe QDs. The different shapes in positive and negative voltage sides were observed due to the different work functions of the electrodes. The positive side forms a Schottky barrier at Ag (4.74eV) and PbSe (4.2eV) junction while negative side forms a near ohmic contact at PbSe (4.82 eV) and ITO (4.6eV ) junction This Schottky barrier formation can be described by a simple diode equation. 4.3 Fig. 4.19: Sample made for photocurrent measurement b y Laser Assisted Spray technique.
69 Where I s is the saturation current, q is the electronic charge, V is the potential drop ac ross the junction, n is the ideality factor, k B temperature. 4.1.5 .2 Photo Generated Current by a Pulse Laser The capacitance of the QD structure that is sandwiched between two metallic electrodes can become an active element of the circuit if the generated current changes with a specific frequency. A circuit has been designed wi th pulsed laser excitation to generate a photocurrent that follows the frequency of the laser pulses. The circuit used for these me asu reme nts i s shown in Fig. 4.21 The photocurrent was measured Fig. 4.20 : The graph of applied voltage vs current density through surfactant free PbSe quantum dots film.
70 under constant bias voltage applied between the device electrodes for various laser pulse intensities. To generate carriers in the quantum dot layer that is sandwiched between the electrodes, laser pulses with a wavelength of 670nm were used. The circuit shown in Fig. 4.21 was used to measure the photo generated current. The g enerated electron s and holes drift to opposite electrodes due to applied electric field across the device structure and flow out of the external circuit as a photocurrent. The generated pho tocurrent was in a power was changed from 0 to 4.2mW by changing the input voltage to th e device f rom 0 to 5V. The l ock in amplifier allows the measurement of only the current generated by the laser pulse, which is the photocurrent. The series resistor in the circuit amplifies and converts the current passing through it to generate the meas ured voltage. Fig. 4.21 : Circuit diagram for detecting photocurrent generated by a pulse d laser
71 The graph in Fig 4.2 2 shows the measured photo generated current at various laser power levels. According to this result the carrier generation rate is proportional to the incident light intensity and did not show any hint of saturation up to the maximum power used. 4 .2. Characteri zation of P3HT Polymer C oatings 4.2.1. Crystallinity Crystallinity of the P3HT polymer was studied by X Ray Diffraction method. Since P3HT has very low angle peaks, the scanning angles were set from 4 o to 30 o with Fig. 4.22 : The graph of photo generated current from the QD device at various laser power levels.
72 0.02 o increments in 3sec/steps. The Fig. 4.23 shows the XRD pattern of the P3HT polymer on the glass substrate deposited by LAS deposition technique that was used for the growth of QD layers, except the aerosol coming out of the no zzle was not heated by the laser. XRD pattern clearly shows the crystalline peaks corresponding to P3HT polymer. The crystallinity of the P3HT films deposited by this method was very similar to the films deposited by spin coating 45 Fig. 4.23 : XRD pattern of P3HT polymer coating deposited by LASP technique on glass substrate heat up to 60 o C
73 4 .2.2. Optical C haracterization The optimum growth temperature was obtained by studying the optical properties of films deposited at different temperatures. Under ideal conditions the absorption by the film should resemble that for a high quality spin coated film. Se veral samples were deposited at substrate temperatures of 200 o C, 100 o C and 60 o C and the absorption s were studied Fig. 4.24 : A bsorption spectrum of P3HT films deposited by Spray Pyrolysis at substrate temperature 200 o C, 100 o C and 60 o C compared to s pin coated fil m ( In collaboration with Dr. Jiang's lab)
74 Fig 4.2 4 shows the absorption spectra of P3HT films deposited by th e spray process at different substrate temperatures. The spectra (excep t at 200 o C) clearly show first transition at 610nm, second transition at 555nm and third transition at 530nm, which confirms the unique absorption property of the P3HT material. But when the substrate temperature was at 200 o C, the material changed its abs orption property due to deco mposition of the polymer at high temperatures. The samples deposited at 100 o C and 60 o C show absorption spectra that are identical to that of spin coated films. But when comparing 100 o C and 60 o C substrate temperatures, the 100 o C shows well defined peaks compared to those of the films deposited at 60 o C. Therefore, one can identify 100 o C as the optimum temperature for P3HT film deposition by the spray p yrolysis technique. 4.2.3. Photo Ge nerated Current by Pulse Laser E xc itation The experime nt in section 4.1.5 .2 was used to determine photo generated current in P3HT polymer structure that is sandwiched between two metallic electrodes The photocurrent was measured with various laser pulse intensities under constant bias vo ltage applied between the device electrodes. The graph i n Fig 4.25 shows the photo generated current value recorded at various laser power levels. The graph shows the linear increment of photo generated current with increasing laser power. This is very goo d evidence of photo generated ca rrie r s in the fabricated polymer device.
75 4.3. Characterization of PbSe QDs/P3HT Hybrid Composite Characterization of PbSe QDs/P3HT hybrid composite was done by using TEM and XRD studies. First, the XRD st udies were performed on thicker samples to determine the crystallinity of the material. Since P3HT has very low angle peaks, the scanning angles were set from 4 o to 60 o with 0.02 o increments in 1sec/step Th e results are shown in Fig. 4.26 Fig. 4.25 : The graph of pho to ge nerated current in the P3HT layer sandwich between ITO and Al electrodes at various laser power levels.
76 XRD pattern clearly shows the crystalline peaks corresponding to P3HT polymer and the PbSe QDs. P3HT shows sharp crystalline peaks due to large c rystalline particles while PbSe shows broad peaks due to n anocrystalline particles Also proportional ized peaks of two mat erials in the films confirm the possibi lity of depositing two materials in Laser Assisted Co Deposition system. The samples for TEM were prepared on a carbon coated TEM grid purchased fro m Electron Microscopy Sciences to investigate the initial formation of PbSe QDs/P3HT hybrid composite. The result i s shown is F ig. 4.27 Fig. 4.26 : XRD pattern o f PbSe QDs/P3HT hybrid composite deposited by laser Assisted Co Deposition.
77 The TEM image shows the initial formation of the hybrid composite on the carbon coated TEM grid. A h igh growth rate of PbSe nanocrystals is visible on the film. Also, the image clea rly shows the uniform distribution of PbSe NC and P3HT in the hybrid composite. 4.3.1. Photo Generated Current Measurements The experiment in section 4.1.5 .2 was repeated for measuring photo generated current in the PbSe QDs/ P3HT polymer hybrid composite th at is sandwiched between two metallic electrodes. As described in that section the photo generated current was Fig. 4.27 : TEM image shows the initial formation of P3HT polymer and the PbSe NC on the film.
78 measured at various laser energy levels. The graph in Fig 4.28 shows the change in photo generated current in the PbSe QDs/ P3HT polymer composit e. The graph in Fig 4.28 shows the normalized photo generated currents in the sampl es of PbSe QDs/P3HT polymer hybrid composite. The graph clear ly shows the photo current generation in the new hybrid composite The photo current of this co mposite ari ses from the efficient charge separation at the polymer NQD/nanorod interface and t he enhanced electron transport in the structure due to the relat ively high intrinsic carrier mobilit y in inorganic semiconductor NQDs. W hen the light is illu minated Fig. 4.28 : The graph of normalized pho to generated current in the PbSe QDs/P3HT layer sandwich between ITO and Al electrodes at various laser power levels.
79 on this PbSe QDs/P3HT composite, both P3HT polymer and the PbSe QDs generate the electron and hole during the light absorption t he hole dissociated to the P3HT polymer due to overlap of wave function between PbSe QD and the P3HT polymer while the electro n hop s b etween the PbSe QDs towards the Al electrodes Since this composite contained surfactant free PbSe QDs the process become s much more efficient compare d to the photo generated current value reported in the literature 45 Therefore, the deposition of PbSe QDs/ P3HT polymer composite by the Laser Assisted co Deposition technique is very promising in producing high ly efficient and low cost hybrid photovoltaic composites 4.3.2. Frequency Dependent Photo Generated Current Measurements Further investigation of the PbSe QDs/P3HT polymer composite w as performed by using frequency dependent photo generated current measurement s. This experiment was designed to study the internal capacitance of the composite layer that consist s of surfactant free PbSe QDs and th e P3HT polymer. The experiment was designed by usin g the setup shown in section 4.1.5 .2 During this experiment, laser power level was kept constant while changing the triggering frequencies of the laser pulse The triggering frequency of the laser was cha nged via lock in amplifier.
80 The graph in Fig 4.29 shows the collected photo generated current values and the values calculated from the model that was developed using the reactive capacitance of the composite The graph shows the increment of photo generated current across the device structure with increasing modulation frequency. This is very interesting since the power incident on the device is independent of the modulation frequency of the laser signal. T his increment of this photo current could b e a result of changing reactive capacitance of the device str ucture In order to investigate the result, we studied the changing reactive capacitance in an equivalent RC (Resistor Capacitor) circuit under same frequency conditions. The voltage and capacito r values for the equivalen t RC circuit were obtained by the equat ion that is derive d below F ig. 4.29 : The graph of calc ulated and ex perimental values for frequency dependent photo generated current through PbSe QDs/P3HT polymer composite.
81 This is similar to y = mx + c; and from the 1/I 2 2 graph one can calculate Sou rce Voltage (V RC ) and the Capacitance (C).
82 From the model that uses an equivalent RC circuit to explain the results t he capacitance and the photo generated voltage of the device can be computed as C = 3.8 10 10 F and V RC = 5.773 10 3 V respectively. T he experiment was extended for different D C bias voltages across the device. The corresponding capacitances were found at each bias voltage. This experiment was performed for investigating the carrier generation, built in vol tage, and depletion region with in this material system. This C V method h as been used by several other groups in calculating interface properties of the semiconductor quantum dots films sandwiched between Alu minum and ITO electr odes 40 The capacitance values obtained in this experiment were used to study the accumulated effect of each PbSe QD/P3HT polymer interfaces The graph of C 1 v s V bias was plotted due to hybrid nature of the interfaces. The f ollowing equation shows the C V relationship in our material system. Fig. 4.30 : Graph of 1/I 2 Vs 1 2
83 w here N a is a carrier density, V bi is built in voltage and q is charge of an electron. The graph in Fig 4.31 shows the C 1 for the device structure at several V bias conditions. Here, the slop e of the graph provides information about ca rrier density while intercept of the graph shows the built in voltage of th e device structure. The graph shows a linear increment of 1/C as the bias voltage changes from positive values to negative values. From the slope of the graph, number of carrier N a and the built in voltage V bi of the device structure can be calculated as 1.0410 9 and 0.833eV, respectively. 4.3 Fig. 4.31: Graph of 1/C Vs bias Voltage applied to the device structure.
84 4.4. Photo Generated Current Measurement of the Solar Cell Structure The photocurrent measurement of the solar device fabricated in section 3.6 was studied by using the experimental set up designed in section 22.214.171.124. The generation of photocurrent at various power levels of the laser was measured. In this experiment, the device was connected in two opposite configurations named toward built in Electric field (the positive side of the DC power supply is conn ected to the Al electrode while negative side is connected to the ITO electrode) and Opposite to built in Electric field (the negative side of the DC power supply is connected to the Al electrode while positive side is connected to the ITO el ectrode). Fig. 4.32 : The graph of photo generated current in the PbSe QDs/P3HT Solar Device at various laser power levels when Forward and Reverse bias condition
85 T he graph in Fig. 4.32 shows the photo generated current values at different laser power levels The graph shows a higher current for forward bias condition and a lower current for reverse bias condition. Also, when comparing the same bias condition with Al /PbSe P3HT/ITO and Al/PbSe P3HT/PEDOT/ITO structures, the one which has PEDOT shows higher current values indicating higher charge extraction. The reason for this higher current can be explained using the work function of ITO layer. The work function of I TO is 4.6eV. When the ITO is coated with PEDOT due to the interface effect, it reduce s the work function of the ITO below 4.6eV Then, the electric field created by the structure containing PEDOT is higher than electric field cr e ated by ITO Al electrodes. At h igher electric fields a higher electron dissociation rate is possible The s low current because of the potential barrier between PbSe QDs and P3HT.
86 CHAPTER 5 DISCUSSION 5.1. Synthesis of PbSe Nanocrystals A noncoordinating solvent method has been used successfully to synthesize PbSe nanocrystals. This method not only elimina tes the extremely toxic organome tallic precursors and most phosphine containing solvents used in other methods but also provides a low cost technique for the growth of high quality semiconductor nanocrystals. Since t he nucleation of PbSe largely depends on the reaction temperature, lowering the temperature can completely s top the nucleation process and thus control the particle size distribution very easily The particles were found to be single crystal as grown and the optical absorption investigations showed them to be in quantum confined states. 5. 2. LAS D eposition of Su rfactant F ree PbSe Nanocrystals The conventional method of synthesizing semiconductor quantum dots always consists of 1 2nm surfactant coating around the quantum dot even after thorough cleaning /washing The surfactant around the quantum dot signifi cantly lowers the conductivity in a coating containing a collection of QDs. The presence of the surfactants
87 is also expected to hinder the dissociation of the excitons generated in photo absorption. Use of the LAS methods enabled the fabrication of PbSe QDs without the surfactant s In this process the droplet s containing PbSe quantum dots and hexane pass through a laser heated region that helps to evaporate the solvent as well as the organic surfactants forming surfactant free QDs. It was shown that th e temperature of the plume can be controlled by controlling the flow rate in the process and the optimum temperature to form closely packed QDs without noticeable recrystallization was found to be between 150 200 o C The deposition of the films at opt imu m temperature showed that the crystallinity and the optical properties of the quantu m dots can be retained. The 40.1 nm absorption peaks shift towards high energy observed in the absorption spectrum of the LAS deposited films corresponds to a reduction in quantum dot size. 5.2.1 Charge Transport in QD F ilms The surfactant around the quantum dots reduce s the electron mobility between the quantum dots The studies have revealed that the low electronic conductivity in semiconductor nanocrystal arrays is b ecause of poor exchange coupling and large concentrations of surface dangling bonds that trap c arriers in mid gap states 4 46 Annealing the nanocrystalline film into a high temperature can increase the film conductan ce due to the removal of t he surfactan t s However, a nnealing has also been shown to depredate the QD I n LAS deposition technique t he removal of the ligand s is completely achieved by the laser interaction, and t herefore, prolonged heating at the substrate is not required. As observed in TEM micrographs t he removal of ligand allows
88 the formation of intimate contact s between adjacent nanocrystals while preserving the distinct excitonic features Also the mobility of the electron is increased due to removal of trap s t ates around the nanocrysta l s and overlap of wavefuctions The current voltage (I V) characteristics of LAS deposited and drop casted films were conducted by depositing films within a 2 m gap between two Ti gold electrodes deposited on a glass substrate. The measurements were ma de at room temperature for films with similar thicknesses. The films do not show a threshold voltage for current flow resulting from Coulomb blockade that has been reported for monolayers of self assembled metal and semiconductor nanoparticle arrays 4,47 If present, such a threshold in our measurements may have been masked by the parasitic conduction in gl as s at room temperature. Fig.4.14 shows the measured current for an applied voltage across the 2 m gap. The current produced by the LAS deposited fil m is more than three orders of magnitude larger than that measured for the drop casted film, which indicates a low resistant carrier percolation path across the electrodes. Several previous experiments on mono and multilayered nanoparticle arrays have sho wn the IV characteristics to follow a functional dependence of the form where is a scaling exponent 39,47,48 The value of the exponent is related to the number of percolation paths available for current flow, and thus depends o n the dimensionality of the junction 49 For 1D arrays where only one path could carry the majority of the current, For 2D arrays of lithographically formed tunnel junctions, the experimentally observed dependence is For a junction of dimensionality slightly higher than 2D the observe d dependency is This mode of transpo rt is in agreement with the 2.19 exponent we observed for the films grown by the LAS process.
89 5.2 2 Photoconduc tivity Generation of photocurrent in both QD and QD/Polymer composite films have been demonstrated in this project. W hen the bias voltage exceeds more than the reverse voltage very low current can be observed due to net movement of electron in the cond uction band towards the potential drop W hen the li ght is illuminated on the sample large current can be observed due to increase of the number of free electron in the conduction To cause this electron excitation to the conduct ion band, the light that st rike s the semiconductor must need a minimum energy. Therefore, illumination of right wavelength is important to observe a large current enhancement compared to the dark condition. In our experiment, the illuminated light shows a considerable improvement o ver the dark current due to high mobility of electrons and holes and less trap states in surfactant free PbSe quantum dots. Since the number of new electron hole pairs generated during the light illumination is proportional to the number of incident photon s, higher intensity will generate higher current. 5.2 3. Temperature Dependent C onductivity According to the Fig 4.18 the electron mobility of the nanocrystals was changed with c hanging temperature. Four distinct regions in the graph were observed at different temperature regions indicating four different transport mechanisms of nanocrystals. When lowering the temperature of the sample each mechanism become s more prominent within that temperature range These mechanisms are independent of each other a nd
90 largely depend on the exchange coupling en ergy between the nanocrystals and the Coulomb energy of the nanocrystal array When temperatu re was reduced from room temperature to 250K the graph show s a linear drop of ln accordance with Arrhenius type beha vior i.e. ln = (E a /k B T) + ln o ; where E a is the activation energy. A ccording to the graph activation energy for surfactant free PbSe is E a = 52.2meV This thermally activated conductance i s due to the effect of electron electron repulsion (Coulomb bloc kage) 4,39 When temperature was reduced further from 250K to 15 0K the graph show s a drop of ln accordance with hopping mechanism ( i.e. ln = ( T o /T) P + ln o ) According to the graph, there are two hopping mechanism s which can be identified From 250K to 24 0K the ln reduced in accordance with P=0.95 while from 240K to 20 0K the ln reduced accordance with P=0.5 When P=0.5 the hopping regime can be identified as Efros Shklovskii variable range hopping (ES VRH) When temperature was reduced from 20 0K to 11 0K the drop becomes almost zero due to diminish ing of hopping mechanism In this region, the tunneling mechanism becomes prominent over other com peting mechanism s Since tunneling contributes weakly to the conductance, it is been neglected in the region f rom 300K to 20 0K T here is a slight increment of electron conduction in the temperature below 11 0K. This c ould be due to semiconductor to metal transition in surfactant free quantum dots which occurs as a result of entering into strong coupling regime at the low temperatures
91 5.3 Fabrication of PbSe QDs/P3HT Hybrid Structure by LA Co Deposition The main advantage of the Laser Assisted C o D eposition technique is co deposition of the surfactant free quantum dots and the polymer T hi s pr ovides a way to deposit the hybrid composite structures with uniform coverage The XRD and TEM resul ts in section 4.3 confirm the formation of PbSe QDs/ P3HT polymer hybrid composite by laser a ssisted co d eposition technique. The photo generated current me asurements of the deposited PbSe QDs/P3HT composites confirm the generation and dissociation of excitons during the light illumination. I n this PbSe QDs/P3HT composite, as confirmed in section 4. 1 and 4.2 both PbSe QDs and the P3HT polymer absorb visible Fig. 5.1 : The versus the inverse of T for PbSe QDs samples deposited by LASP technique comparison with the calculated curves.
92 and NIR spectrum and generat e electrons and holes which eventually separate at each PbSe QD P3HT interface. Th e mechanism taking place in this heterostructure can be explained by using a simplified band diagram as shown in Fig 5.2 The il lustrated band gap energy of PbSe NQDs and P3HT was estimated from the lowest excitonic absorption peak in Fig. 4.6 and Fig. 4.24 respectively It has been assumed that the conduction and valence bands in the nanoparticle shift by the same energies relati ve to the bulk material due to the nearly equal effective masses for holes and electrons in PbSe. 5 0 Since the lowes t unoccupied molecular orbital (LUMO) and the high est occupied molecular orbital ( HOMO ) of P3HT lie above the conduction band ( CB ) and vale n c e band (VB) edges of PbSe NQDs (9.5nm) respectively, the PbSe NQD/P3HT composite forms a type II heterojunction at the nanocrystal/polymer interface. Charge separation is therefore favored between the high electron affinity Fig. 5.2 : Energy level diagram of the PbSe QDs /P3HT based hybrid composite
93 inorganic semiconductors and th e low ionization potential polymer. 51 The ionization potential energy of PEDOT is close to that of P3HT to minimize the hole extracti on barrier at the anode surface by work function of ITO at the interface.
94 REFERE NCES 1. H. Gleiter Progress in Materials Science 33:4 (1990) 2. E. Anisimovasa, A. Matulisa and F.M. Peetersb, Proceedings of the 12th International Symposium UFPS, Vol. 107, page 187 192 (2004). 3. Yu. P. Rakovich, J. F. Donegan, S. A. Filonovich, M J. M. Gomes, D. V. Talapin, A. L. Rogach, A. Eychmller, Physica E 17, 99 100 .2003 4. G. A. Prinz and J. H. Hathaway, Physics Today 48 (4), 24 (1995) ; P. Grnberg Physics Today 54 (5), 31 (2001); G. A. Prinz, Science 282 1660 (1998). 5. D. Vanmaekelberg h and P. Liljeroth Chem. Soc. Rev. 34. Page 299 312 (2005). 6. A. S. Edelstein and R. C. Cammarata, Nanomaterilas: Synthesis, Properties and Applications 7. V. I. Klimov. J. Phys. Chem. B 104, 6112 6123 (2000 ). 8. N. Peyghambarian S. W. Koch, and A. Mysrowi cz Introduction to Semiconductor Optics Englewood Cliffs: Prentice Hall, ( 1993 ) 9. A. Marti, G. L Araujo, Sol Energy Mater. Sol. Cells, Vol. 43, 203 222 (1996) 10. R. D. Schaller M. A. Petruska, and V. I. Klimov, Applied Phys. Lett. 87, 253102 (2005). 11. C .A. Leatherdale, C.R. Kagan, N.Y. Morgan, S.A. Empedocles,M.A. Kastner, and M.G. Bawendi, Phys. Rev. B 62, 2669 (2000). 12. A. Franceschetti and A. Zunger Phys. Rev. Vol 63, 153304 (2000). 13. M.Tamborra, M Striccoli, R Comparelli, M L Curri, A Petrell a and A Agostiano, Nanotechnology Vol. 15, S240 S244 (2004)
95 14. S. J. Kim, W. J. Kim, Y. Sahoo, A. N. Cartwright and P. N. Prasad, App. Phys. Lett, 92, 031107 (2008). 15. J.H. Warner, A. R. Watt, E. Thomsen, N. Heckenberg, P. Meredith, and H. Rubinsztein Dun lop, J. Phys. Chem. B., Vol 109, Page 9001 9005 (2005). 16. N. Greenham, X. Peng, and A. Alivisatos Phys. Rev. B, 54, 17628 (1996). 17. F. Chang, S. Musikin, L. Bakueva, L. Levina, M. Hines, P. Cry, E. Sargent Appl. Phys. Lett., Vol 84, Page 4295 (2004). 18. A. S oloviev, D. Ivanov, R. Tufeu, A. V. Kanaev, J. Materials Science Letters, Vol. 20, Number 10, Page 905 906 (2001). 19. Z.C. Xu, C. M.Shen, C. W. Xiao, T.Z. Yang, H.R. Zhang, J.Q. Li, H.L. Li and H.J. Gao, Nanotechnology 18 115608, Page 5 (2007). 20. E. V. Shlyak hova, N. F. Yudanov, A. V. Okotrub, Yu. V. Shubin, L. I. Yudanova and L. G. Bulusheva, Journal of Inorganic Materials, Vol. 44, No. 3, Page 213 218 (2008) 21. T. Tani, L. Mdler and S. E. Pratsinis Journal of Nanoparticle Research, Vol. 4, No. 4, Page 337 3 43 (2002). 22. H. Maskrot, N. H. Boime, Y. Leconte, K. Jursikova, C. Reynaud and J. Vicens Journal of Nanoparticle Research, Vol. 8 No. 3 4, Page 351 360 (2006). 23. M. Eslamin, M. Ahamed and N. Ashgiz, Nanotecnology, Vol 17, Page 1674 1685 (2006). 24. T. Kyprian idou Leodidou W. Caseri, and V. Suter. J. Phys. Chem 98:8992 (1994). 25. C.C. Wang, Z. Zhang, and J.Y. Ying. Nanostructured Mater 9:583 (1997). 26. T. Gacoin, L. Malier, and J.P. Boilot. Chem. Mater 9:1502 (1997). 27. V. Sankaran J. Yue, R.E. Cahen, R.R. Schro ck, and R.J. Silbey. Chem. Mater. 5:1133 (1993). 28. Y. Yuan J. Fendler, and I. Cabasso. Chem. Mater 4:312 (1992).
96 29. B. L. Justus, R.J. Tonucci, and A.D. Berry. Appl. Phys. Lett 61:3151 (1992). 30. S.A. Majetich, and A.C. Canter. J. Phys. Chem 97:8727 (1993). 31. M. Liu, M.L. Zhou, L. H. Zhai, D.M. Liu, X. Gao and W. Liu., Physica C 386, 366 (2003). 32. N. Herlin, X. Armand, E. Musset, H. Martinengo, M. Luce and M. Cauchetier., J. Eur. Ceram. Soc. 16, 1063 (1996). 33. A. Muller, N. Herlin Boime, F. Tenegal, X. Armand, F. Berger, A. M. Flank, R. Dez, K. Muller, J.Bill and F. Aldinger., J.Eur Ceram. Soc. 23, 37 (2003). 34. P. S. Patil ., Mater. Chem. Phys 59, 185 (1999). 35. C. P. Poole and F. J. Owens, Introduction to Nanotechnology, Page 194 19, Page 90 91 Page 232 36. S. Osta penko, I. Tarasov, J. P. Kaljas, C. Haessler and E. U. Reisner, Semicond. Sci. Technol. Vol. 15, Page 840 848 (2000). 37. E. Bundgaard, S. E. Shaheen, F. C. Kerbs, D. S. Ginley,. Solar Energy Materials & Solar Cells, Vol 91, Page 1631 1637 (2007). 38. D. C. Jian T. Zhu, G. Paradee, S. Ashok, Appl. Phys. Lett. Vol 88, 183111 (2006). 39. E. Sargent, Method of making quantum dots. US Patent Application: 20070174939, (2007) 40. H. E. Remero and M. Drndic, Appl. Rev. Lett. Vol 95, 156801 (2005). 41. J. P. Clifford, K. W. Jo hnston, L. Levina, and E. H. Sargent, Appl. Phys. Lett. Vol 91, 253117 (2007). 42. S. Gunes, K. P. Fritz, H. Neugebauer, N. S. Sariciftci, S. Kumar, and G. D. Scholes, Solar Energy Materials & Solar Cells, Vol 91, Page 420 423 (2007).
97 43. V. A. Tkachenko, Z. D. Kvon ,O. A. Tkachenko, D. G. Baksheev, O. Estibals,and J. C. Portal, Physica E: Low dimensional Systems and Nanostructures, Volume 21, Issues 2 4, Pages 469 473 (2004). 44. G. Markovich, C. P. Collier, H. C. Henriches, F. Remacle, R. D. Levine and J. R. Heath Acc. Chem. Res. 32, 415 (1999). 45. F. Remacle, J. Phys. Chem. A 104, 4739 (2000). 46. D. Cui, J. Xu, T. Zhu, G. Paradee, and S. Ashok, Appl. Phys. Lett. Vol. 88, page 183111 (2006). 47. N. Y. Morgan et al., Phys. Rev. B 66, 075339 (2002). 48. R. Parthasarathy, X. M. Lin, and H. M. Jaeger, Phys. Rev. Letts. 87, 186807 (2001). 49. B. L. Wehernberg, D. Y. Jaisen Ma and P. Guyot Sionnest, J. Phys. Chem. B 109, 20192 (2005). 50. C. T. Black, C. B. Murray, R. L. Sandstrom, and S. Sun, Science 290, 1131 (2000). 51. P. Capper, Narrow Gap II VI Compounds for Optoelectronic and Electromagnetic Applic ations, 1st ed., edited Springer, Berlin, (1997). 52. J. Nozik, Physica E (Amsterdam) 14, 115 (2002). 53. C. B. Murray, S. Sun, W. Gaschler, H. Doyle, T. A. Betley and C. R. Kagan, Organic electr onics, Vol. 45, No. 1 (2001). 54. S.Witanchchi, G. Dedigamuwa and P. Mukherjee, Journal of Material Research, vol 22, page 649 654 ( 2007 )
ABOUT THE AUTHOR Gayan Dedigamuwa was born and raised in Sri Lanka. He attended the University of Kelaniya as an undergraduate where he specialized in Physics. He joined the University of South Florida Department of Physics in the spring of 2003 an d started working as a Research Assistant in Laboratory for Advanced Materials Sciences and Technology (LAMSAT) under Prof. Sarath Witanachchi. In 2005, he finished his Masters of Science degree and continued toward his PhD. He has published several scient ific papers on thin film fabrication and surfactant free QD film fabrication by using Laser Assisted Spray p rocess developed in his laboratory. In summer 2008 he was awarded the Fred L. and Helen M. Tharp Physics Graduate Scholarship In April 2009, he jo ined the United Solar Ovonic located in Auburn Hills, MI as a Process Engineer/Physicist He currently resides in Rochester Hills, MI with his wife, Nilusha and lovely little daughter, Luhansa.