STUDY OF THE IMPACT OF MICROSTRUCTURES AND I N TERFACE ENERGETICS IN PEROVSKITE AND ORGANIC SOLAR CELLS By Chuanpeng Jiang A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Chemistry - Doctor of Philosophy 2018 A BSTRACT STUDY OF THE IMPACT OF MICROSTRUCTURES AND I N TERFACE ENERGETICS IN PEROVSKITE AND ORGANIC SOLAR CELLS By Chuanpeng Jiang To deal with the increasing demand on energy and the concerns about fossil fuels , s olar ener gy has become one of the most promisi ng alternative energy . P hotovoltaic technology has been developed to harness es the solar energy . Different types of solar cells depending on the materials and structures of the devices have been developed, such as cryst alline Si cells , dye sensitized solar cells , perovskite solar cells and organic photovoltaics. Solar cells with high efficienc y , low cost , and excellent stabilit y are desirable for the market. The first part of this study focuses on the organic - inorganic hybrid perovskite photovoltaics . Solar cells consisting of polycrystalline perovskite thin film s have demonstrate d a rapid increase of power conversion efficiency ( PCE ) in the past few years . To further boost the device performance, it is crucial to unders tand how the microstructures, such as the film texture, grains and grain boundaries, impact the electrical properties of the perovskite thin film . The ramp - annealing treatment is ad a pted to tailor t he texture of perovskite film s , where a strong correlat ion between the device performance and the thin film texture is revealed by X - ray diffraction (XRD) and J - V characteristics . Electrochemical impedance spectroscopy (EIS) further suggests that the enhanced texture structure not on ly suppress es recombination at the contact but also improve s the carrier diffusion length , which ultimately contribute s to better device performance . The other important feature of the polycrystalline thin film is grain s and grain boundar ies . To investigate the influence of these micro structures on device perf ormance, photo - conducting atomic force microscopy (pc - AFM) and K elvin probe force microscopy (KPFM) measurements , which provide the nano - scale resolution, are performed on perovskite thin films with columnar structure s . Three discr ete photocurrent levels are identified among perovskite grains, likely corresponding to the crystal orientation of each grain identified by electron backscattering diffraction (EBSD) . Local J - V curves measured on these grains further suggest an anti - correl ation behavior between short - circuit current ( J sc ) and open - circuit voltage ( V oc ). These results suggest the orientation - dependent carrier mobilit y in perovskite thin films. In addition, the photoresponse of perovskite films displays a pronounced heterogen eity across grain boundaries, with low - angle boundaries exhibiting even better performance than the adjacent grain interiors . KPFM further reveals the downward band bending at grain boundaries which draws electrons and repels holes. Thus, the low - angle gra in boundaries facilitate the electron transport and suppress recombination. The second part of this study focuses on the interface engineering of organic photovoltaics (OPV s ). Organic photovoltaics have attracted a significant amount of attention as they offer potential benefits of low cost and mechanical flexibility. It is known that in OPV devices the energy level alignment at the interfaces between metal electrodes and the photoactive layer is critical in determining the charge collection efficiency. He re , zinc oxide ( ZnO ) buffer layer is introduced between the bulk heterojunction (BHJ) organic layer and the cathode material. By varying the processing condition of ZnO layer, the energy level alignment at the contact is tuned and thus the device performan ce. The interfacial energetics is further investigated by KPFM. Schottky barriers with varied widths are identified at ITO/ZnO interface s . With electron s tunneling through the narrow Schottky barrier, the charge collection efficiency at the cathode is impr oved . iv ACKNOWLEDGEMENTS The work detailed in this dissertation was partially supported by Michigan State University Strategic Partnership Grant. In the past 6 years, I have received the help from many people, not only with this work but also in my perso would like to take this chance to acknowledge them for their support. First, I would like to thank Professor Pengpeng Zhang. It has been an honor to work with her. She provides me the opportunity to explore the world of the nanoscale. I really appreciate that she spent a large amount of time in training me on the operation of instruments and instructing me on how to form logical thinking and giving presentations. I really enjoyed every discussion we had which inspired me in my scientific career. I had the opportunity to work with many group mates who help me a lot with my research. I would like to thank Dr. Sean Wagner. He is always willing to help me. I still remembered those la te nights we worked together. It is a great experience working with him. Dr. Jiebing Sun taught me how to make semiconductor devices and helped me with the first publication. I really appreciate that. I have to mention that it is a pleasure to work with P rof. Xianglin Ke, Dr. Tao Zou, and Dr. Mengze Zhu. They offered me different perspectives on my research and provided constructive suggestions. I really enjoyed the time when I sat next to them. We had so many interesting conversations. Please remember to return the tools you borrowed from our lab. Thank you. v I would also like to thank Prof. Thomas Hamann and Dr. Yuling Xie. They offered great help with the impedance measurement and input on the manuscript. Prof. Hamann exposed me to the EIS measurement wh ich has been quite useful throughout my studies. We also had many fruitful conversations about the impedance measurement as well as the design of experiment. Dr. Xie taught me how to use the instrument and helped me reserve the instrument, which I really a ppreciate. In my second publication, we have reported the EBSD results on perovskite thin films for the first time. I have to mention the help from Quan Zhou. He spent lots of effort in the experiment and helped with the data processing. Though we had fai led several times at the beginning, he never gave up and was always willing to help. I really appreciate that. It is an honor to collaborate with Dr. Richard Lunt and his group members. Dr. Lunt is so passionate, creative and inspiring. He also offered gr eat help in making the devices, especially when I was struggling with making perovskite solar cells. I would like to thank his group members: Dr. Dianyi Liu, Chenchen Yang and Dr. Peggy Young as well. Thank you for helping me with using the instruments in the lab and developing new instrument setup. Dr. Liu and Chenchen are also good friends to me and I would like to thank you for your help in my personal life. It would be a regret to me if I do not acknowledge Dr. Reza Loloee and Dr. Baokang Bi. Dr. Loloe e is the first one I would think of when the instrument in the basement went wrong. He is more like a magician who can fix everything. I also enjoyed the time I spent in the clean room with Dr. Bi. He taught me how to use many experimental instruments and skills which are of great help in my scientific career. vi In the last year of my Ph.D. study, I had been doing an internship at BASF - Battery Materials Ovonic. I would like to thank Dr. Kwo Young for providing me this opportunity. He is the smartest person I have ever met but he is also humble and willing to help others. Thank you for all the help at BASF. Also, I would like to acknowledge the coolest lady in the world, Rose Bertolini. I really cannot ask more from a colleague. I had so much fun working with her. Thank you for teaching me the real American culture and taking me in the break - the - week lunch group. I wish you the best! I would also like to thank William Mays, Sherry Hu, Vera Chang, Shuli Yan, Meng Xu, David Huang, Dr. Benjamin Reichman and Peifen g Li. Thank you for all the help you gave. I really appreciate that. I would also like to thank my awesome friends: Dr. Yuan Gao, Dr. Xiaoran Zhang, Yanlian Xin, Shangguan Yang, Lin Song, Yang Zhou, Zhuoqin Yu, Zhao Peng, Dr. Yuling Xie, Ruiqiong Guo, Yuj ue Wang, Zhihui Liu, Lulu Shen and Nan Du. You made my life here wonderful. I cherish all the time we had spent together. It is sad to see some of you leave here but I am also very happy for you guys. I wish everyone the best. I would like to acknowledge my friends from LCCChurch: Pastor Jiarong Peng, Xingran Wang, Yijun Zhao and Lihua Yang. When I first came to the US, they helped me settle down and took me to church. Especially, Uncle Zhao and Aunt Yang offered me so much help and I felt like they treate d me as their own son. And in my most difficult time, they always took care of me and prayed for me. Thank you! Last, I would like to thank my parents. Though they do not understand English, I would like to write down how much I love them. Thank you, mom, for praying for me every day in the early morning when other people are still in their dreams. Thank you, Dad, for working so hard to support me. I am sorry that I cannot always be around you guys. I really appreciate your vii understanding. I know that no mat ter how many thank you I say, it would not be enough. But please allow me wishing you best. Thank you, everyone. viii TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ........................ x LIST OF FIGURES ................................ ................................ ................................ ..................... xi Chapter 1 Introduction ................................ ................................ ................................ ................. 1 1.1 Motivation ................................ ................................ ................................ ....................... 1 1.2 Outline of this work ................................ ................................ ................................ ......... 6 Chapter 2 Solar cell physics ................................ ................................ ................................ ......... 8 2.1 Introduction to semiconductors ................................ ................................ ....................... 8 2.1.1 Basics of semiconductors ................................ ................................ ........................ 8 2.1.2 Intrinsic and extrinsic semiconductors ................................ ................................ .. 10 2.1.3 Equilibrium distribution of carriers ................................ ................................ ....... 12 2.1.4 Junctions ................................ ................................ ................................ ................ 14 2.2 Planar structured solar cells ................................ ................................ ........................... 21 2.2.1 Light absorption and exciton dissociation ................................ ............................. 21 2.2.2 Charge transport ................................ ................................ ................................ .... 23 2.2 .3 Charge Collection ................................ ................................ ................................ .. 25 2.2.4 Recombination, carrier lifetime and diffusion length ................................ ............ 27 2.3 J - V characteristics of photovoltaic de vices ................................ ................................ ... 30 2.3.1 Short - circuit current ( J sc ) ................................ ................................ ....................... 32 2.3.2 Open - circuit voltage ( V oc ) ................................ ................................ ...................... 33 2.3.3 Fill factor, shunt resistance and series resistance ................................ .................. 34 Chapter 3 Perovskite solar cells and organic photovoltaics ................................ .................... 36 3.1 Organic - inorganic hybrid perovskite solar cells ................................ ........................... 36 3.1.1 Basics of perovskite materials ................................ ................................ ............... 36 3.1.2 Structure of solid - state perovskite solar cells ................................ ........................ 39 3.1.3 Fabrication of polycrystalline perovskite thin films ................................ .............. 40 3.1.4 Light absorption and c arrier generation in perovskite solar cells .......................... 45 3.1.5 Carrier transport and collection in perovskite solar cells ................................ ...... 46 3.1.6 Non - rad iative recombination in perovskite solar cells ................................ .......... 47 3.1.7 Remaining challenges in perovskite solar cells ................................ ..................... 50 3.2 Organic photovolta ics and the application of ZnO as a buffer layer ............................. 54 Chapter 4 Device fabrication and experimental methods ................................ ....................... 59 4.1 Device fabrica tion ................................ ................................ ................................ ......... 59 4.1.1 Fabrication of perovskite solar cells ................................ ................................ ...... 59 4.1.2 Fabrication of organic photovoltaics ................................ ................................ ..... 64 4.2 Experimental conditions of characterizing tools ................................ ........................... 64 4.2.1 J - V and EQE measurements ................................ ................................ .................. 64 4.2. 2 SEM and EBSD measurements ................................ ................................ ............. 64 ix 4.2.3 X - ray diffraction measurement ................................ ................................ .............. 65 4.2.4 AFM measurement ................................ ................................ ................................ 66 4 .2.5 EIS measurement ................................ ................................ ................................ ... 66 4.2.6 UV - vis spectroscopy ................................ ................................ ............................. 67 4.3 Working principles of experimental methods ................................ ............................... 67 4.3.1 Atomic force microscopy ................................ ................................ ...................... 67 4.3.2 Electrochemical impedance spectroscopy ................................ ............................. 79 4.3.3 El ectron backscattering diffraction measurement ................................ ................. 84 Chapter 5 Elucidating the impact of thin film texture on charge transport and collection in perovskite solar cells ................................ ................................ ............................. 90 5.1 Ramp annealing treatment and crystalline structure of perovskite thin films ............... 90 5.2 Correlation between device performance and the thin film texture .............................. 94 5.3 Impact of thin film texture on charge collection and transport ................................ ... 100 5.4 Conclusion ................................ ................................ ................................ ................... 109 Chapter 6 Crystalline orientation dependent photoresponse and heterogeneous behaviors of grain boundaries in perovskite solar cells ................................ ........................ 110 6.1 Photocurrent mapping o f perovskite thin films with columnar structures .................. 110 6.2 Identification of grain orientations in perovskite thin films by EBSD ....................... 117 6.3 Photoresponse of low - angle and high angle grain boundaries ................................ .... 123 6.4 Electronic structure of grain boundaries investigated by KPFM ................................ 126 6.5 Conclusion ................................ ................................ ................................ ................... 129 Chapter 7 High - performance inverted solar cells with a controlled ZnO buffer layer ..... 131 7.1 Impac t of ZnO pre paration temperature on device performance of organic photovoltaics ................................ ................................ ................................ ..................... 131 7.2 Origin of the temperature dependence ................................ ................................ ........ 134 7.3 Tuning of the ZnO work function and interface energetics at the cathode ................. 138 7.4 Conclusion ................................ ................................ ................................ ................... 141 Chapter 8 Conclusions and f uture work ................................ ................................ ................. 142 8.1 Summary of results ................................ ................................ ................................ ...... 142 8.2 Future work ................................ ................................ ................................ ................. 145 A PPENDIX ................................ ................................ ................................ ................................ 147 BIBLIOGRAPHY ................................ ................................ ................................ ..................... 153 x LIST OF TABLES Table 4.1: Precursor concentration and film thickness ................................ ................................ . 62 Table 5.1: Characteristic parameters of perovskite solar cells processed under different condictions . ................................ ................................ ................................ ................................ ....................... 96 Table 5.2: Characteristic parameters of perovskite solar cells with different t hickness ............. 105 Table 6.1: Parameters of fitted point J - V curves ................................ ................................ ......... 115 Table 6.2: Miller indices derived from Euler angles ................................ ................................ .. 122 Table 7.1: Summary of the average solar cell performance parameters ................................ ..... 133 Table 7.2: Work functions of ITO, ITO/ZnO, and ITO/ZnO/PCB M measured by KPFM ............. 138 xi LIST OF FIGURES Figure 2.1: Band formation of semiconductors ................................ ................................ .............. 9 Figure 2.2: Schematic of semiconductor doping ................................ ................................ .......... 11 Figure 2.3: Temperature dependence of carrier concentrations in n - type semiconductors. ......... 14 F igure 2.4: Formation of Schottky barriers ................................ ................................ ................... 16 Figure 2.5: Rectifying characteristics of the metal/semiconductor junction. ............................... 17 Figu re 2.6: Formation of Ohmic contacts at the metal/semiconductor junction ........................... 18 Figure 2.7: Schematics of the formation of junctions at the semiconducting organics and metal interface ................................ ................................ ................................ ................................ ......... 19 Figure 2.8: Illustration of the band alignment at (a) p - n and (b) p - i - n junctions. ......................... 20 Figure 2.9: Planar structured solar cells ................................ ................................ ........................ 21 Figure 2.10: Schematic of charge transport driven by the built - in electric - field .......................... 26 Figure 2.11: Schematic of charge collection at the interface ................................ ........................ 27 Figure 2.12: Schematics of the loss mechanisms ................................ ................................ .......... 28 Figure 2.13: Illustration of J - V curves and the characteristic p arameters of photovoltaic devices under dark and illumination conditions. ................................ ................................ ....................... 32 Figure 2.14: Equivalent circuit of photovoltaic devices including series and shunt resistances. . 34 Figure 3.1: Crystal structure and band diagram of organic - inorganic hybrid perovskites ........... 38 Figure 3.2: Illustration of the device architectures of perovskite solar cells ................................ 38 Figure 3.3: Illustration of preparing perovskite thin films by one - step method ........................... 42 Figure 3.4: Illus tration of preparing perovskite thin films by two - step methods ......................... 43 Figure 3.5: Absorption coefficient of ITO/PEDOT:PSS/MAPbI 3 sample. ................................ .. 44 Figure 3.6: Schematic of the ion migration in the perovskite material ................................ ......... 52 xii Figure 3.7: Schematics of photocarrier separation, collection and recombination in OPVs ........ 54 Figure 3.8: Schematics of work function tuning via the formation of interface dipole ................ 58 Figure 4.1: Images of the spin - coater and the hotplate ensem bled in the glovebox. .................... 61 Figure 4.2: Images of the thermal evaporator used for deposition of MoO 3 , C60, BCP and Ag layers. ................................ ................................ ................................ ................................ ............ 63 F igure 4.3: Illustration of the basic setup of AFM and tip - sample force ( F ts ). ............................. 68 Figure 4.4: Images of AFM (MFP - 3D) and the cantilever holder. ................................ ............... 69 Figure 4.5: Examples of Lennard - Jones potential curve and the tip/sample force curve ............. 70 Figure 4.6: Schematics of the contact and tapping modes, and an example of the tip - sam ple force curve ................................ ................................ ................................ ................................ .............. 71 Figure 4.7: Illustration of the resonance frequency shift when the tip - sample interaction falls into attractive and repulsive regimes. ................................ ................................ ................................ ... 74 Figure 4.8: Schematic of pc - AFM setup ................................ ................................ ....................... 75 Figure 4.9: Schematics of KPFM setup and the operation mode. ................................ ................ 76 Figure 4.10: Illustration of the working principle of KPFM. ................................ ...................... 77 Figure 4.11: Cosine - based phasor diagram for a purely capacitive circuit. ................................ .. 80 Figure 4.12: Examples of Nyquist (a) and Bode (b), (c) plots of a RC parallel circuit. .............. 82 Figure 4.13: Schematics of the inelastic and elastic scattering of electrons ................................ . 85 Figure 4.14: Schematic layout of EBSD system. ................................ ................................ .......... 86 Figure 4.15: Illustration of the backscattered electrons diffraction and the fo rmation of Kikuchi bands on the phosphor detector. ................................ ................................ ................................ .... 87 Figure 4.16: Illustration of Hough transform ................................ ................................ ................ 88 Figure 5.1: SEM images of perovskite films processed under different conditions ..................... 91 Figure 5.2: XRD patterns and J - V characteristics of perovskite devices with different annealing treatments ................................ ................................ ................................ ................................ ...... 93 Figure 5.3: Schematics of annealing profiles and nucleation/crystallization processes ............... 95 xiii Figure 5.4: Plot of J - V curves scanned in forward and reverse directions ................................ ... 97 Figure 5.5: Histograms of device characteristics of perovskite solar cells ................................ ... 98 Figure 5.6: Plots of EQE and absorba nce spectra of perovskite devices ................................ ...... 99 Figure 5.7: Plots of EIS results measured under illumination at varied bias .............................. 100 Figure 5.8: Plots of capacitances, resistances and recombination time constants at the open - circuit condition versus the light intensity ................................ ................................ ............................. 101 Figure 5.9: Plots of V oc and J sc versus the light intensity. ................................ ........................... 102 Figure 5.10: Plots of resistances and capacitances at the short - circuit condition versus the light intensity ................................ ................................ ................................ ................................ ....... 104 Figure 5.1 1: Plot of J sc and PCE dependence on the perovskite layer thickness. ....................... 106 Figure 5.12: Plots of capacitances, resistances and recombination time constants of devices with thick perovskite films at the short - circuit condition versus the light intensity ........................... 106 Figure 5.13: Plots of capacitances and resistances of devices with thick perovskite films at the open - circuit condition versus t he light intensity ................................ ................................ ......... 108 Figure 6.1: SEM images of perovskite thin films and J - V characteristics of perovskite devices 111 Figure 6.2: Photocurrent map and line profile of perovskite thin films ................................ ...... 112 Figure 6.3: Illustration of the correlation between the grain area and the photocurrent of different types of grains ................................ ................................ ................................ ............................. 114 Figure 6.4: Plots of averaged photocurrents and point J - V curves of grains .............................. 116 Figure 6.5: Cross - sectional SEM image of the perov skite film with ~1 um thickness. .............. 118 Figure 6.6: Images of EBSD patterns on the same grain. EBSD patterns taken at three different locations within one grain, as marked on the top - view SEM ima ge. ................................ ......... 119 Figure 6.7: Images of EBSD patterns and indexing taken on different grains ........................... 120 Figure 6.8: XRD pattern of perovskite th in film with columnar structures ................................ 121 Figure 6.9: Images of similar EBSD patters taken on different grains ................................ ....... 122 Figure 6.10: Plots of averaged photocurrents and point J - V curves of grain boundaries ........... 125 xiv Figure 6.11: Schematic of the tip - sample contact and the electric field distribution at grain boundaries and grain inter iors. ................................ ................................ ................................ .... 126 Figure 6.12: Images of KPFM results and schematic of band bending at GBs .......................... 128 Figure 7.1: Band diagram and J - V curves of inverted OPVs with ZnO processed at different conditions ................................ ................................ ................................ ................................ ..................... 132 Figure 7.2: J - V characteristics for devices with varied ZnO film thickness. 0 layer, 1 layer, 2 layers and 3 layers of ZnO buff er prepare d at 300 and 450 o C, respectively. ................................ ....... 134 Figure 7.3: Transmittance and XRD spectra of ITO/ZnO films annealed at different temperatures ................................ ................................ ................................ ................................ ..................... 135 Figure 7.4: AFM images of ZnO films processed under different conditions ............................ 137 Figure 7.5: Schematic of the interface energy level alignment at different interfaces ............... 139 Figure A.1: Solvent engineering of spin - coating method preparing halide perovskite thin film149 Figure A.2: SEM images of perovskite films prepared from precursors with different solvents 150 1 Chapter 1 Introduction 1.1 Motivat ion I n the past few decades , m any efforts have been spent o n developing new techniques to u tilize different energy sources to replace fossil fuels due to the limited supply and the severe environmental consequences. T hough the formation of fossil fuels is a natural process, it takes millions of year s to form and the depletion rate is much faster than the formation rate . 1 - 2 In addition, the carbon dioxide emission by human activities has broken the natural balance of carbon cycle which has existed for thousands of years , leading to the s erious greenhouse effect . 3 - 4 In the US, for example, over 80% of the total energy consumption is provided by fossil fuels in 2016 . 5 - 7 The search for renewable energy sources has been going on for years. Considering the above facts, solar energy as a clean and renewable source draw s an increasing amount of attention. First, solar energy is abundant. The energy strik ing the earth with in one h our ( 4.3 10 20 J ) is close to the world energy consumption in 2016 (5.6 10 20 J ) . 8 - 9 Second, it is a clean energy source which satisf ies the nee d of a daunting amount of carbon - neutral energy. La st, the 6.5 - billion - year lifetime of the sun makes solar energy quite sustainable. In order to utilize solar energy , solar cells are employed to convert solar power into electricity . Since the photovoltai c effect was first observed in 1839 by Alexandre - Edmond Becquerel, solar cells have developed into the third generation. The fi rst generation (wafer - based) solar cells are made of single/polycrystalline silicon . 10 - 1 3 And the second - generation cells are called thin film solar cells employing materials such as amorphous silicon 12, 14 - 15 and copper indium gallium selenide (CIGS) 16 - 18 . T he emerging third generation consist s of many different types of systems 2 such as organic photovoltaics , 19 - 21 dye - sensitized solar cells , 22 - 25 and perovskite solar cells 26 - 28 . So far, a large number of effort s have been spent o n the research of third generation cells since they promise to accomplish the goal of low - cost and high - efficiency solar cells. In this work , the study focuses on investigating the impact of microstructures and interface energetics on the device performance and is carried out on perovskite and organic solar cells. Organic - inorganic hybrid perovskite solar cells have drawn a tremen d ous amount of attention due to its low cost, facile fabrication and high efficiency. 27 - 30 Perovskite films are prepare d simply by spin - coating the precursor solution onto the substrates, followed by a low - temperature annealing treatment. 28, 31 The h igh light absorption coefficient has been established in this class of material, where ~ 90 % of light (wavelength: 350 - 750nm) can be absorbed in a 400 nm thick film. 32 - 33 Additionally, the low exciton binding energy on the order of tens of millielectronvolt ( meV ) indicates that free charge carriers are predominantly generated under illumination. 34 - 36 B eyond the light harvesting capability, carrier transport and charge collection also collectively contribute to the efficiency of photovoltaic devices. 37 - 40 Due to the polycrystalline nature of perovskite thin films, it is critical to investigate the impacts of microstructures, such as grain size, crystallinity and texture structure, on the electrical properties of thin films . 30, 41 - 42 More interestingly, studies have s uggested that the texture of the perovskite thin film may play a more significant role in determining the device efficiency than the size of perovskite grains. 32, 43 Thus far, time - resolved photoluminescence(tr - PL) is the most commonly ad o pted technique for investigating the influence of crystal orientation and texture structure on charge carrier dynamics in perovskite solar cells. Although variations of the carrier lifetime are observed in perovskite thin films wit h different textures 39, 41, 44 - 46 , it is challenging to disentangle the texture effects on charge transport and collection processes since the trap states in 3 the bulk and at the surface would both influence the car rier lifetime determined from the tr - PL measurements 47 . By employing the crystallinity and texture structure of perovskite thin film s and thus investigate its correlation with device performance. In addition, through the use of EIS measurements on perovskite devices , th e charge transport and collection processes can be differentiated based on their responses to the light intensity, the applied bias and the frequency. Taken together, the correlation between the crystallinity/texture structure and carrier dynamics in perov skite solar cells can be investigated. The other important feature of polycrystalline thin film s is grain s and grain boundar ies (GB s ), which could demonstrate different photo - responses depending on the crystal orientation of grains and the type of GBs, su ch as high - and low - angle GBs . Thus far, many techniques with spatial resolution s have been employed to investigate the ir role s in perovskite thin films. The g rain - to - grain variation has been revealed in the measurements including the electron beam - induced current and the confocal fluorescence microscopy, suggesting a heterogeneous contribution of crystalline grains to the photocurrent. 48 - 50 T he effect of GBs on the device performance is still quite controversial. O n one hand, a lower photoluminescence (PL) intensity and a shorter local PL lifetime ha ve been identified at GBs as compared with grain interiors, indicating that GBs act as non - radiative recombination centers. 51 On the other hand, the d ensity functional theory (DFT) calculation has predicted that GBs should be benign in halide perovskites due to the dominant shallow defect states, which distinguishes halide perovskites from other existing solar cell absorbers. 52 - 53 Despite much progress, there are several outstanding questions that remain to be addressed. For instance, as mentioned above, studies have suggested that the texture of perovskite thin films may play a more significant role than the si ze of perovskite grains in determining the device 4 efficiency. 43, 54 This indicates that the grain orientation as well as the grain boundary type, i.e., low - angle boundaries formed between grains of the identical cr ystal orientation vs. large - angle boundaries between grains of different orientation s , could lead to drastically different photo - response s . From a fundamental perspective, beyond modulating the carrier diffusion length and transport efficiency, the grain o rientation may also impact the charge transfer and collection efficiency at contacts. 39, 55 Scanning probe studies of perovskite thin films have reported substantial spatial variation s in the local photo - response, including J sc and V oc . 48 - 49, 51, 56 O ne has attributed the significant intra - grain heterogeneity to the facet - dependent density of surface trap states . 56 W hile the interconnected aggregates of several adjacent grains, not just within individual grains, exhibiting similar current levels have also been identified, indicating that there is a network of beneficial and de trimental current pathways. 49 Nevertheless, becau se of the complications arising from the stacking of grains with buried grain boundaries and interfaces, it is challenging to identify, for instance, the high conductivity pathways that promote the transport across interconnected grain aggregates. Thus, it is necessary to carry out local J - V measurements with AFM on perovskite grains of columnar structure s and distinct crystal orientation s , and GBs of the well - classified type to evaluate their contribution s to the overall power conversion efficiency. In ad dition to perovskite solar cells, OPV s have also stood out as an alternative clean energy solution, offer ing potential benefits of low cost and mechanical flexibility 57 - 58 . It is known that in OPV devices the energ y level alignment at the interfaces between metal electrodes and photoactive layers is important in determining J sc , V oc , and fill factor ( FF ) . 59 - 66 At weakly interacting interfaces such as the spin - coated polymers on non - reactive substrates, the vacuum level alignment is often assumed in the OPV design. 67 However, Fermi level pinning to the integer charge - transfer states of organic 5 semiconductors has been observed in a number of systems including poly - 3 - hexylthiophene ( P3HT ) and [6,6] - phenyl C61 butyric acid methyl ester ( PCBM ) . 61, 68 Interfacial buffer layers , therefore , play a critical role in adjusting the contact properties between active layers and electrodes. For instance, in the presence of non - ohmic contacts in BHJ devices, injection barriers may give rise to a reduction of the internal electric field and a decrease in V oc , 69 - 70 while interfacial buffer layers can be readily utilized to optimize these contacts 71 - 73 . Additionally, interfacial buffer layers may contribute to the enhancement of charge collection and the reduction of the inter facial contact resistance and charge recombination, leading to a smaller series resistance ( R series ), a larger shunt resistance ( R shunt ), and hence the improved performance 59, 62 - 63, 74 - 75 . Among the n - type buffer s used in the inverted structures, 59, 76 - 80 ZnO offers the advantages of high conductivity, excellent optical transparency and environmental stability 73, 81 - 82 . Various pre paration methods have been u sed to fabricate high - quality ZnO films in planar and nanowire configurations for solar cell applications 83 - 90 . Among them, the sol - gel method is considered to be cost - effective and comp atible with the solution processing of organic solar cells. The sol - gel method also allows for ZnO - based nanostructuring, 91 - 92 elemental doping, 75, 93 and surface modificat ion, 63, 94 to improve the ZnO functionality. Here we focus on inverted OPVs with the ZnO cathode buffer layer fabricated by the sol - gel method from zinc acetate decomposition. Several parameters can be tuned in the sol - gel process to improve the morphology, as well as its optical and electrical properties of ZnO thin film s . However , there have been conflicting reports of this optimization in the literature 74, 95 . O ptimal ann ealing temperatures reported in the literature range widely from 150 to 450 °C 60, 74, 89, 91, 95 - 102 , although it is recognized that low - temperature annealing is more compatible with the processing of plastic orga nic solar cells. Therefore, a thorough investigation of the physical characteristics of ZnO buffer layers annealed at different temperatures is necessary. 6 1.2 Outline of this work To better understand the work ing principle of solar cells , semiconductor phy sics and physical processes in photovoltaic devices are introduced in Chapter 2. The device architecture based on planar structured solar cells as well as the physical processes such as photo - generation, c arrier transport , charge collection , and recombinat ion will be the focus of the discussion . With a basic understanding of the mechanism of photovoltaics , J - V characteristics are introduced, and the factors that impact the device performance are discussed. After getting acquainted with solar cell physics, perovskite solar cells and OPVs are introduced in Chapter 3 . First, the physical properties of perovskite materials and their contributions to the device performance are discussed. Also, t he current sta tus of perovskite solar cells is reviewed to assess th e challenges and problems to be addressed . D ifferent fabrication methods are also described since they determine the film quality . The second part focuses on the device architecture and physical processes in OPVs. The BHJ based OPVs are described and comp ared with planar perovskite solar cells. The critical role of the electrode/active layer interface is then discussed, and different methods to tune the interface properties are demonstrated . A t last, ZnO and its application in OPVs are illustrated as well as the challenges of tuning its electrical propert y . B efore reaching the experimental results and conclusions, the device fabrication processes and experimental techniques that help to address the physical properties of solar cells are described in Chapte r 4 . The fabrication process is important as it determines the quality of each part in the devices and thus the efficiency. Details about the working principle of AFM, EIS , and EBSD and how to interpret the data are described. Chapter 5 d iscusses the impa ct of microstructure in perovskite thin films on the device performance. T he texture structure and morphology of perovskite thin film s are tuned by ramp 7 annealing treatment, which includes a two - step annealing profile . T he correlation between device perfor mance and the thin film crystallinity /texture is investigated . To elucidate the impact of the texture structure on the carrier dynamic process es in perovskite solar cells, EIS , which is able to disentangle the convoluted charge transport and collection pro cesses , is performed on devices with different thickness and under different light intensities and applied bias es . To further understand how the microstructures, specifically grain s and grain boundar ies , impact the device performance , AFM techniques are u tilized to investigate the correlation between microstructure and nano - scale device performance in Chapter 6. The discrete photocurrent levels across crystalline grains and the anti - correlation of J sc and V oc of each grain are identified by p c - AFM and poin t J - V spectra o n perovskite thin films . With the aid of electron backscattering diffraction (EBSD) measurement , the crystal orientations of grains are revealed. Additionally, GBs in perovskite thin films demonstrate heterogeneous behaviors with low - angle G Bs showing a higher power output than the adjacent grains. KPFM is utilized to investigate the energetics at grain boundaries . Chapter 7 discusses the application of ZnO as a buffer layer between the cathode material , i.e., indium tin oxide (ITO) and the active layer of P3HT and PCBM . The work function of ZnO thin film is tuned , and its impact on the device performance is attributed to the efficient charge collection at the interface. Modulations on the interface energetics are analyzed by KPFM. In Chapte r 8 , the thesis studies are summarized , together with future work from the perspective s of device engineering and instrument ation development . 8 Chapter 2 Solar cell physics In this chapter, the semiconductor physics is introduced in the first part. The b and structures, the doping of semiconductor s and the formation of junctions between metal s and semiconductors are discussed , as well as their applications in photovoltaics. The difference between organic and inorganic semiconductors is also described . In t he second part, the configuration of planar structured solar cell is used as an example to illustrate the physical process occurring in the devices under illumination . The process es from the generation of excitons to the formation of photocurrent are detai led. Lastly, the J - V characterization of solar cells are described. The device performance parameters are defined and t he factors impact ing the characteristic parameters are deliberated . 2.1 Introduction to s emiconductor s 2.1.1 Basics of semiconductors S emiconductors as the core of semiconductor devices are widely used in our daily life owing to the unique electrical properties compared with insulators and conductors. Generally, semiconductors demonstrate certain crystal structure s : a cluster of repetitiv e atom arrangements. When two atoms are brought together, the atomic orbitals would split into two new orbitals with higher and lower energy than that of the original orbitals. With more and more atoms form ing a cluster, the spli t orbitals form bands as sh own in Fig ure 2.1. If the material show s continuous electronic states without a gap, it is a conductor. For semiconductors and insulators, there is a bandgap in the energy diagram . The band above the gap is defined as conduction band (CB) and the one below is the valence band (VB). The bandgap of semiconductors is around 0.5 to 3.0 eV. Materials with bandgap s larger than 3 eV are called insulators. At the low temperature and under the dark condition , semiconductors are barely conductive since not many elect rons can be excited 9 to the CB. However, when enough energy is provided to the semiconductor, such as incident photons and thermal energy , electrons in the VB can be excited to the CB and the material become s conductive For the organic semiconductor s , wh ich are - bonded molecules or polymers , they form relatively dense but discrete levels instead of continuous bands. The highest occupied level is called HOMO, similar to the concept of valence band maximum (VBM) in inorganic semiconductors. The lowest uno ccupied level is called LUMO, corresponding to the conduction band minimum (CBM) . The difference between LUMO and HOMO equals the bandgap of the material . Compared with inorganic semiconductors, organic semiconductors demonstrate smaller dielectric constan ts because of the diminished charge screening cap ability. Figure 2 . 1 : Band formation of semiconductors : as the number of atoms increases, the atomic orbitals split into multiple levels and thus form the conduction band and the valence band. To investigate the carrier population in semiconductors, the density of states and F ermi level function are two key parameters. The density of state s is a prerequisite for determin ing the 10 carrier concentration and energy distributions of carriers in semiconductors. It equals the number of allowed electron states per energy per volume. Through the approximation of a particle in a 3 - D box, the density of s t ates ( D( E ) ) is expressed by : ( 2 . 1 ) ( 2 . 2 ) Here, and are the effective mass of electron and hole , respectively , which is related to the band structure of the semiconductor. In addition to the density of states, the F ermi function is also cr ucial to calculate the number of electron s occupy ing certain electronic states. The F ermi function f( E ) , as a function of energy, is equal to the ratio of filled state s over all allowed state s at a given energy. It can be described by the following equatio n: ( 2 . 3 ) E f is called the F ermi level as shown in Fig ure 2. 1 which lies within the bandgap. For intrinsic semiconductors, the F ermi level is close to the center of the bandgap. If the material is doped, its F ermi level would shift upwards or downwards accordingly. 2.1.2 Intrinsic and extrinsic semiconductors Semiconductors containing an insignificant number of impurity atoms are ref erred as intrinsic semiconductors. Since the only way to generate electron s and holes is to excite electrons in the VB to the CB in intrinsic semiconductors , the population of electron s and hole s are always equal . However, semiconductors with a high electr on (n - type) or hole (p - type) concentration are widely used in semiconductor devices. These types of semiconductors are termed as extrinsic semiconductors and show different electrical properties than intrinsic ones . In order to manipulate 11 the carrier conce ntration, the specific impurity atoms are added into the semiconductor, which is described as doping. Figure 2. 2 : Schematic of semiconductor doping . (a) n - and p - type doping of silicon with phosphorous and boron atoms; (b) the influence of donor states ( E D ) and acceptor states ( E A ) on the position of F ermi level. Figure 2.2 depicts the semiconductor bonding model of Si, which help s us understand the mechanism of the doping process . As a n IV column semiconductor, there are four valence 12 electrons and four neighbors for each Si. By sharing the valence electrons, the covalent bonds are formed between Si atoms. When a silicon atom is replaced by a phosphorous atom, five valence electrons are available for bonding with the adjace nt Si atoms. However, there are only four Si neighbors and each of them needs only one electron to form the covalent bond. The extra electron is loosely bound to the phosphorous atom and can be easily freed from it by the thermal energy provide d at room te mperature. In this case, the phosphorous atom is called donor. In the energy band diagram (Fig ure 2.2) , it can be seen that the energy level ( E D ) introduced by replacing Si with P is very close to the conduction band edge and the electrons can be excited t o the CB at room temperature. Correspondingly, the F ermi level is shifted upwards (closer to CB) compared with the intrinsic case. Vice versa, the replacement of Si with a Boron atom which only provide s three valence electrons would result in an electron vacancy (hole). The adding of Boron atom s ( acceptor ) introduce s the energy levels close to VB ( E A ) which are able to accept electrons excited by the thermal energy at room temperature from the valence band. Thus, the F ermi level shifts downwards to the VB. The doping source can be intrinsic and extrinsic. The example shown above is the extrinsic doping of Si. For solution - processed perovskite materials, both n - type and p - type doping have been identified depend ing on the fabrication process and the types of intrinsic defect s , such as the vacancy/interstitial of I - or MA + . 2.1.3 Equilibrium distribution of carrier s The density of states and the F ermi function have been introduced previously. The carrier concentration in the intrinsic semiconductors can th en be calculated by the following equation s : ( 2 . 4 ) ( 2 . 5 ) 13 Here, n and p are the concentration of electrons in conduction band and holes in valence band, respectively. According to Boltzmann approximation, if the F ermi level is far enough from both band edges, the F ermi fu nction can be simplified by : ( 2 . 6 ) And n and p can thus be written by : ( 2 . 7 ) ( 2 . 8 ) ( 2 . 9 ) ( 2 . 10 ) N C and N V are termed as the effective density of states of the conduction band and t he valence band, respectively. According to the above equations, the product of n and p equals: ( 2 . 11 ) ( 2 . 12 ) Here, we introduce the intrinsic carrier concentration ( n i ): the electron and hole concentration in intrinsic semiconductor s , which is related to the bandgap and the density of states. For example, in Si , the intrinsic carrier concentration at room temperature is 1.05× 10 10 cm - 3 . For the extrinsic semiconductor s , the above equations still hold. However, the electron and hole concentration s are not equal anymore. For n - type semiconductor s , the electron density in the conduction band is given by : 14 ( 2 . 13 ) where is the density of ionized donors. At 0 K, all donor states are full so n = 0. As the temperature increases, donor electrons a re excited to CB and since . As equation (2.11) still holds, the hole density in the valence band is written by : ( 2 . 14 ) At even high temperature s , a large number of electrons in the valence band are excited to the conduction band ( and n equals n i again. Figure 2. 3 : Temperature dependence of carrier concentrations in n - type semiconductors . 2.1.4 Junctions Some basic physical properties of semiconductors have been introduced in the previous content. P hotovoltaic devices consist of multiple layer s of materials in order to generate photocurrent. A junction forms when two materials are in contact, which could facilitate the charge transport and collection in the devices. Thus, it is necessary to talk about the scenario when a semiconductor is in contact with materials such as metals and semiconductors. 15 2.1.4.1 Metal - semico nductor junction Suppose that there is a n n - t ype semiconductor with the work function of and a metal with the work function of ( ) as shown in Fig ure 2. 4 . Before they are in contact, the vacuum level s align indicating that CB, VB and E f of the semiconductor and the metal are independent. However, when these two materials are in contact, the F ermi levels level up to reach the equilibrium , leading to the shift of the vacuum level . E lectrons in the semiconductor flow to the met al and leave the positive charges in the region close to the metal which is called the space charge region . The width of the space charge region depends o n the dielectric constant of the semiconductor ( s ) , the doping level of the semiconductor ( N D ), the vacuum level change at the junction ( V bi ) and the applied bias ( V A ), as described by: ( 2 . 15 ) The n - type semiconductor with a high doping level, for example, demonstrates a narrow depletion region due to the high concentration of electrons. On the other side of the junction, t he corresponding negative charges in the metal is l imited to the interface owing to its high dielectric constant. The separation of electron s and holes forms a n electric field which suppress es the charge transfer at the contact . T he intensity of the electric field increases as more charges are separated. U ltimately, the electric field would stop the charge flow at the junction. Though the space charge region in the semiconductor is positively charged, the conduction and the valence band position s relative to the vacuum level are the same as the regions away from the junction. As the vacuum level in the space charge region changes, the CB and VB change accordingly, the process of which is described as the band bending. Th is junction formed between the n - type semiconductor and the metal is referred as the Scho ttky barrier which shows a rectifying characteristic. 16 Figure 2 . 4 : Formation of Schottky barriers : (a) n - type semiconductor and metal junction when ; (b) p - type semiconductor and mental junction when . Figure 2. 5 illustrates the rectifying characteristic of the Schottky barrier formed between the n - type semiconductor and the metal. As mentioned before, at the thermal equilibrium there is not net current flow (the number of electrons flowing to the metal equals the number of holes to the semiconductor). Under a reverse bias, a positive voltage is applied to the semiconductor which enlarge s the band bending and further suppress es the electrons from flowing to the metal . Consequently, the net curr ent is determined by the leakage current resulting from the movement of hole s to the metal. In n - type semiconductor s , hole s are the minority carrier s , so the current level is quite small. Under a forward bias, a negative voltage is applied to the semicondu ctor, and Schottky barrier is reduced . T hus , the flow of electrons to the metal is enhanced. As the barrier height decreases, the current level increases approximately exponentially with the bias. Similar ly, 17 Schottky barrier is observed at the junction of the p - type semiconductor and the metal when as shown in Fig ure 2. 4 b. Figure 2 . 5 : Rectifying characteristics of the metal/semiconductor junction . The situation is different if the work function of the metal is higher than that of the n - type semiconduc tor. As shown in Fig ure 2. 6 , the electrons flow from the metal to the semiconductor in order to reach the thermal equilibrium . T his type of junction does not show the rectifying characteristics and is called Ohmic contact since the majority carriers is abl e to pass the junction in either direction with a low resistance. Similarly, for the p - type semiconductors and the metals, when , an Ohmic contact form s at the junction. There are other ways to form an Ohmic contact in addition to the satisfaction of or . In the heavily doped semiconductors, for example, the space charge region could be quite narrow ( several nanometers ) and thus the majority carrier s are able to tunnel through the barrier . At the organic semiconductor /metal junction, the situation is different . The formation of the junction , in this case , is described by the Integer Ch arge - Transfer model. 103 The charge transfer 18 at the organic semiconductor/metal interface is dominated by the tunneling of electrons because of the negligible hybridization of - electronic molecular orbitals and the substrate wavefunctions. The tunneling of t he electrons indicates that the transfer process involves an integer amount of charge to the well - defined charge states located at the surface of the organic semiconductor. Figure 2. 6 : Formation of Ohmic contacts at the met al/semiconductor junction . B etween metal and (a) n - type semiconductors when , (b) p - type semiconductors when There are the positive and the negative integer charge - transfer states ( E ICT+ and E ICT - ). The energy of E ICT+ is defined as the energy required to take away one electron from the molecules , and the energy of E ICT - is the energy gained by the molecule when adding an electron to it. The generation of the Integer Charge - Transfer states is caused by the formation of the self - localized states, also called polaron s , when reducing or adding an electron to the organic molecules. As 19 depicted in Figure 2.7, when the work function of the metal, , falls between the E ICT+ and E ICT - , the vacuum level alignment is expected at the interface. If is larger than E ICT+ , electrons would be taken away from the organic molecules and the Fermi level is pinned to the E ICT+ . Similarly, the Fermi l evel would be pinned at the E ICT - when the Fermi level of metal is above E ICT - as a result of the addition of electrons to the organic molecules. It is worth mentioning that this also holds for the semiconducting organic/inorganic interfaces with negligibl e interaction s . Figure 2. 7 : Schematics of the formation of junctions at the semiconducting organics and metal interface: when (a) is smaller than E ICT - , (b) is between E ICT - and E ICT+ , and (c) is larger than E ICT - . 20 Figure 2 . 8 : Illustration of the band alignment at (a) p - n and (b) p - i - n junction s . 2.1.4.2 p - n junction and p - i - n junction The p - n junction is the most widely adopted structure in photovoltaic devices. When an n - type and a p - type semiconductor are in contact, the junction formed is called the p - n junction. As mentioned before, the F ermi levels level up if two materi als are in contact and a n electric field is introduced at the junction . As shown in Fig ure 2. 8 , electrons in the n - type material flow to the p - type material and hole s in the p - type material flow to the n - type material. Thus, the space charge regions exist on both sides of the junction. The junction always behaves as a barrier for the majority carriers and a high conductivity path for the minority carriers, which is beneficial for the charge separation. In solar cells, the separation of electrons and holes i s crucial to the formation of photocurrent. If the carrier diffusion length in the semiconductor is short and photo - generated carrier s in p - or n - type materials are not likely to contribute to the photocurrent, the p - i - n junction is advantageous. The jun ction is formed by inserting a layer of the intrinsic semiconductor between 21 n - and p - type materials. The built - in electric field is extended to a wider region as shown in Fig ure 2. 8 b and drives electrons and holes generated in the intrinsic layer to the op posite directions. Figure 2 . 9 : Planar structure d solar cells: active layer, anode and cathode . Under illumination, the active layer absorbs photons and generate s excitons which are then dissociate d into free carriers. 2 . 2 Planar structured solar cells In this part , planar structured solar cells are used as an example to illustrate the working principle of photovoltaic devices. Planar structured solar cells consist of three parts: active layer, cathode and anode as shown in Figure 2. 9 . The active layer, which is made of semiconductors, absorbs light and converts it into electrons and holes via the photovoltaic effect. T he generated electrons and holes are then transported to the electrodes. As a result of the selectivity of electrode materials, the cathode and anode colle c t electrons and holes, respectively. Then, the free carriers form an electrical current and contribute to the external circuit. 2. 2 .1 Light absorption and exciton dissociation The active layer absorb s photon s and generate s charge carriers . T he number of photons that the active layer is able to absorb is determined by many factors. First, the active layer can only 22 absorb the photons with the energy higher than its bandgap. Semiconductors with low bandga p utilize more of the solar spectrum and thus generate more photocarriers and output a higher current . Second, the quantity of the light absorption depends on the absorption coefficient which is wavelength dependent. Photons with the energy just above the bandgap only excite electrons around the valence band edge to the CB edge , while high energy photons can excite electrons deep in the VB . In addition, the absorption coefficient depends on the density of states in the conduction and valence band of semicon ductors. the excited electrons in the conduction band. Last, the thickness of the active layer influences the absorption of photons. More photons can be absorbed in thicker film s . However, if the film is too thick , photocarriers must travel a long distance to the electrode , which increase s the possibility of be ing nullified by the counter charges (more details are shown in the latter part of this chapter). An optimized film thickness is essential t o balance between the light absorption and carrier transport process in photovoltaic devices. After the absorption of photons , it generates the electron - hole pairs , which are called exciton s, in the semiconductor as shown in Figure 2. 9 . There is a binding force between the electron and hole in the exciton . Consequently, i t requires an external force to dissociate excitons into free carriers . The bound of the electron and the hole is similar to the coulombic interaction between the positive and negative cha rge s proportional to the dielectric constant : ( 2 . 16 ) where r is the distance bet ween the two charges. The energy required to dissociate the exciton is called the binding energy, which is also related to the dielectric constant of the active layer 23 material . 36, 104 S emiconductor s with high diele ctric constant s , such as organic - inorganic hybrid perovskite s , demonstrate low exciton binding energy : tenths of meV . For organic semiconductor s with small dielectric constant s , the binding energy could be as high as hundreds of meV. Consequently , in OPVs, t he donor - acceptor system is introduced to create an energy offset at the interface of the donor and the acceptor to dissociate excitons . 19, 105 More details on the donor - acceptor system will be discussed in next chapter. 2. 2 .2 Charge transport With the excitons being dissociate d into free carriers, the next step is charge transport. Electrons and holes must travel across the active layer to the cathode and anode, respectively. The mechanism of charge transport process in the active layer is the drift and diffusion. 2. 2 .2.1 Drift Drift is defined as the motion of the charged particle in response to an electric field. When an electric field is applied to the device , it drives holes moving towards the direction of the electric field and the electrons in the direction opposite to the electric field. Thus, the drift current is given by: ( 2 . 17 ) ( 2 . 18 ) E is the electric field in the device , and n and p are the carrier mobility for the electron and hole, respectively, with the unit of cm 2 V - 1 s - 1 . The origin of E - field could be an external bias applied to the device or a built - in electric field establi shed between materials of different work function s . For example, in photovoltaic devices, the cathode and anode consist of different materials . The anode material which collect s hole s demonstrates a low work function. For the cathode, material s 24 with high w ork function s are employed to collect electrons. When the cathode, the anode and the active layer are in contact, a built - in electric field form s due to the alignment of F ermi - level s , as illustrated in Figure 2.8 . When solar cells are under illumination, t he built - in electric field draws the photo - generated electrons to the cathode and hole s to the anode, forming the photocurrent . The carrier m obility is the key parameter in determining the performance of solar cells. Carriers with a high mobility suppress the recombination of electrons and holes and improve the charge transport and collection efficiency . One factor that influences the carrier mobility is the scattering process taken place in the active layer. For semiconductor s , the phonon scattering and t he ionized impurity scattering processes tend to be dominant. The p hono n scattering refers to the collision between the carriers and the vibrating lattice atoms. T he ionized impurity scattering is related to the coulombic interaction (attraction or repulsi on) between the carriers and the ionized donor s /acceptors , especially in the highly doped semiconductors . In addition, for polycrystalline materials, like organic - inorganic hybrid perovskite thin films , the defect scattering mus t be taken into consideratio n due to the break of crystal symmetry at the grain boundaries. The carrier mobilities for electron s and hole s could also be different in the same material owing to the different effective mass . The imbalanced carrier mobility requires more efficient charg e collection at the electrode which collect s the carrier s with the lower mobility . 2.2.2.2 Diffusion Diffusion is the process in which particles tend to redistribute to reach a concentration equilibrium as a result of the random thermal motion. Consequen tly , a net movement of carriers goes from the areas of the high concentration to the areas of the low concentration in solar cells . The corresponding current s are expressed by the following equations: ( 2 . 19 ) 25 ( 2 . 20 ) where are the diffusion coefficient of elec tron s and hole s with the unit of cm 2 s - 1 ; are the concentration gradient of electron s and hole s . The diffusion coefficient, similar to the carrier mobility in the drift process, is crucial in determining the carrier diffusion process. The relationship between the diffusion coefficient and the carrier mobility has been built by Einstein: ( 2 . 21 ) In the studies of solar cells, many te chniques , such as Hall measurement, PL q uenching a n d time of flight , have been employed to extract the carrier mobility information in order to characterize the transport properties of devices. 2.2.3 Charge Collection After the photo - generated electrons and holes arrive at the electrode, they must be selectively collected by the electrodes in order to contribute to the external circuit , instead of being the counter charges. The selectivity of cathode s and anode s is enhanced by a selective contact layer inserted between the active layer and the electrodes, named by the electron or hole transport layer (ETL/HTL). Generally, m aterials used in the selective contact layer are highly doped n - type and p - type semiconductors for ETL and HTL, respect ively , with a high carrier mobility . For example, n - type materials employed as ETL usually show high selectivity and conductivity of electrons and a deep VB to block holes . 106 - 107 Another factor to consider here is the energy level alignment at the contact. The interface energetics is critical since it directly impact s the collection efficiency. Ideally, a n Ohmic contact is expected at the active layer/ selective contact and the selective contact/ electrode interfac es to avoid the energy loss. However, the 26 formation of the Schottky barrier is the common case in solar cells. Meanwhile, at the interfaces, there exists a high density of defects which lead s to severe recombination. More details about surface recombinatio n will be discussed in the next section. Figure 2. 10 : Schematic of charge transport driven by the built - in electric - field . (a) I llustration of the carrier transport process in the active layer; (b) the built - in electric fi eld induced by the work function difference of electrodes drives electrons and holes to the cathode and the anode, respectively. 27 Figure 2 . 11 : Schematic of charge collection at the interface . (a) The free carriers are collec ted by the cathode and the anode due to the selectivity of the electrode materials, (b) both interfaces between the active layer and ETL as well as ETL and cathode impact the charge collection efficiency. 2.2.4 Recombination , carrier lifetime and diffusio n length 2.2.4.1 Recombination in photovoltaic devices Recombination is the loss mechanism of photovoltaic devices which refers to the loss of free carriers. For example, an electron annihilates a hole instead of contributing to the photocurrent. There ar e different ways of recombination, such as radiative recombination, Auger recombination and trap - assisted recombination. Radiative recombination is the dominant loss mechanism in the direct bandgap semiconductors. As shown in Fig ure 2. 12 , a n electron in t he conduction band relaxes to the valence band and recombine s with a hole while releasing a photon. This loss mechanism is related to the 28 intrinsic property of the material and is inevitable . Though the radiative recombination is damaging the performance o f photovoltaic devices, the light emitting diode device is a good example of utilizing the radiative recombination mechanism. Figure 2 . 12 : Schematics of the loss mechanisms . (a) B and to band radiative recombination , (b) Au ger recombination , (c) trap - assisted recombination , and (d) surface recombination The other unavoidable loss mechanism is the Auger recombination which involves three charge carriers. Figure 2. 12b depicts the Auger recombination process. When an electron in the CB decays to the VB, instead of releasing the energy in the form of heat or photon s , it transfers the amount of energy, which equals the bandgap, to the other electron in the CB which eventually goes back to the CB edge via thermal relaxation. The a bove process involves the interaction among the carriers, so Auger recombination is dominant in the low bandgap materials with high carrier densities. 29 Recombination via trap states, also called Shockley - Read - Hall (SRH) recombination, denotes the recombina tion of electron s and hole s via an electronic state within the bandgap. For the perfect single - crystal semiconductors , SRH recombination does not occur due to the lack of mid - gap states. As shown in Fig ure 2. 12 c, SRH recombination includes two steps. First , an electron (or hole) is trapped in the mid - gap state which is introduced by the defects of the crystal structures . The trapping and de - trapping processes are reversible since the trapped charges can be re - emitted to the CB or VB by thermal excitation . H owever, if a hole (or electron) is also captured by the mid - gap state before the electron (or hole) is thermally re - emitted to the CB, recombination would happen. For defect states deep into the bandgap, the thermal energy required to de - trap the charge is relative ly high and it is likely to induce recombination since t he trapped charge has a high chance to recombine with a counter charge than being re - emitted. Vice versa, if the defect state is close to the band edge, the trapped charge s can be easily de - t rapped instead of recombining with the counter charges. Surface recombination is a nother critical loss mechanism for photovoltaic devices. The semiconductor surface is an area with a high density of defects as the crystal symmetry is interrupted and a lar ge number of dangling bonds exist there. As the carrier s recombine, a concentration gradient would form which drives more carriers to the surface. However, the surface recombination can be suppressed by passivating the defect sites, such as neutralizing th e charged sites by doping or forming bonds with the unpaired electrons by introducing small molecules or interfacial layer s at the contact . 2.2.4.2 Carrier lifetime and diffusion length Recombination process, as discussed before, is a loss mechanism in photovoltaic devices. It determines how long the carriers survive , how far they travel before recombining, and eventually 30 the number of carriers contributing to the external circuit. The average amount of time which the carrier can survive in an excited st ate is called the carrier lifetime. T he diffusion length is the distance that carriers are able to travel after photo generation and before recombination. I t determines the thickness of the active layer in solar cells. For example, the thickness of the acti ve layer in OPVs is around 100 nm as a result of the low carrier diffusion length. The active layer in perovskite solar cells is around 300 - 500 nm thick since the perovskite materials demonstrate a carrier diffusion length as long as hundreds of nanometers . Overall, the carrier lifetime and carrier diffusion length are the key factors that impact the efficiency of photovoltaic devices. Semiconductor devices with short carrier lifetime and diffusion length would suffer from severe recombination and demonstra te low efficiency. Carrier lifetime ( ) depends on the total recombination rate of the three recombination mechanisms mentioned above. Therefore, it can improve the lifetime of photogenerated carriers by employing semiconductors with insignificant radiative / Auger recombination and manipulati ng the density of defect state s to suppress the SRH recombination. Obviously, carriers with a longer lifetime are able to diffuse a further distance, and the correlation between carrier lifetime and diffusion length ( L ) is written by : ( 2 . 22 ) D is the diffusion coefficient and related to the intrinsic property of the semiconductor. 2. 3 J - V characteristics of photovoltaic device s J - V characteristics are m easured in order to determine the device performance of solar cells . In the dark, photovoltaic devices behave as diodes and the corresponding J - V curve is shown in Fig ure 2.1 3 . The dark current ( I D ) is written by : 31 ( 2 . 23 ) I o is called the saturation current which equals the leakage current of the device in the dark condition. n (usually larger than 1) is the ideality factor as the behavior of real devices deviate s from an ideal diode. Under illumination, the flow of photo - generated carriers in the device form s the photocurrent ( I ph ). Therefore, the total current ( I ) of the device is given by: ( 2 . 24 ) The efficiency of photovoltaic devices ( ) equals the maximum power output ( P max ) over the power input ( P in ) . P in is determined by the incident light intensity. From the J - V characteristics shown in Fig ure 2.1 3 , the power output is calculated by the product of total current and the corresponding voltage. Therefore, the maximum power output is extracted, and the efficiency is written by : ( 2 . 25 ) Here, I sc and V oc are directly read from the J - V curves. At zero bias, the current is named as the short - c ircuit current ( I sc ). When the photocurrent is canceled out by the dark current, the corresponding ly applied bias is called the open - circuit voltage . Then, fill factor can be expressed in terms of I sc and V oc : ( 2 . 26 ) The current density is generally used when talking about the device performance since the current depends on the device area. Hence, in the following discussions, we us e term J sc (short - circuit current density) instead of I sc . 32 Figure 2 . 13 : Illustration of J - V curves and the characteristic parameters of photovoltaic devices under dark and illumination conditions . 2. 3 .1 Short - circuit curre nt ( J sc ) Mathematically, J sc can be written by : ( 2 . 27 ) where f lux (E) is the photon flux of the incident light , and EQE ( the ext ernal quantum efficiency ) is defined by the number of electrons contributing to the photocurrent over the number of incident photons . First, J sc depends on the incident light intensity. The photocurrent increase s when the device is under the illumination o f a higher intensity . EQE , which also determines the photocurrent level , equals: ( 2 . 28 ) 33 LH , CT and CC refer to the efficiency of light harvesting, carrier transport , and charge collection processes as described b efore. Consequently, the active layer material, the selective contacts, the electrode materials , and their interactions coll aboratively impact the photocurrent and the device performance. 2. 3. 2 Open - circuit voltage ( V oc ) The value of V oc depends on the b andgap of the active layer semiconductor as well as the F ermi - level splitting under illumination . In semiconductors with large bandgap s , the excited electrons and holes possess more energy leading to a higher open - circuit voltage. Under the strong illumina tion , the extent of the F ermi level splitting increase s and thus the V oc . Since the photo - generated carrier s are then collected by the electrodes, the interface energetics also impact the value of V oc . For example, if an injection barrier presents or sever e recombination happens at the interface, the V oc decrease s drastically. As the V oc equals the voltage when the corresponding total current is zero, we obtain : ( 2 . 29 ) Since and depends on the light intensity , the light intensity dependence of V oc can be simplifie d as follows: ( 2 . 30 ) X is the relative light intensity with the unit of 1. By plotting V oc versus X , the ideality factor can be calculate d, which shine s light on the recombination mechanism in the devices. 34 2. 3. 3 Fill factor, shunt resistance and series resistance According to Fig ure 2.1 3 , FF represent ing the squareness of the J - V curve is related to the shunt resistance ( R sh ) and the seri es resistance ( R s ) . The introduction of R sh and R s modif ies the total current equation by: ( 2 . 31 ) The s eries resistance refers to the total resistance of the device to the movement of photocarriers. Thus, the resistance s coming from the active layer, the electrodes and the interface s between different layers contribute to R s . R s impact s the squareness of the J - V curve close to the open - circuit condition because of the result ing voltage drop across the device. Figure 2 . 14 : Equivalent circuit of photovoltaic device s including series and shunt resistances . The s hunt resistance impact s FF and the device performance since it provides an alternative path for the photocurrent as shown in Figure 2.14 . For example, when the active layer is not compact, the contact of ETL and HTL act s as a leakage path for photocurrent with the electrons in ETL recombin ing with holes in HTL . Therefore, R sh mainly affect s the shape of the J - V curv e closer to the short - circuit condition. 35 In this chapter, the semiconductor physics and the working principle of photovoltaic devices as well as the J - V characteristics are introduced. The discussion over the factors that impact the photo - generation, char ge transport and collection, and recombination processes is helpful to understand the strategies of boosting the device performance. In the next chapter, the focus is introducing the physical processes in perovskite solar cells and OPVs. 36 Chapter 3 Perov skite solar cells and organic photovoltaics In this chapter, the physical processes in perovskite solar cells and OPVs are introduced. First, the physical properties of perovskite material s and their contribution s to the remarkable device performance of pe rovskite solar cells are discussed. In addition, the current stage , as well as the challenges of perovskite solar cells , are reviewed. The working principle of organic photovoltaics is described in the second part. Especially, the critical role of the inte rfacial layer in OPVs is illustrated together with the methods to improve the charge collection at the interface. The physical properties of ZnO and its application in OPVs are presented at last. 3.1 Organic - inorganic hybrid perovskite solar cells In 200 9, organic - inorganic hybrid perovskite materials were first applied in dye - sensitized solar cells by Kojima et al. , yielding a n efficiency of 3.8 %. 108 Three years later, Snaith et al. employed the perovskite materials in solid state photovoltaic devices, which increased the efficiency to over 10%. 28 By the year of 2017, the highest certified efficiency of perovskite solar cells has reached 22.7%. 109 The drastic improvemen t of the device performance , together with the low - cost and facile fabrication of perovskite materials, has drawn tremendous attention to this type of material s in the past few years. P erovskite solar cells have been considered as one of the most promising candidate s for the next generation photovoltaic technology. 3.1.1 Basics of perovskite materials The most well - studied organic - inorganic hybrid perovskite material is methylammonium lead iodide (MAPbI 3 ) which adopts an ABX 3 formula and the perovskite str ucture a s shown in Fig ure 3.1 . In MAPbI 3 perovskite, A site is composed of methylammonium (MA) cation. The lead (B site) and iodide (X site) form an oc tahedral cage with lead in the center and iodide at the corners. 37 There are many other options to fill A, B and X positions of the perovskite materials. For A site, formamidinium (FA), Cs + and Rb + have been used to replace or partially substitute MA. Sn 2+ has been found to be a possible substitute for Pb 2+ at B site. As to X site, Cl - and Br - were mixed with iodide. Perovskite materials with different compositions demonstrate varied physical properties. For example, according to Shockley - Queisser limit, the optimal bandgap of a single junction photovoltaic device is around 1.1 to 1.4 eV which is smaller than t he bandgap of MAPbI 3 . By replacing or mixing the cations or anions of the perovskite material, the bandgap can be effectively tuned. So far, the major concerns about the commercialization of perovskite solar cells are the toxicity of lead, the stability an d hystere tic J - V characteristics of devices . The above concerns c an be addressed by modifying the composition of perovskite materials. In this study, t he focus is to investigate the electrical property of the microstructures in MAPbI 3 and its contribution to the associated device performance . The crystal structure of MAPbI 3 is tetragonal at room temperature with the octahedral PbI 6 cage corner - connected and MA cations filling the openings. 110 At 330 K, the tetragonal phase undergo es a phase transition to the cubic phase. 111 In photovolt aic devices, perovskite materials are employed in the form of polycrystalline thin films. The bandgap is a bout 1.5 to 1.6 eV for the polycrystalline MAPbI 3 thin films . According to the density functional theory (DFT) calculation, t he conduction band minimu m (around 3.9 eV) is composed of the non - bonding state of the 6 p orbitals of Pb and 5 p orbitals of iodide, while the valence band maximum (around 4.5 eV) consists of the anti - bonding of 6 s orbitals of Pb and 5 p orbitals of iodide in the Pb - I chains. 112 - 113 Though MA + does no t directly participate in the formation of band structure s , the orientation of MA cation still impact s the electronic properties of perovskite materials . 114 - 115 38 Figure 3. 1 : Crystal s tructure and band diagram of organic - inorganic hybrid perovskites . (a) Crystal structure of organic - inorganic hybrid perovskites (ABX 3 ) with a tetragonal phase and (b) the band diagram of MAPbI 3 . In the perovskite structure, A site: MA, FA; B site: Pb, Sn; X site: Cl, Br, I. Figure 3 . 2 : Illustration of the device architectures of perovskite solar cells : meso - porous structure, inverted (p - i - n) structure and conventional (n - i - p) structure. 39 3.1. 2 S tructure of solid - state perovs kite solar cell s As mentioned above, the efficiency of perovskite solar cells took off after the electrolyte was replaced by solid - state materials. In general, there are two different types of perovskite solar cells: meso - porous structured and planar stru ctured as shown in Fig ure 3. 2 . 27 - 28, 31 - 32, 40 - 41, 44, 107 The meso - porous structure is usually composed of meso - porous titanium oxide (TiO 2 ) as the electron transport layer and the scaffold of the perovskite layer . 28, 31, 44, 48 B y filling into the porous structure, it is more likely to form a continuous and compact film which is essential for high efficiency devices. One has discovered that perovskite materials exhibit an excellent carrier diffusion length (> 100 nm for MAPbI 3 ) and the ambipolar charge transport properties, indicating that photo - generated electrons and holes are able to travel hundreds of nanometers within the perovskite material before they recombine . 116 - 120 Therefore, the planar structured perovskite solar cells have drawn a n increasing amount of attention. 27, 32, 40 - 41, 107, 120 According to the configuration of planar s tructured solar cells, there are two types of planar devices: conventional and inverted structured solar cells as shown in Fig ure 3. 2 . The inverted structured ( p - i - n ) perovskite solar cells usually employ a thin layer of conducting metal oxide, such as ind ium tin oxide (ITO) or fluorine doped tin oxide (FTO) coated glass as the cathode to collect electron s . The selectivity of the electrode is modified by inserting an n - type material (ETL), such as TiO 2 and ZnO . 27, 41 , 107 The most widely used HTL in perovskite solar cells is 2,2(7,7(tetrakis - (N,N - di - pmethoxyphenylamine)9,9( - spirobifluorene))) ( spiro - OMeTAD ) which is first dissolved in chlorobenzene with additives to improve the conductivity and then spin - coated on to p of the perovskite layer. Other inorganic and organic hole transport materials have also been employed in perovskit e solar cells such as copper thiocyanate (CuSCN), nickel oxide (NiO x ) , P3HT, poly - triarylamine (PTAA). 121 - 124 The disadvantage of the inverted structure is that the 40 ETLs require a high processing temperature, which increase s the complexity and cost of the device fabrication and limits the application on the flexible substrates . In contrast to the inver ted structure, the conventional structured solar cells employ the n - i - p configuration. The most widely used hole transport material is PEDOT:PSS which is spin - coated onto ITO or FTO from water solution and require s a low annealing temperature ( around 100 o C ) . Due to the organic nature of PEDOT:PSS, it can also be coated onto soft substrates in order to fabricate flexible devices. A s for ETL s , C 60 fullerene and its derivatives , like phenyl - C 61 - butyric acid methyl ester ( PC 61 BM ) , are commonly used in the conv entional structure d solar cells. These organic molecules are dissolved in chlorobenzene and then cast on top of perovskite layer. 3.1. 3 Fabrication of polycrystalline perovskit e thin film s Prior to discussing the physical process in perovskit e solar cell s, it is necessary to introduce the fabrication of perovskit e layer s since the film qualit y is significantly impacted by the processing methods. The simplest way to make MAPb X 3 perovskite thin film s is spin - coating the precursor solution , which is composed of lead iodide/chloride (PbI 2 /PbCl 2 ) and methylammonium hal ide (MA X ) with certain ratios in dimethylformamide (DMF)/dimethyl sulfoxide (DMSO) , onto the substrate followed by annealing at 100 o C as shown in Fig ure 3.3 . 28, 31, 125 To prepare MAPbI 3 , the ratio of MAI and PbI 2 is 1.0. T he incorporation of Cl in the perovskite material require s the mixture of MAI and PbCl 2 with a ratio of 3:1. This method works fine for meso - porous structured devices but not the planar structured devices with respect to the morphology and surface coverage of perovski te layer s . It is challenging to form a continuous and pin - hole free perovskit e film on the planar substrates. As mentioned in the previous chapter, it is critical to prepare a compact film without any pinholes in order to achieve high efficienc ies . One way to improve the film quality is adjusting the precursor solution. The influences of the concentration of precursor solution , the 41 types of lead precursor, and the ratio of lea d precursor and organic cation on film quality have been studied by several groups. 125 - 129 One has reported that a higher concentration lead s to a compact thin film and the large r grain size with improved crosslin king of grains, which contribute to the enhanced characteristic parameters of device performance. 128 Meanwhile, it has been proposed that the first step of the crystallization process involves the removal of organic molecules which are the product of the reaction between lead source and organic salt. 126 Thus, lead acetate w as employed, instead of lead iodide, to form methylammonium acetate (MAAc). Since MAAc requires a lower temperature to decompose, the fast nucleation and crystallization are induced during the annealing treat ment resulting in an ultra - smooth thin film. The ratio of lead precursor and organic salt is another factor that impact s the quality of perovskite thin film s . An e xcess amount of PbI 2 influence s not only the grain size but also the electronic structure of perovskit e thin film s . 129 In addition, anti - solvent methods, washing the generic film during spin - coating process to introduce over - saturation, have also been reported to effective ly improve the film quality by inducing the fast nucleation and crystallization. 130 - 131 A similar but simpler way to over - saturate the perovskite layer and induce the homogeneous nucleation is applying a vacuum - fla sh treatment which pumps away the remaining solvent molecules in the film right after spin - coating . 30 While the one - step method s evolve, two - step fabrication , also called sequential deposition, start s to draw more and more attention since it gene rates a compact film on the planar substrates . In this method, t he first step is to deposit a compact PbI 2 layer. The PbI 2 is dissolved in DMF/DMSO and then cast onto the substrate. On top of the compact PbI 2 layer, MAI is subsequently deposited , followed by th e thermal annealing process to allow the inter - diffusion of PbI 2 and MAI to form the perovskit e phase. There are several different ways to deposit the MAI layer. Burschka et al. and Liu et al. dipped the PbI 2 coated sample into MAI contained isopropanol 42 (I PA) solution. 107, 132 Another approach to incorporate MAI into the active layer is spin - coating the MAI solution onto the PbI 2 layer. 133 - 134 In this scheme, the critical pa rt is to find a n orthogonal solvent , which dissolve s MAI but not PbI 2 . The most commonly used candidate is IP A. Due to the low melting point of MAI, Chen et al. developed a vapor - assisted solution process which introduces the organic part into the perovski t e layer by vaporizing MAI. 135 Figure 3. 3 : Illustration of preparing perovskite thin films by one - step method . A few drops of the precursor solution are dripped onto the substr ate, and then the substrate is spun at a high speed. The thin film is annealed at about 100 o C leading to the formation of the perovskite structure. Generally, the challenge s of two - step method s are to full y convert PbI 2 into the perovskite structure and control the grain size . The concerns mentioned above are related to the fast crystallization of PbI 2 which lead s to random ly sized crystallites. 134 I t is difficult for MAI to penetrate into the la rge crystallites due to the edge - on nature of PbI 2 crystals. Therefore, it is critical to control the grain size of the PbI 2 layer . One has discovered that the crystallization process can be effectively controlled through retarding the crystallization of P bI 2 via its strong coordination with solvent molecules, such as DMSO. 134, 136 The solvent molecules interact with Pb by donat ing a pair of electrons, which slow down the crystallization of PbI 2 and stretch the PbI 2 lattice for the incorporation of MAI. The other way to control the grain size is the solvent annealing process. 137 After depositing the MAI layer, the sample is transferred to the hotplate. A petri - dish with a fe w 43 microliters of DMF covers the sample. During the annealing process, DMF molecules are vaporized and fill the petri - dish. The diffusivity of PbI 2 and MAI are enhanced in the solvent atmosphere leading to a large grain size in micron scale and high film qu ality. Figure 3. 4 : Illustration of preparing perovskite thin films by two - step methods . The first step is to spin - coat a layer of PbI 2 . The introduction of the MAI layer can be achieved in different ways, such as (1) immers ing in the MAI/IPA solution, (2) spin - coating the MAI/IPA solution on top of PbI 2 , and (3) thermal evaporation of MAI. As mentioned at the beginning of this chapter, the efficiency of single junction perovskite solar cells has reached over 20%. The thin film preparation process es, which control the film quality, play a critical role in determining the device performance. L arge - area devices with high - quality perovskite films are required in order to commercialize perovskite solar cells. However, the majori ty of the methods mentioned above result in low film quality and increase the complexity 44 of the fabrication process when it comes to the fabrication of large devices. Several techniques have been developed to make large - area perovskite thin film s with high quality so far. 30, 138 - 139 T he device s with the area around 1 cm 2 have achieved an efficiency close to 20%. 30 Kim et al. have reported an even larger device (16 cm 2 ) with an efficiency of 12%. 139 Still, it is imperative to keep boosting the device performance and simplifying the fabrication processes of large - area perovskite photovoltaic devices. Overall, photovoltaic devices employing perovskite materials demonstrate excellent capab ilities for the commercialized solar cells. In the following part of this section, the outstanding optical and electrical properties of perovskit e solar cells, as well as the remaining challenges, will be discussed. Figure 3 . 5 : A bsorption coefficient of ITO/PEDOT:PSS/MAPbI 3 sample . Even though ITO and PEDOT:PSS layers contribute to the calculated value, the high absorption coefficient, which is comparable to that of GaAs, is still dominated by the perovskite material. Inset: an image of ITO/PEDOT:PSS/MAPbI 3 . 45 3.1. 4 Light absorption and carrier generation in perovskite solar cells The perovskite thin film s with a bandgap of ~ 1. 5 eV show a dark brown color. M ost of the high - efficiency perovskite solar cells obtain a 300 - 600 n m thick perovskite film which is able to absorb 90% of the visible light . 32 - 33, 140 One has reported that p olycrystalline MAPbI 3 perovskites demonstrate a sharp absorption onset around 800 nm, a high absorption coe fficient ( around 10 5 cm - 1 ) and the Urbach energy as small as 15 meV . 52, 141 Figure 3.5 depicts the absorption coefficient of ITO/PEDOT:PSS/MAPbI 3 sample. T he high absorption coefficient of perovskite material is re lated to its direct bandgap nature and the stronger p - p orbitals interaction of the conduction band compared with s - p interaction s and indirect bandgap semiconductors. 52 T he unoccupied p - orbitals of lead contribute to the high density of states at the conduction band minimum and the sharp absorption onset. 142 The small Urbach energy indicates a well - order ed structure of perovskite thin film s and a low density of mid - gap states. The high absorption coefficient contributes to not only the high photocurrent but the small V oc loss . With a high light harvesting efficiency , perovskite solar cells require a thin active layer which reduce s the recombination - induced saturated dark current . 140 A fter photons are absorbed by the perovskite layer, excitons are generated. It has been reported that the exciton binding energy is around a few tenths of meV in perovskite materials which is comparable to the thermal energy provided at room temperature, implying that majority of the excitons are spontaneously dissociate d into free carriers upon light absorption. 36, 143 T he efficient free car rier generation under illumination also contributes to the remarkable device performance of perovskite solar cells through avoiding the recombination within the excitons . It is worth mentioning that the co - existence of excitons and free carriers has also b een observed , 46 especially under a high incident photon density, by et al. and Nah et al. , indicating the wide perspective of perovskite materials in different types of optoelectronic devices. 34, 144 3.1 . 5 Carrier transport and collection in perovskite solar cells The transport of holes and electrons is critical in determining the device performance as discussed before. So far, MAPbI 3 and MAPbI 3 - x Cl x have demonstrated superior charge transport properties with ambipolar characteristics , balanced charge transport of electrons and holes . 117, 119 - 120, 145 - 146 One has reported the carrier diffusion length to be over 100 µm for both types of carriers , and the mobilities of electron s and hole s are around 100 and 20 cm 2 V - 1 s - 1 in MAPbI 3 single crystals , measured by time - of - flight (TOF) and space charge limited current (SCLC) . 117 I n polycrystalline perovskite thin films, the carrier diffusion length is over 100 nm for MAPbI 3 and 1 m for MAPbI 3 - x Cl x , 119, 146 and t he carrier mobilities drop to below 20 cm 2 V - 1 s - 1 for both electron s and hole s . 119, 145 The superior charge transport and ambipolar chara cteristic of perovskite materials ha ve been attributed to the similarly small effective mass of electron s and hole s as discover ed by DFT calculations. 118, 147 In addition, the low doping level of perovskite materia ls (weakly p - type) explain s the excellent carrier mobility and diffusion length with the diminished carrier scattering and recombination. 117 It is worth mentioning that the reported carrier mobilities vary in a considerable ra n ge due to not only the utilization of different measurements but also the diverse film qualit y , since t he carrier mobility is potentially influenced by grain size s , crystallinity , grain boundar ies and electronic traps in the perovskite film s . 148 It has been reported that devices with larger grain size s yie ld a higher efficiency due to the improved quality of the film and the redu ced density of grain boundaries (GBs) since GBs act as not only recombination centers but the carrier scattering center s . 149 - 151 Neverthele ss, the concern here is that the increase of grain size is al so accompanied by the change of the texture structure, 47 the degree of preferential crystal orientation. Docampo et al. have found that the crystal orientation plays a more significant role tha n th e grain size in determining the film quality and thus the charge transport property. 43, 54 Besides, o ne has reported that the electron mobility varies along different crystal orientations. 152 Therefore, it is necessary to take grain size s , crystal orientation s and GBs into consideration when investig ating the charge transport in perovskite thin films, and employ techniques with high spatial resolution s in order to distinguish the role of GBs from grain interiors. The charge collection process at the interface between the active layer and ETL/HTL is a nother critical factor that impacts the device performanc e . The carrier injection barrier at the interfaces has been identified in perovskite solar cells. 40, 153 - 154 To effectively collect carriers, it is essential to form an Ohmic contact via the matching energy levels between ETL/HTL and perovskite materials. Additionally, the conductivity of ETL/HTL materials influences FF and thus the device performance by contributing to R s . One has reported that the charge inj ection from perovskite materials to spiro - OMeTAD is more efficient than to polymeric HTLs due to the matching energy levels and the high carrier mobilit y in the former. 155 The insertion of an interface layer between TiO 2 and MAPbI 3 and doping o f ETL are also employed to enhance the charge collection by tuning the injection barrier and the conductivity of ETL . 156 - 157 At the surface of perovskite materials, there also exist defect states due to the broken symmetry and the dangling bonds. The trapping of electron s and hole s at the surface would result in recombination , which hamper s the cha rge collection process . 158 - 159 More details about surface recombination and the passivation of surface defects will be discussed in next session. 3.1.6 Non - radiative r ecombination in perovskite solar cells In this section, the discussion focus es on the non - radiative recombination in perovskite solar cells. Typically, non - radiative recombination happens via trap states within the bandgap. The 48 trap states, induced by the crystal structure defects, are able to trap a nd de - trap electron/hole. I f the carrier cannot be de - trapped from the mid - gap state before encountering an opposite charge, recombination happen s . Thus, the density of trap states, trap/de - trap kinetics and carrier concentrations impact the non - radiative recombination process. The structural defects exist in the bulk of the perovskite thin film s , as well as at the sur face and grain boundaries. As mentioned before, the small Urbach energy (~15 meV) indicates a well - ordered structure and the low density of mid - gap states in the polycrystalline perovskite thin films. It has been reported that the polycrystalline perovskite thin film demonstrates a bulk trap density around 10 15 - 10 16 cm - 3 which is comparable to the solution - processed CIGS films . 51, 160 Several theoretical studies have revealed the unusual defect physics in perovskite thin films. One has discovered that the defects generate shallow levels close to CB or VB. 52, 16 1 - 162 Consequently, these defect states act as unintentional doping sources instead of recombination centers. In addition , it has been pointed out that the defects in MAPbI 3 that create deep levels require high formation energy and barely exist in low - tem perature processed perovskite films . 162 The low density of trap state s in perovskite thin films is also ascribed to the low crystallization activation energy (less than 100 kJ mol - 1 ) compared with that of amorphous silicon, which is as high as 400 kJ mol - 1 . 127 Though the polycrystalline perovskite thin films show a low density of trap state s , it has been suggested that the trap - assisted recombination in the bulk perovskite still i mpact s the power output of devices. 51, 140, 163 - 164 By comparing with single crystals, it is evident that polycrystalline perovskite thin films demonstrate a short diffusion length, a low carrier mobility and a hig h defect density . And these characteristics collectively impact to the device performance of perovskite solar cells . Besides the point defects, the defects located at grain boundaries also impact the density of trap states in the bulk perovskite thin film s. Different from point defects, GBs possess more 49 complicated structural defects, 2 - dimensional planar defects. It is likely that there are various stoichiometries at GBs. Guo et al. have carried out DFT calculation s to investigate the electronic structure of GBs in MAPbI 3 thin film s . 53 The same conclusion as point defects has been made th at 2D planar defects at GBs do not generate deep level electronic states. Experimentally, the non - detrimental , even beneficial, role of grain boundaries has been observed by employing the techniques with spatial resolution s such as AFM, microwave impedanc e microscopy (MIM) and KPFM. 48, 165 - 167 However, de Quilettes et al. and Mamun et al. have employed confocal and wide - field fluorescence microscopy and discovered the lower PL intensity and faster PL decay at GBs c ompared to grain interiors in perovskite thin films , suggesting the GB - induced non - radiative recombination . 51, 168 - 169 Meanwhile, one has found the heterogeneous behavior s of GBs: some GBs ateral carrier transport while others possess an energy barrier for charge transport across grains. 49, 170 More interestingly, MacDonald et al. have found the depth - dependent electrical behaviors of GBs due to the faster degradation of MAPbI 3 close to the top surface. 170 Though most of the techniques mentioned above provide a high resolution, the characterization of the power output of the microstructures in perovs kite thin films , which provide s direct evidence on their contribution s to the device performance, ha s not been performed due to the challenges brought by the roughness of perovskite thin film s and the stacking of grains and GBs. Contrary to the controver sial role of GBs, non - radiative recombination at the interface has been criticized as the dominant carrier loss mechanism in perovskite solar cells. T he broken symmetry at the perovskite surface introduce s trap states and thus result in severe recombinatio n. Trap - mediated recombination has been identified in both perovskite/ETL and perovskite/HTL interfaces. A large portion of recombination occurs at the interface of perovskite/PEDOT:PSS in n - i - p planar structured devices . T he adoption of nickel oxide nanop - bis(4 - 50 butylphenyl) - - bis(phenyl) - benzidine (poly - TPD) enhance s the passivation of surface states and thus reduces the V oc loss . 171 - 172 Also, Correa - Baena et al. discovered that in p - i - n s tructured perovskite solar cells the r ecombination dynamics are strongly correlated with the dopant concentration in spiro - OMeTAD layer , and a V oc as high as 1.22 V has been achieved by controlling the amount of dop ants . 173 T he passivation of perovskite/ETL interface s has been accomplished by employing tin oxide (SnO 2 ), chlorine - capped TiO 2 , magnesium oxide ( MgO ) coated ZnO nanoparticles as ETL, or inserting an ultrathin polymer - fullerene film. 173 - 176 Similarly, Yang et al. have inserted an organic self - assemble d mon olayer (Si - bearing 3 - aminopropyltriethoxysilane) to introduce an interfacial dipole, which boost s the interaction of perovskite materials and ETL and tune s the band energy alignment to enhance the charge extraction. 177 In addition, due to the interfacial charge transfer anisotropy of perovskite films, tuning the texture structure and preferential crystal orientation improve s the charge collection and su ppress es the recombination at the interface. Overall, the passivation of interface recombination is especially beneficial to reduce the V oc loss and improve the efficiency of perovskite solar cells. 3.1. 7 Remaining c hallenges in perovskit e solar cells T hough the device performance of perovskite solar cells has reached over 20 %, there are still some challenges in the commercialization process, such as the J - V hysteresis characteristics and device stability. The hysteresis characteristics of perovskite s olar cells refer to the discrepancy of J - V curves when sweeping the voltage in the forward and reverse diractions . 178 Therefore, the existence of the hysterical phenome non poses challenges in accurately determining the device performance of the devices. The hysteresis behavior stems from the devices themselves and is impacted by the 51 scanning conditions. From the inside, the hysteresis behavior depends on the perovskite m aterials and the selective transport layers. 172, 179 - 182 It has been discovered that inverted structured perovskite solar cells demonstrate a negligible hysteresis due to the passivation effect of PCBM. 183 O n the outside, the scanning conditions such as the scan rate , the external electric field, and the pre - conditioning of devices also impact the hysteresis of J - V curves. 184 - 186 Since the discover y of the hysteresis behavior in perovskite solar cells, many mechanisms have been proposed , such as the ferroelectricity of MAPbI 3 , the trapping/de - trapping process, and ion migration s . 179 - 180, 187 - 189 S everal studies have revealed that ion migration is the dominant mechanism of the hysteresis behavior in perovskite solar cells. 179 - 180, 186, 190 The ion migration is driven by the b uilt - in and/or external electric field , result ing in the ionic accumulation at the interface . Consequently, the built - in electric field in the bulk perovskite is enhanced or diminished depending on the specie s of the accumulated ions and the interface. Whe n sweeping the voltage in different directions, even under the same bias, the effective electric field in the device would be different due to the slow response of the ion migration, leading to the mismatch of the photo - respons e . The time - scale of the ion i c m ovement also matches with the observed hysteresis response to the scan rate in perovskite solar cells . Figure 3.6 describes the ion migration process due to the built - in electric field and its impact on the distribution of the electric field. The accumu lated ions at the interfaces form an electric field in the opposite direction to the built - in electric field, which diminishes the intensity of the effective electric fi e ld in the bulk perovskite layer . As discussed before, the electric field influences th e separation of photogenerated electrons and holes. Hence, the diminished field cause s a less efficient charge separation in perovskite solar cells . The ions at the interface modify the charge extraction barrier as well , as depicted in Fig ure 3.6. The spec ies of ion s migrati ng in the perovskite material are likely to be I - 52 and MA + ions due to the low activation energy of migration comparing with that of Pb 2+ ion, as calculated by Meloni et al. and Eamea et al . 180, 19 0 Experimentally, the concentration contrast of MA + ions before and after poling the device has been observed by Yuan et al . , indicating the migration of MA + under the external electric field . 191 Also, Li et al. have observed the I - migration and characterized the migration kinetics in perovskite solar cells with the aid of wide - field PL microscopy. 192 Figure 3 . 6 : Schematic of the ion migration in the perovskite material . Illustration of (a) t he ex istence of mobile ions in perovskite layer s and their moving directions (negative ions: blue arrow, positive ions: red arrow) under the built - in electric field when sandwiched between the anode and the cathode with different work functions; (b) the accumulation of ions at the interface and t he modified (diminished) built - in electric field in the bulk of perovskite layer . In order to suppress the ion migration, it is necessary to understand ion migrat ion paths in the perovskite photovoltaic devices. According to DFT study , the ions migrate al ong the defects in the perovskite layer by hopping between neighbor sites. 180, 190 Additionally , Shao et al. and Yun et al . have reported a faster ion migration along GBs and attributed it to the high density of de fects with the aid of c - AFM and KPFM techniques. 193 - 194 Therefore, it is important to improve the 53 crystallinity , reduce the concentration of mobile ions , and grow large grains to decrease the density of mobile ions and grain boundaries in perovskite thin films. 183, 195 - 196 Likewise, it is important to passivate the perovskite surface to suppress the ion migration . 172, 197 The device stability is the other challenge that perovskite solar cells are facing on the way to the market of photovoltaic devices. It is well - know n that MAPbI 3 is sensitive to moisture and heat. 198 - 201 The moisture in the air accelerates the decomposition of perovskite materials into HI, MA and PbI 2 . The encapsulation of perovskite solar cells has successfully protect ed the devices from the humid environment and improve d the device stability. 202 - 203 The other external factor that impacts the stability is the heat induced by illumination under working condition. The low thermal stability lead s to the decomposition of MAPbI 3 under constant illumination. In addition, as discussed above , a transition from the tetragonal to the cubic phase of MAPbI 3 happens around 50 o C which is with in the range of the operation temperature for photovoltaic devices. T he organic part (MA) in the perovskite materials has been accused of hampering the therma l stability. Several studies have demonstrated that by introducing organic or even inorganic cations, such as FA and Cs , the stability of the corresponding perovskite materials is greatly improved as well as the phase transition temperature. 30, 204 - 205 Recently, one has reported the migration of the electrode metals , such as Au, to the active layer through small molecule HTL s under thermal stress, leading to the decomposition of MAPbI 3 . 173, 206 To suppress the migration of Au, n ovel and stable hole transport materials have been employed to replace spiro - OMeTAD and the thermal stability is improved. 205, 207 I n order to compete with s ilicon solar cells, perovskite solar cells have to demonstrate a degradation rate of 0.5 % per year for up to 20 years and comparable or even lower costs. 200 54 Accordingly, more efforts are needed to make high - quality perovskite materials with excellent stability, and cost - effective ETL/HTLs in order to commercialize perovskite photovoltaic devices. 3.2 Organic photovoltaics and the applicatio n of ZnO as a buffer layer T his part focuses on the application of ZnO nanoparticles in organic photovoltaics. Before talking about ZnO, it is necessary to briefly talk about organic photovoltaics. OPVs demonstrate a low cost, facile manufacturing , and th e mechanical flexibility. 57 - 58 Majority of the OPVs are using polymers as light absorbers which offer a high absorption coefficient (usually 10 5 cm - 1 ). T he chemical flexibility of the modifications facilitates the tuning of optoelectrical properties in organic semiconductors. Here, the P3 HT:PCBM blend , one of the most widely studied active layer in OPVs, is used as an example. Figure 3 . 7 : Schematics of photocarrier separation, collection and recombination in OPVs . Excitons generated close to the donor/acceptor interface are efficiently dissociated into free carriers. The electrons/holes then travel within the ac ceptor/donor domains to the cathode/anode. As illustrated above, the isolated domains confine the free carriers and eventually they recombine at the donor/acceptor interface. 55 P3HT is a p - type polymer based on polythiophene , demonstrating an extended deloc alized - electron system with the bandgap around 2. 0 eV . PCBM is a n n - type small molecule based on C60 . 106 The general physical processes in OPVs are illustrated in Fig ure 3. 7 . Compared with perovskite materials, the dielectric constants of P3HT and other organic semiconductors are small. Consequently, upon illumination, the excitons generated in P3HT show a high binding energy (0. 3 eV - 0.7 eV) and thus cannot be spontaneously di ssociated at room temperature. 208 - 211 One way to efficiently separate the photogenerated electron and hole pairs is to introduce a potential drop, for example , the P3HT and PCBM interface as discussed in Chapter 2. At the heterojunction of P3HT:PCBM, electron s and hole s are separated via the transfer of electrons to PCBM. Accordingly, P3HT and PCBM are called the donor and the acceptor, respectively. There are many ways to create the donor - acceptor interface . For ex ample, it can be achieved in a layered structure: one layer of P3HT followed by one layer of PCBM. However, the concern is that the excitons must diffuse to the heterojunction and then be dissociated. T he average distance that excitons travel before recomb ination , also called exciton diffusion length, is usually less than 20 nm in organic materials . 212 Therefore, it requires the thickness of P3HT layer to be around the same range to efficiently dissociate excitons, which is obviously not thick enough to make the most out of the incident photons. The bulk heterojunction (BHJ) configuration solves the rivalry of the exciton diffusion length and the film thickness by intimately mixing the donor and accept o r. In order to form the BHJ, P3HT and PCBM are dissolved together in o rganic solvents, such as chlorobenzene, followed by spin - coating onto the substrate. As illustrated in Fig ure 3 .7 , donor s and accept o r s form segregated domains in the active layer. Under illumination, excitons are generated in the donor, and then they diff use to the interface and g e t separated. In this scenario , as long as the domain size is within the same range of the exciton diffusion length, excitons can be efficiently dissociated at 56 the donor/acceptor interface. The challenges for this structure are co ntrolling the domain size and forming the percolated path for both donor s and acceptor s . If the domain size is much larger than the carrier diffusion length, the excitons that are generated deep with in the domains recombine within the acceptor before reach ing the interface. W ithout the percolated path, the free carriers dissociated at the interface would be spatially confined inside the domain and cannot be collected by the electrodes, as illustrated in Fig ure 3 .7 . After the exciton dissociation process, f ree electrons and holes are transported within donor s and acceptor s separately. The carrier mobilities of organic semiconductors are generally low, ranging from 10 - 5 to 1 cm 2 V - 1 s - 1 , which limit the thickness of the active layer and pose certain requireme nt s on the length of the percolated pathways. 213 For example, if th e percolated path of P3HT is longer than the hole diffusion length , it increase s the possibility in the recombination of separated electrons and holes at P3HT/PCBM interface. The interfacial layers play a significant role in determining the device perform ance of OPVs . First, interfacial layers act as selective contacts. As depicted in Fig ure 3. 7 , the P3HT:PCBM blend would directly contact the electrodes without the assistance of interfacial layers. E lectrons in PCBM and holes collected by the anode are lik ely to recombine at PCBM/anode contact. A similar situation holds for the P3HT/cathode contact. Therefore, an interfacial layer is necessary to form a selective contact and suppress the recombination at the contact . In OPVs, the commonly used electron sele ctive materials include ZnO, TiO 2 , fullerene - based molecules. And p - type materials, such as PEDOT:PSS, molybdenum oxide (MoO 3 ) and NiO, are employed as the interfacial layer at the anode/active layer interface. Second, the interfacial layers optimize the c ontact by tuning the working function of the electrode and reducing the injection barrier. From the perspective of energetics , in OPVs, t he built - in electric field , which depends on the work 57 function difference of the anode and cathode materials , not only assist s the dissociat ion of excitons but facilitate s the charge transport. T he existence of injection barrier s at the active layer/electrode interface s ha s been observed and attributed to the energy level mis - alignment. 214 - 216 By inserting an interfacial layer , it has been demonstrated to effectively tune the work function of electrodes and thus optimize the contact , resulting in high device performance . 59, 62, 106, 217 - 220 An alternative way to tune the work function is via the formation of a dipole layer at the contact. Figure 3. 8 depicts the impact of the interfacial dipole on the work function of the electrode. When an interfacial dipole points out of the electrode, it re duces the work function and vice versa. In addition, materials, like bathocuproine ( BCP ) and MoO 3 , with bandgaps larger than that of the donor and the acceptor act as the exciton blocking material since it requires extra energy to transfer the exciton to t he interfacial layer. 221 - 224 ZnO is a direct bandgap material with the bandgap around 3.4 eV. 225 The CB M and VB M of ZnO are 4.3 and 7.7 eV , respectively. 225 ZnO frequently exhibits the n - type conductivity due to the existence of intrinsic defects in the material. 226 - 227 The intrinsic defects include zinc vacancy/interstitial, oxygen vacancy/interstitial, and antisites of Zn and oxygen. One has found that the zinc vacancies , which result in deep acceptor states with low formation energ y under n - type conditions , are likely to be the compensating defects in n - type samples. And t he deep VBM enables ZnO acting as an effective hole blocking material . In addition , the application of ZnO interfacial layer in OPVs has been proved to improve the stability of the devices. 85, 228 - 229 However, the control over the electrical conductivity of ZnO is still challenging, which hinders the application of ZnO in electronic devices. 81 In order to tune the conductivity, it is critical to control the intrinsic and extrinsic defects, which is still a major issue for ZnO materials. 58 Figure 3. 8 : Schematics of work function tuning via the formation of interface dipole . The interface dipole ( ) pointing out of the surface induces a vacuum level shift and thus reduces the work function of ITO. 59 Chapter 4 Device fabrication and e xperimental m ethods To character ize the photovoltaic devices and investigate the dynamic processes , many techniques have been employed in this study. I t is necessary to understand the principles of the experimental instruments in order to interpret the results. In this chapter, first, th e detailed procedures of sample preparation are described . Different preparation methods have been developed and resulted in different film qualit y , especially for perovskite thin films. Hence, it is important to know the fabrication process of the devices . Second , t he experimental conditions and instrument settings are illustrated for cross - referencing the experimental results between different studies . Particularly, the working principle s of AFM, EIS and EBSD measurements are introduced in the last sessio n for a better understanding of the experimental results. 4.1 Device fabrication 4.1.1 Fabrication of perovskite solar cells In this study, the conventional planar structured ( n - i - p ) perovskite solar cells were fabricated and investigated. In this archi tecture, the MAPbI 3 layer i s sandwiched between ITO/PEDOT:PSS (a n ode) and PCBM/BCP/Ag ( cathode ). The fabrication process involves multiple deposition techniques , such as solution process es and physical vapor deposition s . 4.1. 1.1 Anode preparation The ano de preparation was done in the cleanroom. ITO/Glass substrates from Xinyan Technology Ltd. were pre - cleaned in alconox, deionized (DI) water, acetone and IPA, respectively, with sonication for 10 min. Then, the substrates were blown dry with nitrogen. In o rder to tune the work function of ITO and remove the organic molecules adsorbed on the surface, the substrates 60 were transferred to the vacuum chamber and processed with oxygen plasma for 5 min at 150 W (PX - 250, March Instruments). The spin - coating of PEDOT :PSS layer was done in the air. PEDOT:PSS solution (Heraeus Technology Group) was filtered and sonicated for 30 min before used for spin - coating. The solution was kept in the fridge (around 5 o C) when not in use. 50 uL of PEDOT:PSS solution was dripped on top of ITO/glass and then spin - coated at 6000 RPM for 20s. A highly transparent film with light blue color was formed on the substrate. Right after spin - coating, the substrate was transferred onto a hotplate which was pre - heated to 150 o C. After 5 min, the residual water from the solution was completely removed from PEDOT:PSS film s . The samples were immediately transferred into the glovebox where the water and oxygen levels were below 0.1 ppm for the next step: the deposition of the perovskite layer. 4.1.1. 2 Preparation of MAPbI 3 thin film In this study, both one - step and two - step methods were employed to fabricate the pure MAPbI 3 thin film of different thickness and grain sizes. The formation of the perovskite phase is through combining the lead source an d CH 3 NH 3 I with the aid of thermal annealing. The perovskite layers were prepared in the glovebox due to its sensitivity to the ambient condition. Details of the fabrication process are described below. To synthesize CH 3 NH 3 I, 10 mL of hydroiodic acid (57wt% in water, Sigma) was added to 24 mL of methylamine (33 wt.% in ethanol, Sigma) dropwise into a round - bottom flask which was immersed in ice/water bath (0 o C). After completing the adding of acid, the solution turned dark brown and was stirred for another 2 hr at 0 o C, Then, the solution was transferred to a rotary evaporator in a water bath at 50 o C. The water and ethanol were gradually removed from the solution leaving the yellow precipitate in the flask. Next, the raw product was dissolved in the 61 minimum amount of ethanol and then precipitated by adding diethyl ether. The precipitate with light yellow color was then filtered. The recrystallization process was repeated two more times to purify the CH 3 NH 3 I product into white crystallites. Figure 4. 1 : Images of the spin - coater and the hotplate ensembled in the glovebox . In one - step method, both lead source and MAI were dissolved in DMF and/or DMSO. To make MAPbI 3 , 2.88 g of PbI 2 (Sigma) and 0.636 g of CH 3 NH 3 I were dissolved i n 5 mL mixture of dimethylformamide (DMF) and dimethyl sulfoxide (DMSO) and stirred at 60 °C for 12 hr in the glovebox. The solids gradually dissolved, forming a light - yellow solution. The precursor solution prepared above was then spun onto the PEDOT:PSS /ITO substrate at 1000 RPM for 15 s and 5000 rpm for another 30 s. A light - yellow film was formed on the substrate. The samples were then annealed at 100 °C for 10 min, which w as referred as fast annealed samples. The films turned into dark brown after hea ted for 5 s. As to ramp annealing process, the samples were first annealed at a lower temperature T 0 for 10 min, and then the temperature was ramped up to 100 °C at different ramp rates, followed by 10 min annealing at 100 °C. The color of the film gradual ly turned into dark brown and the film became highly 62 reflective. The thickness of MAPbI 3 thin films was tuned by adjusting the concentration of the precursor solution as listed in Table 4.1. Table 4 . 1 : Precursor concentration and film thickness . Solution Concentration mol L - 1 Solution temperature °C Film thickness nm 0.63 25 110 1.25 25 200 1.87 25 290 2.50 70 450 In Chapter 6, the samples were prepared by two - step method followed by the solvent annealing process in orde r to grow columnar - structured grains. First, 230 mg of PbI 2 and 15 mg of MAI were dissolved in 1 mL of DMF and MAI, respectively. The solutions were stirred at room temperature in the glovebox for 2 hr. The as - prepared PbI 2 solution was spun onto PEDOT:PSS /ITO substrate at 4000 RPM for 30 s. Then, MAI was spun on top of PbI 2 at 4000 RPM for 20 s. A smooth film with light brown color was formed immediately after spin - coating. The samples were immediately transferred to a hotplate which was preheated at 100 ° C. A petri - during the annealing (30 min). The resulting film thickness was around 110 nm as measured by AFM. In order to increase the film thickness to 300 nm and c haracterize the film quality, the concentration of PbI 2 and MAI were increased to 500 mg mL - 1 and 40 mg mL - 1 , respectively, with the annealing duration extended to one hour. To prepare the sample for EBSD measurement, the same procedure was adopted while u sing a higher concentration of PbI 2 (1000 mg mL - 1 , 100 °C) and MAI (40 mg mL - 1 , 70 °C). And the solvent annealing condition was adjusted to 2 hr wit h the - filled glovebox. 63 4.1.1.3 Cathode preparation The preparation of cathode involves the solution process and the thermal evaporation. PCBM (2 wt% in chlorobenzene) was spun onto the pe rovskite film at 1500 RMP for 30 s. The resulting samples were transferred to a vacuum deposition chamber (Angstrom Engineering). First, 5 nm of BCP was thermally evaporated and deposited on top of perovskite at the rate of 0.5 Å s - 1 . The sample stage was rotating during the deposition to guarantee the full coverage of BCP. Then, Ag (100 nm) was thermally evaporated at increasing rates (0.5, 1.0 and 2.0 Å s - 1 ) with a patterned shadow mask covering the samples. The stage was still to avoid the shadow effect which could potentially enlarge the device area. The deposition of BCP and Ag was under a pressure of 3.0 10 - 6 Torr. Figure 4 . 2 : Images of the thermal evaporator used for deposition of MoO 3 , C60, BCP and Ag layers . 64 4.1. 2 Fabrication of organic photovoltaics To make ZnO thin film , z inc acetate solution (0.1 mol L - 1 ) was prepared in 2 - methoxyethanol (MXL) and monoethanolamine (MEA) with a molar ratio of 1:1 between zinc acetate dihydrate and MEA. 230 All materials were purchased from Sigma - Aldrich. The addition of MEA is to increase the solubility of zinc acetate in MXL by th e chelation between - OH/ - NH 2 group and zinc. The solution was spin - coated at 4000 RPM for 40 s onto ITO/Glass substrates (Xinyan Technology Ltd.) which were pre - cleaned with sonication in acetone and isopropanol. The samples were then annealed for 10 min a t different temperatures to decompose the acetate precursor and crystallize the ZnO film. These fi lms were subsequently rinsed in DI water and ethanol, followed by annealing at 150 °C for 10 min to dry the sample. 90 ZnO fi lm preparation was followed by spin - coating a P 3HT and PCBM blend ( 1: 0.8 in weight) in chlorobenzene (27 mg ml - 1 in total) at 800 RPM for 30 s. P3HT and PCBM were purchased from Rieke Metal and American Dye Sources Inc., respectively. The samples were then heated to 105 °C at a rate of 20 °C min - 1 and annealed at this temperature for 20 min, followed by cooling at a ramping rate of 10 °C min - 1 to room temperature. The entire process was carried out in a glove box whe re the oxygen and water impurity levels were below 0.1 ppm. The resulting samples were transferred to a vacuum deposition chamber (Angstrom Engineering), where 10 nm MoO 3 and 100 nm Ag were thermally deposited with shadow masks under a deposition pressure of 3.0 10 - 6 Torr. 4.2 Experimental conditions of characterizing tools 4.2.1 J - V and EQE measurements Solar cell efficiencies were characterized in air using a Keithley 2420 source - meter and a Newport solar simulator under 100 mW cm - 2 illumination measure d with an NREL calibrated mc - 65 Si detector with KG5 filter. Filters were applied to adjust the light intensities from 0.0092 to 1 sun for the J - V measurements at varied light intensities. The scan rate of the J - V characteristics of perovskite solar cells was 50 mV s - 1 . The EQE measurements were carried out on a setup comprising a Xe lamp, a monochromator, a current - voltage preamplifier, and a lock - in amplifier. The light spectrum was determined with a monocrystalline photodetector calibrated by the National I nstitute of Standard and Technology (NIST) . Devices were corrected for the spectral mismatch (M) with values of approximately 0.98 0.2 sun). 164 The diode quality factors further accounts for the larger FF which is consistently measured for the ramp - annealed compared to the fast annealed - samples. ` Figure 5. 10 : Plots of resistances and capacitances at the short - circuit condition versus the light intensity . Dependence of (a) resistances and (b) capacitances on the light intensity at the short - circuit condition. The measurements are taken on samples with the perovskite layer thickness of 110 nm. The J sc is found to increase linearly with light intensity for both samples (Fig ure 5. 9 b). The interfacial recombination time constants are also determined at short circuit, where s,s c is found to be larger than s,o c for both samples at all light intensities (Fig ure 5. 8 c). Unlike at open circuit, however, the time constants of the ramp - annealed and fast annealed samples are fairly close at short circuit (see Figure 5.10 for resist ances and capacitances at varying intensity under the short circuit) . This suggests that interfacial recombination may not account for charge collection losses. 105 It was suggested that the collection efficiency can be calculated from the ratio of the recombi nation resistance ( R 1 + R 2 ) at the open circuit and the short circuit according to cc = 1 - R oc / R sc . 244 This analysis produces nominally identical values of cc = 0.985 for the r amp - annealed samples and cc = 0.978 for the fast - anneal ed samples. Thus, differences in charge collection efficiencies cannot completely account for the discrepancy in J sc observed. Instead, if this photocurrent discrepancy mainly originates from the bulk diffusion / recombination processes through differences in the carrier diffusion length, it should be reflected in the device performance as a function of the perovskite layer thickness. 19, 119, 146 Table 5 . 2 : C haracteristic parameters of perovskite solar cells with different thickness . T he devices are constructed of ITO/PEDOT:PSS/CH 3 NH 3 PbI 3 /PCBM/BCP/Ag with the varying thickness of the perovskite layer processed at the fast - or ramp - annealing conditions. The optimized T 0 and ramping rate a re adapted in the latter. A nnealing p rocess film thickness nm J sc mA cm - 2 V oc V FF PCE % Fast 110 10.6±1.1 0.84±0.02 0.58±0.03 5.1±0.5 Fast 200 15.0±1.5 0.90±0.01 0.52±0.01 7.0±0.7 Fast 290 8.9±0.9 0.88±0.01 0.70±0.01 5.4±0.5 Fast 450 8.8±0.9 0.85±0.0 1 0.67±0.02 5.0±0.5 Ramp 110 13.7±1.4 0.89±0.01 0.66±0.01 8.0±0.8 Ramp 200 17.2±1.7 0.87±0.01 0.65±0.01 9.7±1.0 Ramp 290 22.0±2.2 0.83±0.01 0.66±0.01 12.1±1.2 Ramp 450 16.5±1.7 0.88±0.01 0.67±0.01 9.7±1.0 106 Figure 5 . 11 : P lot of J sc and PCE dependence on the perovskite layer thickness . Figure 5. 12 : Plots of capacitances, resistances and recombination time constants of devices with thick perovskite films at the short - circuit condition versu s the light intensity . Dependence of capacitances (a) and resistances (b) on light intensity under the short - circuit condition for both ramp - and fast - annealed devices with the 290 nm thick perovskite layer . (c) Calculated interface recombination time cons tants under the short and open circuit conditions at different light intensities . 107 By varying the concentration of the precursor solution, film thickness can be controlled between 1 1 0 nm and 450 nm as listed in Table 5.2 . Indeed, t he highest PCE for the fa st - and ramp - annealed samples is achieved at the thickness of 200 nm (7.0 %) and 290 nm (12.1%), respectively, as illustrated in Fig ure 5.11, suggesting a longer carrier diffusion length associated with the latter . Because the J sc and PCE variations betwee n the ramp - and fast - annealed samples are most pronounced at the perovskite layer thickness of 290 nm, corresponding EIS measurements are performed at this thickness which might provide more information on the recombination parameters under the short - circu it condition. The associated capacitances and resistances, as well as the derived recombination time constants are plotted in Fig ure 5 .12 . In contrast to the comparable s,sc in thin devices, s,sc of the 290 nm thick ramp - and fast - annealed samples differ s by 2 order s of magnitude, as shown in Fig ure 5 .1 2 c, which originates from the larger R 1 and R 2 in the former (Fig ure 5 .1 2 b). I t has been argued that the applied bias in EIS measurements can lead to ionic transport and polarization of the interfaces in de vices with the regular structure , which consequently interferes with the EIS measurements and changes the mechanism of recombination from bulk - to surface - dominant. 247 The same ionic processes will al so lead to a gradual increase of R s (defined in Fig ure 5. 7 b ) over time and introduce hysteresis in J - V curves. 247 - 248 Nevertheless, as shown in Fig ures 5 . 4 and 5. 7b , no obvious J - V hysteresis or change in R s is obs erved in our experiments, indicating that the ion migration/polarization issue is not as severe in our inverted - structured devices. This can be attributed to the surface passivation effect of fullerene molecules , 183 as well as the avoidance of chemical reactions as those occur at the interfaces between perovskite and TiO 2 /Spiro - OMeTAD in devices with the regular structure. 247 Therefor e, the larger R 1 and R 2 as measured in the ramp - annealed sample under the short - circuit condition (Fig ure 5 .1 2 b) provide a strong indication that beyond the surface recombination at contacts , bulk recombination process 108 is also suppressed in the ramp - anneal ed devices, which ultimately leads to the longer carrier diffusion length , as illustrated in Fig ure 5.1 1 . Note that similar to the thin devices , significant variation in s,oc , mainly contributed by the modulation in C 1 , that is , interface capacitance, is also observed between the ramp - and fast - annealed 290 nm thick samples, as illustrated in Fig ure 5.1 3 . Figure 5 . 13 : Plots of capacitance s and resistance s of devices with thick perovskite films at the open - circuit condition versus the light intensity . Dependence of resistances (a) and capacitances (b) on light intensity under the open - circuit condition measured on samples with the 290 nm th ick perovskite layer, the trend of which is similar to that of the devices with 110 nm thick perovskite. Although there are theoretical predictions that defects in the perovskite layer mainly contribute to the shallow traps and thus are not detrimental to the device performance, 52, 161, 249 our study suggests that it is still crucial to improve the crystallinity and texture of the perovskite layer to boost the device performance. On the one hand, the recombination a t the contacts is expected to be the dominant loss mechanism in perovskite solar cells. 38, 172, 250 Instead of introducing interfacial layer at the contact 172, 250 - 251 or p assivating the surface trap states with small molecules 158 - 159 which could complicate the device fabrication process, our study demonstrates 109 that the recombination kinetics at the interfaces can be effectively supp ressed simply by the ramp - annealing treatment via controlling the surface orientations or terminations of perovskite grains. On the other hand, it is likely that the preferential crystal orientation in the ramp - annealed sample yields an increase in the den sity of low - angle grain boundaries in the polycrystalline perovskite thin film, which, as compared to the large - angle grain boundaries, exhibit better carrier transport properties with minimized bulk carrier recombination. 5.4 Conclusion In this chapter , it is demonstrate d that the texture of perovskite thin film influences both the surface recombination at the contacts and the carrier diffusion length in the bulk . The combination of the two effects lead s to enhanced performance in devices constructed of p referentially oriented perovskite thin films. These findings could aid in the simple design and fabrication of planar - structured high - efficiency perovskite solar cells. However, it is still not clear how the texture structure enhances the charge transport and suppresses the surface recombination. As mentioned above, one possibility is that the low - angle grain boundaries are beneficial in perovskite films compared with high - angle grain boundaries. Also, the carrier mobility along different grain orientation could be different. Since CBM of perovskite mainly consists of p - orbitals of lead and iodide, different crystal orientation would impact the distribution of electro n cloud and therefore the charge transport . Therefore, it is critical to investigate the cor relation between crystal orientation and carrier mobility as well as the role of grain boundaries in perovskite thin films, which is the focus of next chapter. 110 Chapter 6 Crystalline o rientation d ependent p hotoresponse and h eterogeneous b ehaviors of g rai n b oundaries in p erovskite s olar c ells In this chapter , photoconductive atomic force microscopy and Kelvin probe force microscopy are employed to study the photoresponse of microstructures in perovskite thin films. The discrete photocurrent levels across c rystalline grains along with the anti - correlated behavior between the local J sc and V oc are identified in perovskite thin films for the first time, revealing an orientation - dependent transport. Additionally, t he photoelectrical properties of low - angle grai n boundaries from that of large - angle boundaries are distinguished , with the former even displaying higher J sc and V oc than adjacent grain interiors. It is worth mentioning that t he high - resolution photocurrent mapping and diode - shaped point J - V curves es tablished in this study allow the extraction of local device parameters, thus providing new insights into the correlation between microstructures of the film and properties/performance of the device. Unraveling such correlation will aid the fabrication of high - efficiency hybrid perovskite solar cells. 6.1 Photocurrent mapping of perovskite thin film s with columnar structure s Figure 6. 1a , b show the top view and cross - sectional view SEM images, respectively, of the CH 3 NH 3 PbI 3 /PEDOT:PSS/ITO sample prepared w ith sequential deposition followed by solvent annealing. 137 One - step spin - coating induces small perovskite grains due to the fast crystallization of perovskite films. Compared with one - step method, the sequential d eposition retards the crystallization process, which facilitates the mass transport and the growth of perovskite grains . The DMF vapors during the solvent annealing process further promote the ion diffusivity and contribute to the improved grain sizes. The refore, f ilms processed under such condition are constructed of single grains along the perpendicular direction with the lateral grain size varying 111 between 100 and 500 nm. For studies aiming at unraveling the photoresponse of individual grain or GB, such c olumn - structured thin film is a prerequisite which ensures that the photocurrent measured on the top surface is not convoluted by the stacking of grains or GBs underneath. Additionally, instead of a thicker film (300 nm) as in the device configuration (see Fig ure 6.1c), a thinner film (110 nm) is employed in the pc - AFM studies as the roughness of the film increases with the thickness which can cause artifacts in photocurrent mapping. Figure 6. 1 : SEM images of perovskite th in films and J - V characteristics of perovskite devices . (a) Top view and (b) c ross - sectional SEM images of CH 3 NH 3 PbI 3 /PEDOT:PSS/ITO sample; (c) J - V characteristics of devices with the structure of ITO/PEDOT:PSS/CH 3 NH 3 PbI 3 (~300nm)/C60/ BCP/Ag, yielding a d evice performance of 13.0 % under 1 sun illumination. 112 Figure 6. 2 : Photocurrent map and line profile of perovskite thin films . Simultaneously obtained (a) topology and (b) photocurrent map of CH 3 NH 3 PbI 3 /PEDOT:PSS/ITO under the illumination of a green laser while applying a - 2V bias to the sample. The three discrete photocurrent levels correspond to grains of different types, as marked by A, B, and C in (b). (c) and (d) show the line profile along the red mark in (b). Note t hat the streaky features on the bottom of (b) indicates a slight tip change during the scan. 113 Figure 6.2 shows the photocurrent map along with the simultaneously taken topography image of the CH 3 NH 3 PbI 3 /PEDOT:PSS/ITO sample, while - 2V bias is applied to the ITO bottom electrode with the tip grounded. This imaging condition is chosen because the work functions (measured by KPFM using highly ordered pyrolytic graphite (HOPG) as the reference) of the ITO electrode and the Pt/Ir coated AFM tip are ~5.0 eV and ~4 .6 eV, respectively, indicating that the built - in electric field for collecting photo - generated holes by ITO through the hole transport layer (PEDOT:PSS) and electrons by the AFM tip is rather weak. Photocarriers are thus more effectively extracted with th e aid of an external electric field. It can be seen in Fig ure 6.2 that the grain and GB features are more distinguishable in the photocurrent map than in the topography image. Line profiles as depicted in Fig ure 6.2c , d further illustrate that the photocur rent contrast does not originate from the height variation. It is worth pointing out that in the pc - AFM setup even though the tip makes a nano - contact with the film surface, the current spreads out beneath the tip due to the electric field distribution. 252 There are studies suggesting that more carriers will be collected in larger grains simply due to the larger volumes. 253 In order to explore the correlation between the grain area and the photocurrent, the photocurrent levels are pl otted against grain areas for the six type A grains, ten type B grains, and ten type C grains (Fig ure 6.3a). The Pearson correlation coefficient ( R ) is - 0.67 , and R 2 equals 0.45. It is not surprising that there is a weak correlation between the grain area/ volume and the photocurrent level, attributable to the confinement of photoexcited carriers due to the energy barriers for the lateral transport imposed by the grain boundaries. 253 However, if the photocurrent level is predominantly determined by the grain area, one would expect a broadly distributed photocurrent for each grain type since the areas of the grains vary significantly within each type. 114 Figure 6. 3 : Illustration of the correlation between the grain area and the photocurre nt of different types of grains . (a) Scatter plot of photocurrent levels versus grain areas, with the Pearson correlation coefficient of - 0.67 ; (b) r anges of photocurrent (black solid dots) and grain area (gray floating columns) for different types of grai ns. As shown in Fig ure 6.3 , the photocurrents are overall more narrowly distributed in comparison with the grain area s, such that the discrete photocurrent levels can be discerned in the pc - AFM image (Fig ure 6. 2b). Meanwhile, the photocurrent levels associ ated with grain A, B, and 115 C barely overlap to each other despite the considerable overlap ping of their grain areas (Fig ure 6.3 b). Note that AFM is not an ensemble averaging technique. Thus, it is important to consider the distribution, instead of just the averaged value, of the measured parameters. T he distribution of the photocurrent levels observed among the different grain types leads to the conclu sion that the grain area is not the predominant factor in determining the grain photoconductivity. The super ior spatial resolution of the photocurrent map, as evidenced by the uniform photocurrent on each grain and the sharp photocurrent contrast across grains of different type s , is likely attributable to the photoinduced giant dielectric constant in CH 3 NH 3 PbI 3 that constraints the electric field distribution. 254 Table 6. 1 : Parameters of fitted point J - V curves. Fitted parameters of the averaged point J - V curves corresponding to grain A, B, C and grai n boundary AA Type J sc A V oc V FF Power W A - 4.67 10 - 10 0.61 0.32 9.13 10 - 1 1 B - 2.28 10 - 10 0.70 0.38 6.05 10 - 1 1 C - 2.95 10 - 1 1 0.75 0.30 6.65 10 - 1 2 GB AA - 1.50 10 - 9 0.83 0.34 4.29 10 - 10 Based on the magnitude of the photocurrent, perovskite grains can be categorized into three types as shown in Fig ure 6.4a: type A ( - 2 ~ - 3 nA), type B ( - 1 ~ - 2 nA) , and type C (less than - 1 nA). To further investigate the contribution of grains to the overall device performance, point J - V spectrum measurements are perfo rmed to extract the parameters of nanoscale J SC , V OC and FF . Figure 6.4b displays the characteristic J - V spectra collected on the different grain types. Since 116 these spectra exhibit diode behaviors, similar to the macroscopic J - V curves measured on function al perovskite solar cells (Fig ure 6.1c), the diode equation, is applied to the fitting, where J SC , J 0 , q and represent photocurrent, dark current, elemental charge, and ideality factor of the diode, resp ectively. The extracted parameters are listed in Table 6.1. Note that the dependence of J SC on the grain type is consistent with that observed in the photocurrent mapping, despite the overall smaller values because of the absence of applied external field at the short - circuit condition. Figure 6. 4 : Plots of averaged photocurrents and point J - V curves of grains . (a) Average and standard deviation of photocurrent measured on the grains of different types, summarized from the photocurrent map shown in Fig ure 6.2b ; (b) a veraged point J - V curves at ten different locations for each grain type under illumination. The most striking feature revealed in Fig ure 6.4b and Table 6.1 is the anti - correlation behavior between J SC and V OC am ong the grains of different type s , that is , ones of larger J SC are associated with lower V OC . This observation is counterintuitive at first because if it is the bulk recombination that limits the device performance , one would expect a correlating trend bet ween 117 the two parameters . Nevertheless, the sample layout in the pc - AFM studies, CH 3 NH 3 PbI 3 /PEDOT:PSS/ITO, lacks an electron transport layer as compared to the complete device architecture (Fig ure 6. 1 c ). Without the selective contact layer, severe recombina tion would occur at the tip (cathode) and sample interface under the open - circuit condition. Indeed, fit of the device J - V curve to the diode equation reveals a significantly enhanced FF compared to those obtained from the pc - AFM point spectra (Table 6. 1), suggesting that surface recombination at the contact plays a more dominant role in the latter. 255 Since surface recombination is determined by the m inority carrier density, the recombination current at the tip - sample contact can be described as : , where q is the elemental charge, is the excess hole carrier density at the contact, and S is the effective recombination velocity which i s proportional to the capture coefficient for holes and the surface density of recombination centers. 164 Because of the large magnitude of S expected at the non - selective tip - perovskite contact, the surface recombination will be limited by the diffusion of holes to the contact interface via its impact on . 256 Thus, the higher the carrier mobility is, the more significant the surface recombination and the smaller the V OC will be. On the other hand, higher carrier mobility will lead to enhanced J SC under the short - circuit condition. T hereby, the anti - correlation behavior between J SC and V OC of the pc - AFM point spectra is attributed to diffusion - limited surface recombination, where gra in A carries the highest mobility followed by grain s B and C. 6.2 Identification of grain orientations in perovskite thin film s by EBSD The mobility variation is likely associated with the distinct crystal orientation of each grain type. It has been pred icted that the charge carrier mobility is a function of the crystal orientation, 152 which u ltimately leads to varied carrier migration path along different perovskite grains. 257 In order to pinpoint the crystalline orientation of each grain, electron backscatter diffraction (EBSD) 118 measurements are conducted on columned perovskite thin films. 258 Due to the damage of the sample by electron beam and the impact of surface topography on the detection o f backscattered electrons, EBSD typically yields extremely diminished signals on perovskite samples. To alleviate the challenge, perovskite films with the thickness of ~ 1µm is employed in order to allow a larger interaction volume with the electron beam. It is worthwhile to point out that the cross - sectional SEM image (Fig ure 6.5 ) confirms the columnar structure of the perovskite layer, like the thin film case. Figure 6 . 5 : C ross - sectional SEM image of the perovskite film wit h ~1 um thickness . Fig ure 6.6 present s the EBSD patterns taken at different locations of the same grain. These patterns show the same Kikuchi bands but with slight shift and contrast variation, indicating that each grain is a single crystalline domain. The shift and contrast variation can be attributed to the surface topography effect. For instance, when the electron beam strikes on the valley features of the surface, no EBSD signals can be detected as a result of the blocking of backscattered electrons. Du ring the EBSD experiment, the damage of the sample by electron beam is observed , as evidenced by the presence of pits next to the red markers shown on the top grain of the SEM image (Fig ure 6.6 ). 119 Figure 6 . 6 : Image s of EBSD patterns on the same grain . EBSD patterns taken at three different locations within one grain, as marked on the top - view SEM image. In F ig ure 6.7 , the distinct EBSD patterns taken on three grains are depicted . The indexed orientations of these grains are presented in Euler angles ), that is , a triplet of rotations that describe the grain crystalline orientation with respect to a reference coordinate system, as specified on the bottom of the indexed patterns. 120 Figure 6 . 7 : Images of EBSD patterns and indexing taken on different grains . EBSD patterns (middle panels, with and without line markers), along with their indexes (right), taken on the three different grains as marked by green cross and circled by dotte d line on the top - view SEM images (left). To guide the comparison between the EBSD patterns and their indexes, main features of the EBSD patterns are outlined with the same color as the corresponding indexed lines. Due to the weak backscattered electron in tensity from the sample, not all Kikuchi lines can be observed experimentally. Euler angles of each grain are listed on the bottom of the indexes. From the Euler angles, the Miller indices of the perovskite grains can be calculated using the following equa tions: 121 ( 6 . 1 ) ( 6 . 2 ) where are Euler angles, a , b and c are lattice constants ( a = b in this case ), and h , k and l are miller indices. 259 represents the in - plane rotation and thus will not impact the miller indices. The Miller indices derived from the EBSD measurements ( Table 6 . 2 ) match with the (110), (310) and (202) diffraction peaks observed in the XRD pattern ( Figure 6 . 8 ). Note that the angles formed between the corresponding crystalline planes are around 10 o . This experimental uncertainty is comparable with that of the EBSD measurement performed on a single crystal CH 3 NH 3 PbBr 3 with a much smoother surface. 258 Figure 6 . 8 : XRD pattern of perovskite thin film with columnar structures . The diffraction pattern demonstrat es its polycrystalline nature. Peaks labelled with grey stars correspond to the x - ray diffraction of ITO/ glass substrate. 122 Table 6 . 2 : Miller indices derived from Euler angles. The extracted M iller indices of perovskite grains match with the predominant diffraction peaks observed in the XRD pattern. The angles between the correspond ing crystalline planes are included. ) Figure 6 . 9 : Images of similar EBSD patters taken on different grains . Different grains (marked by green cross and circled by dotted line on the to p - view SEM images, left) can exhibit similar EBSD patterns (right panels, with and without line markers) with slight shift, indicating that the film has a textured structure with certain preferred crystalline orientations. To guide the view, common feature s between the two EBSD patterns are outlined. These patterns correspond to (202) crystalline orientation of the perovskite grains. 123 Lastly, similar EBSD patterns can also be identified among various grains (Fig ure 6.9), indicating that the as - prepared film demonstrates a textured structure with certain preferred crystalline orientations. In conjunction with XRD and SEM data, EBSD results convincingly support the proposed mechanism of diffusion - limited surface recombination, which correlates the discrete phot oconductivity levels with the crystalline orientation of each grain. Unfortunately, due to the roughness of the perovskite film and the damage caused by the electron beam, EBSD mapping could not be performed on such films, which could potentially elucidate the distribution of preferential crystal orientations as observed in the XRD data. Novel preparation methods are needed in order to achieve smooth surfaces on perovskite films. The other way to investigate the local crystal orientation is transmission ele ctron microscopy (TEM). The primary way to prepare the thin films for TEM tests is using focused ion beam (FIB). However, the damage caused by the FIB could be even worse than that of the electron beams. If the perovskite film could directly grow on the TE M grid, the TEM measurement is also able to elucidate the crystal orientation information of perovskite domains. In this case, it is important to choose the TEM grid with proper coating materials in order to achieve a compact thin film, similar to the film grown on the glass substrates. 6.3 Photoresponse of low - angle and high angle grain boundaries Next, the photo - response of GBs that are categorized based on the relative photoconductivity level of the adjacent grains is examined . For instance, GBs formed between adjacent grains of the same photoconductivity level are termed as boundar ies AA, BB, and CC. The other type s of GBs, including type s AB, AC, and BC, emerge between grains of different photoconductivity levels. The photocurrents of more than sevent y GBs and five data points for each GB are extracted from the photocurrent map (Fig ure 6.2) with the statistics of the analysis presented in Fig ure 6.10a. It can be seen that AA, BB and CC types of GBs carry higher 124 photocurrents than that of the adjacent grains, whereas AB, BC or AC GBs yield a photocurrent that lies between the levels of the adjacent grains. To further illustrate the impacts of GBs, point J - V measurement is performed on the AA type boundary as shown in Fig ure 6.10b. Compared with grain A , the AA boundary exhibits a higher J SC , consistent with the observation in the photocurrent map. More interestingly, V OC is also larger than that of the grain interior. With a similar FF , the maximum power output of AA boundary is around five times of the grain A, suggesting a beneficial role of such GB to the overall photovoltaic performance. Parameters extracted from the fit to the diode equation are listed in Table 6.1. According to the previous discussion, it is likely that AA, BB and CC boundaries are low - angle GBs formed between two perovskite domains of the same crystalline orientation. And AB, BC and AC boundaries are high - angle GBs located between grains that exhibit different crystalline orientations. There is one concern, which is a common iss ue in AFM techniques , about the tip artifact . The contact area between tip and sample influenced by the surface topography and the spread - out of the electric field would both impact the current measured in pc - AFM. For instance, if a side contact is establi shed, for example , at steep valley features, more photocarriers would be collected to the tip. Here , pc - AFM measurement i s conducted on perovskite thin films (110 nm) with relatively smooth surface in order to avoid the contact issue. A typical line profil e of the surface is shown in Fig ure 6.2 , where a rising angle of ~10 o is observed at the boundary between B - B grains (also illustrated in Fig ure 6.11 ). Owing to the smoothness of the surface, the contact area does not vary much when tip lands on top of GBs as compared to that landing on the grain interior (Fig ure 6.3 ). Regarding the spread - out of the electric field, the field is well confined in the perovskite layer as evidenced by the uniform photocurrent on each grain and the sharp photocurrent contrast a cross grains of different type s . To be more quantitative, it is estimate d that 125 the field spread - out is around 30 nm based on the spatial resolution of the photocurrent map (Fig ure 6. 2b). Figure 6. 10 : Plots of average d photocurrents and point J - V curves of grain boundaries . (a) Average and standard deviation of photocurrent measured at GBs of different types, summarized from the photocurrent map shown in Fig ure 6.2b. The averaged photocurrents of grain interiors are ma rked on the graph as reference. (b) Point J - V curves of type AA GBs in comparison with that of grain A (the same curve as displayed in Fig ure 6.4b). 126 Figure 6 . 11 : Schematic of the tip - sample contact and the electric field di stribution at grain boundaries and grain interiors . 6.4 Electronic structure of grain boundaries investigated by KPFM To elucidate the electronic structures at GBs, KPFM characterization is performed on CH 3 NH 3 PbI 3 /PEDOT:PSS/ITO samples. KPFM measures the work function difference between tip and sample, which is also termed as contact potential difference, i.e., CPD=W tip - W sample . The spatial variation of CPD indicates inhomogeneous sample work function which could be associated with the local bending of ban d structures and thus impacts the charge separation/transport processes. 165, 260 - 261 Figure 6.1 2 illustrates the topography and simultaneously obtained KPFM image of the same area (except some drift) before and aft er light illumination. The shift of the averaged CPD from - 110 to - 320 mV is likely induced by the spatial separation of photo - generated carriers in the perovskite film, that is , the built - up of electrons on the top surface and holes on the bottom interfac e (CH 3 NH 3 PbI 3 /PEDOT:PSS/ITO is hole selective). The purpose of KPFM studies here is to illustrate the behavior of grain boundaries, where the adjacent grains act as the reference. It can be seen in Fig ure 6.12a that the majority of the GBs 127 show a lower CP D (more negative) in comparison with that of the grains under the dark condition ; however, this contrast is reversed upon illumination. First, this observation indicates that the artifact associated with the topography convolution is negligible. Otherwise, the contrast between grains and GBs should remain consistent regardless of light illumination. Second, a downward bending of the band structure at the GBs, as illustrated in Fig ure 6.12b, can be inferred from the CPD contrast observed under illumination. This band bending will facilitate charge separation as photo - generated electrons are attracted to the GBs, and holes are repelled from them. Because electrons and holes are spatially separated, the recombination process is suppressed at the GBs, potentiall y leading to an improvement in V OC and J SC . Moreover, a reduced surface recombination is also expected at the tip - GB contact because of the lower concentration of photo - generated holes, which could further contribute to enhanc ing V OC . Last, it must be ment ioned that pc - AFM and KPFM cannot be performed at the same location as the setups for these two measurements are different. Although KPFM is not capable of distinguishing the low - angle from the large - angle GBs, it elucidates that the GBs, in general, may not be detrimental to the device performance of perovskite solar cells. The origin of the downward band bending at GBs, however, is not clear at the moment. One possibility is the unintentional doping of GBs by defects. 161, 262 - 263 It has been predicted that the intrinsic defects at surfaces and grain boundaries of CH 3 NH 3 PbI 3 do not produce deep level states, because of the anti - bonding coupling between Pb lone - pair s and I p orbitals, as well as the high ionicity a nd large lattice constants of the material. 50, 161, 264 - 265 Rather, the energy levels of these structural defects are positioned close to the band edge of CH 3 NH 3 PbI 3 . Thus, they act as unintentional doping sources instead of trapping/non - radiative recombination centers, owing to the delocalization nature of their wave functions. I t is necessary to point out that even though 128 grains with similar photocurrent levels are likely correlated to the same crystal orientation as illustrated in the EBSD and XRD results, the mis - orientation between adjacent grains of similar photocurrent could be large due to the difference in - plane rotations . Therefore, further studies on the in - plane azimuthal orientations of the grains are ne cessary in order to elucidate the detailed structures of the grain boundaries . Figure 6. 12 : Images of KPFM results and schematic of band bending at GBs . (a) Topology (left) and CPD (right) images obtained on CH 3 NH 3 PbI 3 /P EDOT:PSS/ ITO under dark (top) and illumination (bottom) conditions . (b) Proposed band alignment between grains and GBs based on the KPFM measurements. 129 Nevertheless, some studies have suggested that GBs play a detrimental role in the device performance si nce devices with larger grains demonstrate higher efficiency. 149 - 151 speculate d that whether GBs play a beneficial or detrimental role in the PV performance should heavily rely on the layout of the GB network in the perovskite film. According to the band alignment picture (Fig ure 6.12b), GBs, on the one hand, facilitate charge separation and act as electron transport channels; on the other hand, they pose energy barriers for the lateral charge transport across different grains. Therefore, in perovskite thin films composed of small grains stacking on top of each other, horizontally positioned GBs will impede the flow of charge between the top and bottom electrodes. As the grain size increases together with the en hanced crystallinity and texture of the film, grain stacking will be suppressed, and grain boundaries will align mainly along the perpendicular direction, leading to improved device performance. Lastly, future investigations into the atomic structures, che mical and electronic states of the different types of GBs, that is , low - angle vs. large - angle, are warranted to obtain a mechanistic understanding of the impacts of these GBs on the PV performance. 6.5 Conclusion P c - AFM and KPFM are performed to study the local electrical properties of grains and GBs in organic - inorganic hybrid perovskite thin films. Three discrete photoconductivity levels are identified among perovskite grains, likely corresponding to the crystal orientation of each grain. Local J - V curves recorded on these grains further suggest an anti - correlation behavior between J SC and V OC , which can be attributed to diffusion - limited surface recombination at the non - selective perovskite - tip contact. In addition, the photo - response of perovskite films displays a pronounced heterogeneity across the grain boundaries, with the boundaries formed between grains of the same photoconductivity level display ing even enhanced photocurrent and open circuit voltage compared 130 to those of the adjacent grain interiors. These observations demonstrate that the texture structure impact the charge transport process due the orientation - dependent carrier mobility and the beneficial role of low - angle grain boundaries in perovskite thin film. Overall, the results in Chapter 5 a nd 6 highlight the significance of controlling the microstructure of perovskite thin films in order to push the efficiencies to the limit. 131 Chapter 7 High - performance inverted solar cells with a controlled ZnO buffer layer In this chapter, the impact o f the processing temperature of ZnO cathode buffer layers on the device performance of OPVs is investigated. Using the sol gel method, it is found that the processing temperature of ZnO cathode buffer layers significantly influences the device performance of inverted polymer OPVs composed of blended films of P3HT and PCBM. In particular, ZnO processed at relatively low temperatures results in better device performance than those processed at higher temperatures despite the improved crystallinity and electr on mobility of the latter. This finding is attributed to the tuning of the ZnO work function with the annealing temperature, which determines the interface energetics at the cathode and thus influences the open - circuit voltage, series resistance and fill f actor. 7.1 Impact of ZnO preparation temperature on device performance of organic photovoltaics Figure 7.1(a) illustrates the electronic structure of the inverted solar cells . 266 In this electrically in verted structure, electrons are transported in PCBM to the ITO/ZnO cathode and holes in P3HT toward the MoO 3 /Ag anode. The concentration of zinc acetate (0.1 M), the thickness of ZnO (8 nm) and MoO 3 (10 nm) films, and the annealing temperature of the P3HT/ PCBM active layer (105 o C ) were all optimized. The J - V characteristic curves for devices with a single layer of ZnO buffer prepared at 300 and 450 o C are shown in Fig ure 7.1(b). The average power conversion efficiency, and the corresponding J sc , V oc , and F F are included in Table 7.1. The low - temperature annealing results in significant improvements of V oc , FF , and device performance. Figure 7.1(b) also suggests that R series increases with the temperature, leading to a smaller slope of the J - V curve at the o pen - circuit condition. Indeed, R series which is extracted by fitting the photocurrent J - V 132 curves to the Shockley diode equation 267 is 10 cm 2 for the 300 o C case and 16 cm 2 for the 450 o C case. Figure 7. 1 : Band diagram and J - V curves of inverted OPVs with ZnO processed at different conditions . (a) Layout of the inverted organic solar cell with ZnO as th e cathode buffer layer and the schematics of energy levels before reaching the equilibrium (vacuum level alignment). (b) J - V characteristics for devices based on a single layer ZnO cathode buffer prepared at 300 o C and 450 o C , respectively. (c) J - V charact eristics for two devices based on a single layer ZnO cathode buffer prepared at 300 o C with one piece of the ITO substrate was preheated at 450 o C . 133 It is well known that the optical and electrical properties of the ITO substrate can be sensitive to high te mperature annealing. Accordingly, to eliminate the impact of ITO at higher annealing temperatures , control experiments were performed with zinc acetate spun on pre - annealed (450 o C) and pristine ITO substrates, followed by 300 o C annealing of both to creat e ZnO thin films. The J - V curves of the two devices are plotted in Fig ure 7.1(c), which show nearly identical performance, suggesting that any changes in the optical and electronic properties of ITO from the higher temperature annealing do not play a signi ficant role here, distinct from an earlier report. 102 Table 7 . 1 : Summary of the average solar cell performance parameters. J sc , V oc , FF , and PCE for devices made of 0 layer (0L), 1 layer (1L), 2 layers (2L) and 3 layers (3L) ZnO buffer prepared at 300 and 450 °C, respectively. The performance parameters for 1L 3L samples are averaged over more than 10 devices at each condition , and the parameters for 0L samples are averaged over 4 devices (the device area is 4.84 mm 2 ), with the standard deviations included . ZnO buffer layer J sc mA cm - 2 V oc V F F PCE % 0L at 300 o C 9.7 ± 0.2 0.33 ± 0.01 0.43 ± 0.01 1.4 ± 0.1 1L at 300 o C 10.5 ± 0.8 0.62 ± 0.01 0.60 ± 0.02 3.9 ± 0.3 2L at 300 o C 10.4 ± 1.0 0.61 ± 0.02 0.58 ± 0.04 3.7 ± 0.6 3L at 300 o C 10.2 ± 1.1 0.61 ± 0.02 0.59 ± 0.06 3.8 ± 0.8 0L at 450 o C 9.6 ± 0.2 0.32 ± 0.01 0.41 ± 0.01 1.3 ± 0.1 1L at 450 o C 9.7 ± 1.0 0.53 ± 0.04 0.52 ± 0.03 2.7 ± 0.5 2L at 450 o C 9.7 ± 0.3 0.53 ± 0.03 0.52 ± 0.01 2.7 ± 0.2 3L at 450 o C 9.6 ± 0.4 0.51 ± 0.04 0.51 ± 0.02 2.5 ± 0.3 1L at 300 o C/ITO at 450 o C 10.1 ± 0. 7 0.60 ± 0.02 0.58 ± 0.03 3.6 ± 0.3 To further distinguish the contributions of the interface quality and the ZnO thin film conductivity, inverted devices with the ZnO buffer layer of various thickness were fabricated via multiple spin - coatings followed by thermal annealing at each step. As presented in Fig ure 7.2 and 134 Table 7.1, devices without ZnO buffer layer show diminished performance, likely due to the enhanced recombination at the cathode which decreases the shunt resistance, leading to reduced V oc and FF . Additionally, devices with the ZnO buffer layer processed at 300 o C consistently outperform those processed at 450 o C at each corresponding buffer layer thickness, where the different ZnO film thicknesses under a given annealing condition show simi lar performance in V oc and FF . This limited thickness dependence suggests that interface properties at the cathode are markedly influenced by the ZnO processing temperature , and the resistance of the ZnO thin film plays a lesser role in determining the per formance of the inverted devices. Figure 7 . 2 : J - V characteristics for devices with varied ZnO film thickness . 0 layer, 1 layer, 2 layers and 3 layers of ZnO buffer prepare d at 300 and 450 o C, respectively. 7.2 Origin of the temperature dependence To clarify the underlying mechanisms of the improved device performance in inverted solar cells with the ZnO buffer annealed at 300 o C , comprehensive experiments were performed to investigate the potential effects of thermal anneal ing on the properties of ZnO thin films , including 135 optical transmittance, thin film crystallinity, surface morphology, and work function. After examining all these factors, it is discovered that interface energetics at the cathode plays the most dominating factor in determining the device performance. Figure 7. 3 : Transmittance and XRD spectra of ITO/ZnO films annealed at different temperatures . (a) Optical transmittance for bare ITO annealed at 450 o C and ZnO/ITO prepared at 300 and 450 o C , respectively. The thickness of ZnO layer is about 8 nm. (b) XRD 2 spectra for ZnO films prepared at 300 and 450 o C on glass. The peak near 34.4 degree is characteristic for the diffraction of ZnO wurtzite phase. Inset: TGA curve of th e zinc acetate gel. The significant mass loss near 300 o C corresponds to the complete thermal decomposition of zinc acetate precursor. 136 The optical transmittance of ITO/ZnO substrates is depicted in Fig ure 7.3(a) where the low - temperature annealed substrate yields a higher transmittance than the high - temperature annealed one in the visible range. However, such a difference mainly originates from the ITO substrate, as illustrated by the similar transmittance between the bare ITO and the ITO/ZnO substrates ann ealed at 450 o C , respectively. This observation implies a limited impact of the ZnO optical properties on the OPV device performance as the films are very thin. It has been well established that higher annealing temperatures lead to improved thin film cry stallinity and enhanced electron mobility. 97, 230, 268 As presented in the XRD data of ZnO thin films in Fig ure 7.3(b), samples annealed at 300 o C show amorphous structures, whereas a pronounced ZnO (002) peak is o bserved in thin films processed at 450 o C . One may expect to obtain better electron transport, reduced R series and therefore enhanced fill factors in devices with high - temperature processed ZnO films. However, the devices processed at 300 o C show better pe rformance, including fill factor which suggests that there are other dominant factors compensating the crystallinity and mobility effect. It is worth noting that annealing the ZnO buffer layer below 300 o C results in significantly reduced device performanc e, which likely stems from the incorporation of residue zinc acetate in the ZnO film as indicated by the thermogravimetric analysis (TGA) of zinc acetate (inset of Fig ure 7.3(b)). T he roughness and homogeneity of ZnO thin films are also investigated with AFM. Figure 7.4(a) and (b) are AFM images of the single layer ZnO film prepared at 300 and 450 o C . The film prepared at 450 o C is rougher with a root mean square (rms) roughness of 2.2 nm as compared to the film prepared at 300 o C (rms of 1.7 nm), which mi ght result in a higher leakage current (smaller R shunt ) and enhanced recombination between injected holes and photo - generated electrons at the cathode. However, this difference in roughness is unlikely to be the driving force for the 137 performance difference observed between various annealing conditions since the OPV device performance is not strongly dependent on the ZnO film thickness, as suggested in Fig ure 7.2, even though the multiple layer coating is expected to improve the compactness of the ZnO film a nd thus reduces the leakage paths. Figure 7 . 4 : AFM images of ZnO films processed under different conditions . (a) and (b) are tapping - mode AFM morphology images of single layer ZnO thin film deposited on the ITO substrate an 2 . 138 7 .3 Tuning of the ZnO work function and interface energetics at the cathode In the inverted bulk heterojunctions where the exciton dissociation occurs predominantl y at the P3HT / PCBM interface, the capability of ZnO as the cathode buffer layer to collect photo - generated electrons from the PCBM directly determines the charge collection efficiency . Thus, the energy alignment (or rather, collection barriers) at the ITO/ ZnO and ZnO/PCBM interfaces is crucial to the device performance. KPFM is a useful tool to measure the work function of electrodes and the interface energetics in solar cells. 61, 269 - 270 It is worth mentioning tha t oxygen molecules adsorbed on ZnO grain boundaries can trap free electrons and cause a depletion layer near the surface. 271 - 273 Accordingly, the interface energetics between ZnO and photoactive materials and the r esultant transport properties (in dark) in solar cell devices can be impacted. It is found that light soaking using the solar illuminator is effective at removing these surface states, after which the dark current and the photocurrent merge together in the forward bias. The work functions obtained at such conditions are summarized in Table 7.2. Table 7 . 2 : Work functions (in eV ) of ITO, ITO/ZnO, and ITO/ZnO/PCBM measured by KPFM. The ITO and ITO/ZnO are annealed at 300 and 450 ° C, respectively. A representative error bar is ±0.04 eV ZnO preparation temperature ITO ITO/ZnO ITO/ZnO/PCBM 300 °C 4.77 4.36 4.34 450 °C 4.76 4.53 4.38 The energy level alignments between ITO, ZnO and PCBM from KPFM are shown in Fig ure 7.5 , where the depletion widths and band bending are inferred. One can see that the ZnO 139 work function has been tuned by thermal treatment where the Fermi level in the 300 o C annealed film is positioned in closer proximity to the conduction band edge, as compared to the o ne annealed at 450 o C . This implies that the former sample is more heavily n - doped by native defects, including Zn interstitials and oxygen vacancies, that have been partially annihilated by annealing at higher temperatures because of the improved thin fil m crystallinity. 268, 274 Consequently, the width of the Schottky barrier formed at the ITO/ZnO interface may be significantly reduced in the samples annealed at 300 o C , as shown in Fig ure 7.5. Figure 7 . 5 : Schematic of the interface energy level alignment at different interfaces . (a) and (b) are schematic band diagrams to illustrate the energy level alignment at the interfaces for devices based on the 300 and 450 °C processed ZnO fil ms, respectively, as deduced from KPFM measurements. In addition, it is found that the Fermi level of the ITO/ZnO cathode is pinned at the negative integer charge transfer state ( E ICT - ) of PCBM for both samples within the experimental error. As shown in Fi g ure 7.5(a), in the 300 o C case, a neutral contact is formed at the ZnO/PCBM interface 140 as a result of the alignment of energy levels between the cathode and PCBM. While for the ZnO film annealed at 450 o C , although its work function falls well within the t ransport gap of PCBM so that one may expect the vacuum level alignment at the interface, 68 KPFM results suggest that the Fermi level is still pinned at the E ICT - of PCBM, leading to the formation of an interface dipole. T wo possibilities are proposed to reconcile the discrepancy. First, the charge transfer behavior between ZnO and PCBM may be disturbed by the existence of interface gap states. 67 Second, if the PCBM film is unintentionally doped by impurities, the imbalan ce in work functions upon contact with the ITO/ZnO cathode (450 o C ) can be compensated by the electron flow from PCBM to the cathode. 269 - 270 Finally, the impact of the interface energetics on the solar cell device performance is discussed . As presented in Table 7. 1, the devices made of the 300 o C annealed ZnO cathode buffer display a higher FF , a larger PCE and an optimal V oc as obtained in the P3HT:PCBM systems with ohmic contacts. 72, 275 On the contrary, V oc , FF , and PCE are reduced by about 15 %, 13 %, and 31 %, respectively , in the devices composed of the 450 o C annealed ZnO film. These findings suggest that there are several factors contributing to the enhanced performa nce in devices composed of the ZnO buffer layer processed at 300 o C . (1) Charge collection at the ITO/ZnO interface may be improved by electron tunneling through the Schottky barrier of reduced width; (2) the Fermi level of the ITO/ZnO cathode lines up wit h the E ICT - of PCBM which enhances the electronic coupling at the interface and minimizes the V oc loss. In contrast, the extraction barrier at the ZnO (450 o C )/ PCBM interface may result in a significant charge accumulation and the consequent recombination loss at the interface; (3) these two effects also contribute to the low contact resistance, thereby a smaller R series and a larger FF in the 300 o C case. 141 7.4 Conclusion The inverted solar cells with controlled ZnO cathode buffer layers in the ITO/ZnO/ P3 HT:PCBM/MoO 3 /Ag structure are fabricated, which are comparable to the best conventional cells. Through comprehensive characterization of the surface morphology, thin film crystallinity and optical and electrical properties, it is discovered that the tuning of the ITO/ZnO work function and the interface energetics play a dominant role in determining the device performance for sol gel processed ZnO. These findings could aid in the design and interface engineering of high quality OPVs incorporating ZnO buffer layers on low temperature, flexible substrates 142 Chapter 8 Conclusion s and future work 8.1 Summary of results To boost the device performance of photovoltaic devices, it is critical to understand t he impact of microstructures , such as the thin film texture, grain s and grain boundar ies , on the device performance, especially for devices composed of polycrystalline films. A strong correlation between the efficiency and microstructure is demonstrated in o rganic - inorganic hybrid perovskite solar cells, wh ich is one of the most promising candidate s for the next generation of commercialized photovoltaic devices . As shown in Chapter 5, b y adopting the ramp annealing process, both the morphology and the texture structure of polycrystalline perovskite thin fil m s can be effectively tailored. Compared with the complicated two - step or solvent/vapor annealing methods as introduced in Chapter 4 , the ramp annealing treatment reduces the complexity during the thin film processing yet still enables high - quality perovsk ite devices , which is advantageous for the large - scale manufacturing process. By employing EIS measurements , the impact of texture structure on the carrier dynamic processes , such as carrier transport in the bulk and charge collection at the interface , is elucidated. Though the interface , instead of the bulk, has been argued to render severe recombination in perovskite devices , t his finding indicates the importance of controlling the crystallinity /texture of perovskite thin film s . Also, t he enhanced texture structure in perovskite thin film s is accompanied by a n increase in the number of low - angle grain boundaries. It is well - known that there is a higher density of defects at GBs due to the broken symmetry of the crystal planes. T he defects in perovskite thi n films ha ve been theoretically investigated , and the results show that they do not generate mid - gap state s or act as recombination centers. However , the detrimental role of them ha s been observed 143 experimentally. Taking these factors into consideration, it is necessary to investigate the impact s of different types of grain boundaries on the performance of perovskite solar cells . P c - AFM and KPFM , which provide nano - scale resolution on charge transport, are employed to investigate the contributions of grain s and grain boundaries to the overall efficiency of perovskite device s. For the perovskite thin film consisting of multiple grains stacked in the vertical direction, the measured current would be convoluted by grains and grain boundaries in the film. Unfor tunately, t he issue of grain and grain boundary stacking has been largely neglected in most of the previous AFM studies on perovskite thin films. To isolate the contributions of different types of grains and grain boundaries , it is crucial to employ perovs kite thin films with columnar structures , where t he orientation dependent photoresponse of grains and the heterogeneous behaviors of grain boundaries are observed . In details, t he discrete photocurrent levels and anti - correlation behavior between the photo current and photovoltage measured by the local J - V curves on different grains indicate that the crystal line orientation of grains (extracted separately by the EBSD measurement) impact s carrier transport properties. These observations support the propositio n of the correlation between crystal orientation and carrier mobility. In addition, pc - AFM results indicate the non - detrimental role of grain boundaries in perovskite thin films with low - angle grain boundaries showing improved photocurrent and photovoltage . The downward band bending at GB, as revealed by KPFM, serves to attract electrons and repel holes. Consequently, GBs enhance the carrier transport and suppress the recombination. These discoveries point out the importance of controlling the grain orienta tion and low - angle grain boundaries in perovskite thin films for achiev ing efficient devices. In addition to perovskite solar cells, OPVs have also drawn a significant amount of attention due to the low cost and mechanical flexibility. Besides the active layer, t he interface is 144 also crucial in determining the device performance. In bulk heterojunction solar cells, it requires a high selectivity at the electrodes due to the existence of both donor and acceptor at the interface. By tuning the post - annealing temperature of sol - gel processed ZnO film, the work function of the cathode is tuned. The mechanism of work function tuning is the variation in the density of intrinsic defects, which acts as the dopant, with annealing temperature. By controlling the defe ct density, a sharp Schottky barrier is formed between ITO (cathode) and ZnO (electron transport layer) , which electrons can tunnel through. Accordingly , the charge injection efficiency is improved. These findings not only aid in the design of OPVs but dem onstrate the importance of interfacial engineering process which helps reduc e the energy loss and improve the device performance. Overall, the carrier transport and charge collection process es have been investigated in different types of solar cells. For devices composed of polycrystalline films , like perovskite solar cells , it is critical to understand the influence of microstructures on device performance . In Chapter 5, the correlation between device performance and texture structure of perovskite thin film is discovered in MAPbI 3 based perovskite solar cells. The ramp annealing treatment is an effective way to tune the crystallinity. Nevertheless, i n order to largely enhance the texture structure of perovskite thin film s , many methods can be conducted s uch as doping the material with chloride. For GBs, only low - angle GBs in perovskite thin films demonstrate a beneficial role as revealed in this work, which is likely related to the varied defect chemistry . By modifying the defect chemistry, such as adding extra CH 3 NH 3 I, the majority of the GBs may be activated and show a higher conductivity. 276 Additionall y , interfacial engineering can effectively enhance the charge collection at the interface and thus contribute to the device performance. Ideally, the formation of Ohmic contact s at the interface s is expected for efficient charge collection. In our study in Chapter 7 , t hough a Schottky barrier is formed at the interface , by controlling the width of the barrier , an 145 effective charge collection is enabled with electrons tunnel ing th r ough the barrier. These discoveries provide deep physical insight s into the imp act of microstructures on the physical processes in photovoltaic devices and offer new strategies of device engineering to improve the overall efficiency . 8.2 Future work As we have extensively discussed, AFM measurements are capable of providing nano - sc ale information. However, taking the c - AFM or pc - AFM measurements as an example, the transport properties measured locally are convoluted between the surface/contact and the bulk. On one hand, a s discussed in Chapter 5 , EIS is capable of disentangling the physical process occurring in the bulk and at the contact of solar cell devices at the macroscopic scale. Thus, the incorporation of EIS into AFM will be powerful at revealing the relevant physical processes at the nanoscale, offering unprecedent insights into the operation of solar cells and the limiting factors of the device efficiency. As mentioned in Chapter 3, one of the challenge s faced in perovskite solar cells is the hysteresis issue which is likely associated with the ionic migration. It has bee n suggested that the ions move faster along grain boundaries due to the high density of local defects. 193, 277 Beyond the R s which would increase upon t he accumulation of ions at the interface , the interfacial capa citance ( C 1 ) in the equivalent circuit as shown in Chapter 5 is also likely impacted by the ion accumulation at the interface. Therefore, b y performing EIS measurement s at grain interiors and grain boundaries, the potential impacts of grain boundaries on t he ion migration could be revealed. In addition, similar to the carrier transport, the migration path of ions could also depend on the crystal orientation. By comparing the impedance responses on various grains of the different (hkl) orientation, informati on on the orientation - dependency of ion migration could be illustrated. 146 To commercialize perovskite solar cells, it is essential to improve the stability of the devices. Due to the broken symmetry, the moisture in the ambient atmosphere is more likely to diffuse into the perovskite film at GBs. The consequence of the decomposition is the formation of volatile CH 3 NH 2 and hydroiodic acid (HI), leaving PbI 2 in the film. PbI 2 demonstrates a large bandgap and poor carrier transport properties. Therefore, the d ecomposition of perovskite materials would diminish the conductivity resulting in the increase of R s . Furthermore, the dielectric constant of PbI 2 and perovskite materials is different, and the C 2 as defined in the equivalent circuit would be affected. By comparing the impedance response of grain and grain boundary, the decomposition process in the perovskite thin films could be spatially resolved. performed w ith AFM due to the point - contact between tip and sample and the spread - out of the electric field. Thus, it might be challenging to solely analyze the capacitance and resistance since such as time constant, would be more characteristic since the variation in the effective area is canceled out by the multiplication of resistance and capacitance. 147 APPENDIX 148 APPENDIX This part would focus on the spin - coating method and the fabrication process of perovskite solar cells. The film quality, such as morphology, surface coverage, crystallinity and defect density, impacts the device performance. Therefore, it is essential to prepare films with high quality. The most widely used method for the fabrication of perovskite films and devices is the spin - coating deposition. Spin - coating deposition is one of the most widely used techniques to prepare thin films with the thickness ranging from a few nanometers to microns on a flat substrate. The advant ages of spin - on deposition are the easy setup and the high uniformity. Generally, two steps are involved in the spin - coating process. The first step is to dispense the precursor solution onto the substrates. To ensure the surface coverage and uniformity, i t requires the solution to fully cover the substrate, which is determined by the surface tension of the solution and the surface energy o the substrates. Precursors with a low surface tension tend to spread - out well over the substrates. Also, when the surf ace energy is high, it is more favorable to cover the substrate with precursors to minimize the surface energy. A good example to illustrate the dispensing issue is the preparation of PEDOT:PSS layer. PEDOT:PSS (organic salt) is dissolved in water. The so lution demonstrates a high surface tension, which poses the challenge in the dispensing process. To overcome this issue, ITO substrates are treated with oxygen plasma which increases the surface energy by adding - OH groups to the surface. Yet, it enhances the interaction of the solvent (H 2 O) and the ITO surface via the formation of hydrogen bond. Therefore, the PEDOT:PSS film with high surface coverage is achieved. The second step is spinning which includes two stages. At first, most of the precursor solut ion is removed when the spinning starts. Then, the remaining solute and solvent start to 149 densify with further removal of the solvent molecules. In this step, the film quality such as surface coverage and crystallinity is further impact by the solvent used in the precursor solution. When the boiling point of the solvent is low, it evaporates fast during the spin - coating, leading to films with high surface coverage but low crystallinity. On the contrary, the long time is required to remove the solvent molecul es with a high boiling point, which promotes the order of the molecular arrangement (crystallinity). However, due to the strong interaction between precursor molecules compared with precursor and substrates, the solute molecules tend to grow into islands, which reduces the surface coverage. To achieve thin films with both high surface coverage and crystallinity, the mixture of solvents with high and low boiling points could be adopted. The densification of the film is completed while the solvent with a low boiling point is removed. The remaining solvent molecules with a high boiling point would assist the arrangement of the precursor molecules. Figure A. 1 : Solvent engineering of spin - coating method preparing halide perovskite t hin film : the impact of the boiling point of the solvents. - butyrolactone (GBL). In this work, PbI 2 and CH 3 NH 3 I are first dissolved in DMF to fabricate the 150 planar - structured perovskit e solar cells. But the surface coverage is quite low when spin - coated onto PEDOT:PSS substrates. Two strategies are developed to improve the surface coverage. The first way is by adding toluene (2%) to DMF. The boiling point of toluene is 110.6 o C which is lower than that of DMF , which accelerates the evaporation of the solvent. Also, the solubility of PbI 2 in toluene is very low, leading to the fast densification of precursors on the substrates. The SEM images of perovskite films prepared with different pr ecursor solutions are illustrate d in Figure A.2. In addition to the improvement of surface coverage, the grain size also increases. A further study on the impact of toluene on the film quality of perovskite, such as grain size and crystallinity , is necessa ry in the future . Figure A. 2 : SEM images of perovskite films prepared from precursors with different solvents. (a) and (b) are the top - view and cross - section SEM images of perovskite films prepared from DMF. (c) and (d) are t he top - view SEM images of perovskite film prepared from DMF/Toluene and DMF/DMSO solvents. 151 The other way is by using the mixture of DMF and DMSO. Though the boiling point of DMSO is higher than DMF, it is discovered that DMSO reduces the surface tension o f the precursor solution and induces the formation of intermediate phase. As discussed in Chapter 5, the intermediate phase demonstrates a large d - spacing compared with PbI 2 , which facilitate the formation of perovskite phase by promoting the incorporatio n of organic cations (CH 3 NH 3 + ). Consequently, the addition of DMSO improves the dispensing of precursor solution and accelerates the crystallization of perovskite materials. The corresponding SEM images are shown in Figure A.2. The smaller grain size of pe rovskite films processed with DMF/DMSO is likely due to the fast crystallization induced by the intermediate phase, which limit s the mass transport and growth of perovskite grains. In Chapter 6, to grow perovskite films with columnar structures, two - step spin - coating is adopted together with solvent annealing. The purpose of using two - step instead of one - step is to retard the crystallization of perovskite films. As shown in Figure A.2, the accelerated nucleation results in small grain sizes. Therefore, by spin - coating PbI 2 and CH 3 NH 3 I separately, it requires Pb 2+ , I - , and CH 3 NH 3 + to diffuse across the film in order to form the perovskite phase. To further increase the grain size, the solvent annealing is employed. By annealing the perovskite film in the DM F vapor, the diffusivity of ions is improved which enhance the growth of each grain. It is worth mentioning that the perovskite film is very sensitive to the DMF vapor. Too much DMF would lead to the decompose of perovskite phase into PbI 2 . Therefore, the amount of DMF depends on preparing the film of 100 nm, 10 microliter s DMF is enough to make columnar grains. However, when preparing films with 1 - micron thickness, 20 - 3 0 microliter DMF is needed. In addition, the annealing time extends from 30 min to 5 hours. Other solvents, such as DMSO and GBL, are also 152 used in the solvent annealing. But the film quality is not comparable to that of DMF. It is likely due to their highe r boiling point which requires a higher annealing temperature. And the perovskite film would decompose at elevated temperatures. Another factor that would impact the film quality is the morphology of PbI 2 layer. We have observed that the film quality is be tter when the PbI 2 layer is annealing for 1 - 5 min. Further systematic studies on the impact of PbI 2 (morphology and crystallinity) on the final film quality in the solvent annealing is necessary. In this work, the thickness of perovskite films has also be en tuned by controlling the concentration of the precursor solution. The film thickness increases linearly with the concentration within a certain range. However, the concentration is also limited by the solubility. When depositing perovskite films of 1 mi cron, the precursor solution is heated at 80 o C. The preheat treatment of precursor may also reduce the viscosity of the solution which benefit s the dispensing of the precursor. The other approach is to adjust the spin - speed. The film thickness is reversel y proportional to the square root of spin - speed. So, the film thickness varies 3 times when the spin - speed varies 10 times. The concern is that spin - coating at low speed could reduce the uniformity of the film. Consequently, it is recommended to tune the f ilm thickness via adjusting the concentration. The film quality is critical in determining the efficiency of photovoltaic devices. Compared with other deposition methods, the spin - coating method is simple and facile. And t he thin film properties such as m orphology and crystallinity can be effectively engineered by solvent formulation and process control. 153 BIBLIOGRAPHY 154 BIBLIOGRAPHY 1. Hoel, M.; Kverndokk, S., Depletion of fossil fuels and the impacts of global warming. Resource and Ener gy Economics 1996, 18 (2), 115 - 136. 2. Hall, C. A. S., Fossil Fuels. In Energy Return on Investment: A Unifying Principle for Biology, Economics, and Sustainability , Springer International Publishing: Cham, 2017; 95 - 105. 3. Bolin, B.; Doos, B.; WARRICK, R. ; JAEGER, J., The greenhouse effect, climatic change, and ecosystems((Book)). Chichester, England and New York, John Wiley and Sons(SCOPE 1986, (29). 4. Schneider, S. H., The greenhouse effect - Science and policy. Science 1989, 243 (4892), 771 - 781. 5. Le Quéré, C.; Andres, R. J.; Boden, T.; Conway, T.; Houghton, R. A.; House, J. I.; Marland, G.; Peters, G. P.; van der Werf, G. R.; Ahlström, A.; Andrew, R. M.; Bopp, L.; Canadell, J. G.; Ciais, P.; Doney, S. C.; Enright, C.; Friedlingstein, P.; Huntingford, C.; Jain, A. K.; Jourdain, C.; Kato, E.; Keeling, R. F.; Klein Goldewijk, K.; Levis, S.; Levy, P.; Lomas, M.; Poulter, B.; Raupach, M. R.; Schwinger, J.; Sitch, S.; Stocker, B. D.; Viovy, N.; Zaehle, S.; Zeng, N., The global carbon budget 1959 2011. Earth Syst. Sci. Data 2013, 5 (1), 165 - 185. 6. Arneth, A.; Sitch, S.; Pongratz, J.; Stocker, B. D.; Ciais, P.; Poulter, B.; Bayer, A. D.; Bondeau, A.; Calle, L.; Chini, L. P.; Gasser, T.; Fader, M.; Friedlingstein, P.; Kato, E.; Li, W.; Lindeskog, M.; Nabel, J. E. M. S.; Pugh, T. A. M.; Robertson, E.; Viovy, N.; Yue, C.; Zaehle, S., Historical carbon dioxide emissions caused by land - use changes are possibly larger than assumed. Nature Geosci 2017, 10 (2), 79 - 84. 7. Administration, U. S. E. I. U.S. Energy Consumpt ion by Energy Source, 2016. https://www.eia.gov/energyexplained/?page=us_energy_home (accessed October). 8. Lewis, N. S.; Nocera, D. G., Powering the planet: Chemical challenges in so lar energy utilization. Proceedings of the National Academy of Sciences 2006, 103 (43), 15729 - 15735. 9. Global Energy Statistical Yearbook 2017,Accessed November 2017 . 10. multicrystalline and 24.4% monocrystalline silicon solar cells. Applied Physics Letters 1998, 73 (14), 1991 - 1993. 11. Fabre, E.; Baudet, Y. In Polycrystalline silicon solar cells , Photovoltaic Solar Energy Conference, 1978; 178 - 186. 12. Schropp, R. E.; Ca rius, R.; Beaucarne, G., Amorphous silicon, microcrystalline silicon, and thin - film polycrystalline silicon solar cells. MRS bulletin 2007, 32 (3), 219 - 224. 155 13. Pillai, S.; Catchpole, K.; Trupke, T.; Green, M., Surface plasmon enhanced silicon solar cells. Journal of applied physics 2007, 101 (9), 093105. 14. Carlson, D. E.; Wronski, C. R., Amorphous silicon solar cell. Applied Physics Letters 1976, 28 (11), 671 - 673. 15. Derkacs, D.; Lim, S.; Matheu, P.; Mar, W.; Yu, E., Improved performance of amorphous si licon solar cells via scattering from surface plasmon polaritons in nearby metallic nanoparticles. Applied Physics Letters 2006, 89 (9), 093103. 16. Repins, I.; Contreras, M. A.; Egaas, B.; DeHart, C.; Scharf, J.; Perkins, C. L.; To, B.; Noufi, R., 19 . 9% e fficient ZnO/CdS/CuInGaSe2 solar cell with 81 . 2% fill factor. Progress in Photovoltaics: Research and applications 2008, 16 (3), 235 - 239. 17. Contreras, M. A.; Egaas, B.; Ramanathan, K.; Hiltner, J.; Swartzlander, A.; Hasoon, F.; Noufi, R., Progress toward 20% efficiency in Cu (In, Ga) Se2 polycrystalline thin film solar cells. Progress in Photovoltaics: Research and applications 1999, 7 (4), 311 - 316. 18. Ramanathan, K.; Contreras, M. A.; Perkins, C. L.; Asher, S.; Hasoon, F. S.; Keane, J.; Young, D.; Romer o, M.; Metzger, W.; Noufi, R., Properties of 19.2% efficiency ZnO/CdS/CuInGaSe2 thin film solar cells. Progress in Photovoltaics: research and applications 2003, 11 (4), 225 - 230. 19. Nelson, J., Polymer:fullerene bulk heterojunction solar cells. Materials Today 2011, 14 (10), 462 - 470. 20. Günes, S.; Neugebauer, H.; Sariciftci, N. S., Conjugated polymer - based organic solar cells. Chemical reviews 2007, 107 (4), 1324 - 1338. 21. Hoppe, H.; Sariciftci, N. S., Organic solar cells: An overview. Journal of material s research 2004, 19 (7), 1924 - 1945. 22. O'Regan, B.; Gratzel, M., A low - cost, high - efficiency solar cell based on dye - sensitized colloidal TiO2 films. Nature 1991, 353 (6346), 737 - 740. 23. Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson, H., Dye - sensitized solar cells. Chem. Rev 2010, 110 (11), 6595 - 6663. 24. Grätzel, M., Dye - sensitized solar cells. Journal of Photochemistry and Photobiology C: Photochemistry Reviews 2003, 4 (2), 145 - 153. 25. Grätzel, M., Conversion of sunlight to electric power b y nanocrystalline dye - sensitized solar cells. Journal of Photochemistry and Photobiology A: Chemistry 2004, 164 (1), 3 - 14. 26. Green, M. A.; Ho - Baillie, A.; Snaith, H. J., The emergence of perovskite solar cells. Nat Photon 2014, 8 (7), 506 - 514. 156 27. Liu, M .; Johnston, M. B.; Snaith, H. J., Efficient planar heterojunction perovskite solar cells by vapour deposition. Nature 2013, 501 (7467), 395 - 398. 28. Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J., Efficient hybrid solar cells based on meso - superstructured organometal halide perovskites. Science 2012, 338 (6107), 643 - 647. 29. Hodes, G., Applied physics. Perovskite - based solar cells. Science 2013, 342 (6156), 317 - 8. 30. Li, X.; Bi, D.; Yi, C.; Décoppet, J. - D.; Luo, J.; Zakeeruddin, S. M.; Hagfeldt, A.; Grätzel, M., A vacuum flash assisted solution process for high - efficiency large - area perovskite solar cells. Science 2016 . 31. Ball, J. M.; Lee, M. M.; Hey, A.; Snaith, H. J., Low - temperature processed meso - superstructured to thin - film p erovskite solar cells. Energy & Environmental Science 2013, 6 (6), 1739. 32. Docampo, P.; Ball, J. M.; Darwich, M.; Eperon, G. E.; Snaith, H. J., Efficient organometal trihalide perovskite planar - heterojunction solar cells on flexible polymer substrates. N at Commun 2013, 4 , 2761. 33. Sun, S.; Salim, T.; Mathews, N.; Duchamp, M.; Boothroyd, C.; Xing, G.; Sum, T. C.; Lam, Y. M., The origin of high efficiency in low - temperature solution - processable bilayer organometal halide hybrid solar cells. Energy Environ. Sci. 2014, 7 (1), 399 - 407. 34. M. M.; Lanzani, G.; Snaith, H. J.; Petrozza, A., Excitons versus free charges in organo - lead tri - halide perovskites. Nat Commun 2014, 5 , 3586. 35. Galkowski, K.; Mitioglu, A.; Miyata, A.; Plochocka, P.; Portugall, O.; Eperon, G. E.; Wang, J. T. - W.; Stergiopoulos, T.; Stranks, S. D.; Snaith, H. J.; Nicholas, R. J., Determination of the exciton binding energy and effective masses for methy lammonium and formamidinium lead tri - halide perovskite semiconductors. Energy & Environmental Science 2016, 9 (3), 962 - 970. 36. Miyata, A.; Mitioglu, A.; Plochocka, P.; Portugall, O.; Wang, J. T. - W.; Stranks, S. D.; Snaith, H. J.; Nicholas, R. J., Direct m easurement of the exciton binding energy and effective masses for charge carriers in organic - inorganic tri - halide perovskites. Nat Phys 2015, 11 (7), 582 - 587. 37. Zhao, T.; Shi, W.; Xi, J.; Wang, D.; Shuai, Z., Intrinsic and Extrinsic Charge Transport in C H3NH3PbI3 Perovskites Predicted from First - Principles. Scientific Reports 2016, 7 , 19968. 38. Lee, Y. H.; Luo, J.; Son, M. - K.; Gao, P.; Cho, K. T.; Seo, J.; Zakeeruddin, S. M.; Grätzel, M.; Nazeeruddin, M. K., Enhanced Charge Collection with Passivation La yers in Perovskite Solar Cells. Advanced Materials 2016, 28 (20), 3966 - 3972. 157 39. Bae, S.; Park, J. - S.; Han, I. K.; Shin, T. J.; Jo, W. H., CH3NH3PbI3 crystal orientation and photovoltaic performance of planar heterojunction perovskite solar cells. Solar En ergy Materials and Solar Cells 2017, 160 , 77 - 84. 40. Correa Baena, J. P.; Steier, L.; Tress, W.; Saliba, M.; Neutzner, S.; Matsui, T.; Giordano, F.; Jacobsson, T. J.; Srimath Kandada, A. R.; Zakeeruddin, S. M.; Petrozza, A.; Abate, A.; Nazeeruddin, M. K.; Gratzel, M.; Hagfeldt, A., Highly efficient planar perovskite solar cells through band alignment engineering. Energy & Environmental Science 2015, 8 (10), 2928 - 2934. 41. Li, S. - S.; Chang, C. - H.; Wang, Y. - C.; Lin, C. - W.; Wang, D. - Y.; Lin, J. - C.; Chen, C. - C. ; Sheu, H. - S.; Chia, H. - C.; Wu, W. - R.; Jeng, U. S.; Liang, C. - T.; Sankar, R.; Chou, F. - C.; Chen, C. - W., Intermixing - seeded growth for high - performance planar heterojunction perovskite solar cells assisted by precursor - capped nanoparticles. Energy & Environ mental Science 2016, 9 (4), 1282 - 1289. 42. Bi, D.; Yi, C.; Luo, J.; Décoppet, J. - D.; Zhang, F.; Zakeeruddin, Shaik M.; Li, X.; Hagfeldt, A.; Grätzel, M., Polymer - templated nucleation and crystal growth of perovskite films for solar cells with efficiency gr eater than 21%. Nature Energy 2016, 1 , 16142. 43. Giesbrecht, N.; Schlipf, J.; Oesinghaus, L.; Binek, A.; Bein, T.; Muller - Buschbaum, P.; Docampo, P., Synthesis of Perfectly Oriented and Micrometer - Sized MAPbBr3 Perovskite Crystals for Thin - Film Photovolta ic Applications. Acs Energy Letters 2016, 1 (1), 150 - 154. 44. Chen, H.; Zheng, X.; Li, Q.; Yang, Y.; Xiao, S.; Hu, C.; Bai, Y.; Zhang, T.; Wong, K. S.; Yang, S., An amorphous precursor route to the conformable oriented crystallization of CH3NH3PbBr3 in mes oporous scaffolds: toward efficient and thermally stable carbon - based perovskite solar cells. Journal of Materials Chemistry A 2016, 4 (33), 12897 - 12912. 45. Niu, G.; Yu, H.; Li, J.; Wang, D.; Wang, L., Controlled orientation of perovskite films through mi xed cations toward high performance perovskite solar cells. Nano Energy 2016, 27 , 87 - 94. 46. Ye, J.; Zheng, H.; Zhu, L.; Zhang, X.; Jiang, L.; Chen, W.; Liu, G.; Pan, X.; Dai, S., High - temperature shaping perovskite film crystallization for solar cell fast preparation. Solar Energy Materials and Solar Cells 2017, 160 , 60 - 66. 47. Wu, B.; Nguyen, H. T.; Ku, Z.; Han, G.; Giovanni, D.; Mathews, N.; Fan, H. J.; Sum, T. C., Discerning the Surface and Bulk Recombination Kinetics of Organic Inorganic Halide Perovsk ite Single Crystals. Advanced Energy Materials 2016, 6 (14), 1600551. 48. Edri, E.; Kirmayer, S.; Henning, A.; Mukhopadhyay, S.; Gartsman, K.; Rosenwaks, Y.; Hodes, G.; Cahen, D., Why lead methylammonium tri - iodide perovskite - based solar cells require a me soporous electron transporting scaffold (but not necessarily a hole conductor). Nano Lett 2014, 14 (2), 1000 - 4. 49. Kutes, Y.; Zhou, Y.; Bosse, J. L.; Steffes, J.; Padture, N. P.; Huey, B. D., Mapping the Photoresponse of CH3NH3PbI3 Hybrid Perovskite Thin Films at the Nanoscale. Nano Letters 2016, 16 (6), 3434 - 3441. 158 50. Yang, M.; Zeng, Y.; Li, Z.; Kim, D. H.; Jiang, C. - S.; van de Lagemaat, J.; Zhu, K., Do grain boundaries dominate non - radiative recombination in CH3NH3PbI3 perovskite thin films? Physical Che mistry Chemical Physics 2017, 19 (7), 5043 - 5050. 51. deQuilettes, D. W.; Vorpahl, S. M.; Stranks, S. D.; Nagaoka, H.; Eperon, G. E.; Ziffer, M. E.; Snaith, H. J.; Ginger, D. S., Impact of microstructure on local carrier lifetime in perovskite solar cells. Science 2015, 348 (6235), 683 - 686. 52. Yin, W. - J.; Shi, T.; Yan, Y., Unique Properties of Halide Perovskites as Possible Origins of the Superior Solar Cell Performance. Advanced Materials 2014, 26 (27), 4653 - 4658. 53. Guo, Y.; Wang, Q.; Saidi, W. A., Struc tural Stabilities and Electronic Properties of High - Angle Grain Boundaries in Perovskite Cesium Lead Halides. The Journal of Physical Chemistry C 2017, 121 (3), 1715 - 1722. 54. Docampo, P.; Hanusch, F. C.; Giesbrecht, N.; Angloher, P.; Ivanova, A.; Bein, T. , Influence of the orientation of methylammonium lead iodide perovskite crystals on solar cell performance. Apl Mater 2014, 2 (8), 081508. 55. Yin, J.; Cortecchia, D.; Krishna, A.; Chen, S.; Mathews, N.; Grimsdale, A. C.; Soci, C., Interfacial Charge Trans fer Anisotropy in Polycrystalline Lead Iodide Perovskite Films. J Phys Chem Lett 2015, 6 (8), 1396 - 1402. 56. Leblebici, S. Y.; Leppert, L.; Li, Y.; Reyes - Lillo, S. E.; Wickenburg, S.; Wong, E.; Lee, J.; Melli, M.; Ziegler, D.; Angell, D. K.; Ogletree, D. F .; Ashby, Paul D.; Toma, F. M.; Neaton, J. B.; Sharp, I. D.; Weber - Bargioni, A., Facet - dependent photovoltaic efficiency variations in single grains of hybrid halide perovskite. Nature Energy 2016, 1 , 16093. 57. Brabec, C. J.; Sariciftci, N. S.; Hummelen, J. C., Plastic Solar Cells. Advanced Functional Materials 2001, 11 (1), 15 - 26. 58. Coakley, K. M.; McGehee, M. D., Conjugated Polymer Photovoltaic Cells. Chemistry of Materials 2004, 16 (23), 4533 - 4542. 59. Hau, S. K.; Yip, H. - L.; Acton, O.; Baek, N. S.; M a, H.; Jen, A. K. Y., Interfacial modification to improve inverted polymer solar cells. Journal of Materials Chemistry 2008, 18 (42), 5113 - 5119. 60. Kuwabara, T.; Kawahara, Y.; Yamaguchi, T.; Takahashi, K., Characterization of Inverted - Type Organic Solar C ells with a ZnO Layer as the Electron Collection Electrode by ac Impedance Spectroscopy. ACS Applied Materials & Interfaces 2009, 1 (10), 2107 - 2110. 61. Davis, R. J.; Lloyd, M. T.; Ferreira, S. R.; Bruzek, M. J.; Watkins, S. E.; Lindell, L.; Sehati, P.; Fa hlman, M.; Anthony, J. E.; Hsu, J. W. P., Determination of energy level alignment at interfaces of hybrid and organic solar cells under ambient environment. Journal of Materials Chemistry 2011, 21 (6), 1721 - 1729. 159 62. Hau, S. K.; Yip, H. - L.; Ma, H.; Jen, A. K. - Y., High performance ambient processed inverted polymer solar cells through interfacial modification with a fullerene self - assembled monolayer. Applied Physics Letters 2008, 93 (23), 233304. 63. Stubhan, T.; Salinas, M.; Ebel, A.; Krebs, F. C.; Hirsch, A.; Halik, M.; Brabec, C. J., Increasing the Fill Factor of Inverted P3HT:PCBM Solar Cells Through Surface Modification of Al - Doped ZnO via Phosphonic Acid - Anchored C60 SAMs. Advanced Energy Materials 2012, 2 (5), 532 - 535. 64. Bai, S.; Wu, Z.; Xu, X.; Jin , Y.; Sun, B.; Guo, X.; He, S.; Wang, X.; Ye, Z.; Wei, H.; Han, X.; Ma, W., Inverted organic solar cells based on aqueous processed ZnO interlayers at low temperature. Applied Physics Letters 2012, 100 (20), 203906. 65. Jie Luo, H. W., Chao He. Aiyuan Li, Wei Yang, and Yong Cao, Enhanced open - circuit voltage in polymer solar cells. Applied Physics Letters 2009, 95 (4), 043301. 66. Chen, L. - M.; Xu, Z.; Hong, Z.; Yang, Y., Interface investigation and engineering - achieving high performance polymer photovolta ic devices. Journal of Materials Chemistry 2010, 20 (13), 2575 - 2598. 67. Hwang, J.; Wan, A.; Kahn, A., Energetics of metal organic interfaces: New experiments and assessment of the field. Materials Science and Engineering: R: Reports 2009, 64 (1), 1 - 31. 68 . Braun, S.; Salaneck, W. R.; Fahlman, M., Energy - Level Alignment at Organic/Metal and Organic/Organic Interfaces. Advanced Materials 2009, 21 (14 - 15), 1450 - 1472. 69. Mihailetchi, V. D.; Blom, P. W. M.; Hummelen, J. C.; Rispens, M. T., Cathode dependence o f the open - circuit voltage of polymer:fullerene bulk heterojunction solar cells. Journal of Applied Physics 2003, 94 (10), 6849 - 6854. 70. Tress, W.; Leo, K.; Riede, M., Influence of Hole - Transport Layers and Donor Materials on Open - Circuit Voltage and Shap e of I V Curves of Organic Solar Cells. Advanced Functional Materials 2011, 21 (11), 2140 - 2149. 71. He, C.; Zhong, C.; Wu, H.; Yang, R.; Yang, W.; Huang, F.; Bazan, G. C.; Cao, Y., Origin of the enhanced open - circuit voltage in polymer solar cells via inte rfacial modification using conjugated polyelectrolytes. Journal of Materials Chemistry 2010, 20 (13), 2617 - 2622. 72. - J.; Yip, H. - L.; Ma, H.; Jen, A. K. - Y., Optimization of active layer and anode electrode for high - per formance inverted bulk - heterojunction solar cells. IEEE Journal of Selected Topics in Quantum Electronics 2010, 16 (6), 1665 - 1675. 73. Steim, R.; Kogler, F. R.; Brabec, C. J., Interface materials for organic solar cells. Journal of Materials Chemistry 2010 , 20 (13), 2499 - 2512. 74. Liang, Z.; Zhang, Q.; Wiranwetchayan, O.; Xi, J.; Yang, Z.; Park, K.; Li, C.; Cao, G., Effects of the Morphology of a ZnO Buffer Layer on the Photovoltaic Performance of Inverted Polymer Solar Cells. Advanced Functional Materials 2012, 22 (10), 2194 - 2201. 160 75. Gadisa, A.; Liu, Y.; Samulski, E. T.; Lopez, R., Minimizing interfacial losses in inverted organic solar cells comprising Al - doped ZnO. Applied Physics Letters 2012, 100 (25), 253903. 76. Steim, R.; Choulis, S. A.; Schilinsky, P.; Brabec, C. J., Interface modification for highly efficient organic photovoltaics. Applied Physics Letters 2008, 92 (9), 093303. 77. Waldauf, C.; Morana, M.; Denk, P.; Schilinsky, P.; Coakley, K.; Choulis, S. A.; Brabec, C. J., Highly efficient inverte d organic photovoltaics using solution based titanium oxide as electron selective contact. Applied Physics Letters 2006, 89 (23), 233517. 78. Li, G.; Chu, C. W.; Shrotriya, V.; Huang, J.; Yang, Y., Efficient inverted polymer solar cells. Applied Physics Le tters 2006, 88 (25), 253503. 79. Liao, H. - H.; Chen, L. - M.; Xu, Z.; Li, G.; Yang, Y., Highly efficient inverted polymer solar cell by low temperature annealing of Cs2CO3 interlayer. Applied Physics Letters 2008, 92 (17), 173303. 80. Zhao, D. W.; Liu, P.; Su n, X. W.; Tan, S. T.; Ke, L.; Kyaw, A. K. K., An inverted organic solar cell with an ultrathin Ca electron - transporting layer and MoO3 hole - transporting layer. Applied Physics Letters 2009, 95 (15), 153304. 81. Janotti, A.; Van de Walle, C. G., Fundamental s of zinc oxide as a semiconductor. Reports on progress in physics 2009, 72 (12), 126501. 82. Huang, J.; Yin, Z.; Zheng, Q., Applications of ZnO in organic and hybrid solar cells. Energy & Environmental Science 2011, 4 (10), 3861 - 3877. 83. Hu, J.; Wu, Z.; Wei, H.; Song, T.; Sun, B., Effects of ZnO fabricating process on the performance of inverted organic solar cells. Organic Electronics 2012, 13 (7), 1171 - 1177. 84. Umeda, T.; Hashimoto, Y.; Mizukami, H.; Shirakawa, T.; Fujii, A.; Yoshino, K., Improvement o f characteristics on polymer photovoltaic cells composed of conducting polymer - fullerene systems. Synthetic Metals 2005, 152 (1), 93 - 96. 85. Schumann, S.; Da Campo, R.; Illy, B.; Cruickshank, A. C.; McLachlan, M. A.; Ryan, M. P.; Riley, D. J.; McComb, D. W.; Jones, T. S., Inverted organic photovoltaic devices with high efficiency and stability based on metal oxide charge extraction layers. Journal of Materials Chemistry 2011, 21 (7), 2381 - 2386. 86. Chang, C. - Y.; Tsai, F. - Y., Efficient and air - stable plast ics - based polymer solar cells enabled by atomic layer deposition. Journal of Materials Chemistry 2011, 21 (15), 5710 - 5715. 87. Kang, Y. - J.; Lim, K.; Jung, S.; Kim, D. - G.; Kim, J. - K.; Kim, C. - S.; Kim, S. H.; Kang, J. - W., Spray - coated ZnO electron transport layer for air - stable inverted organic solar cells. Solar Energy Materials and Solar Cells 2012, 96 (Supplement C), 137 - 140. 161 88. Cheng, Y. - J.; Hsieh, C. - H.; He, Y.; Hsu, C. - S.; Li, Y., Combination of Indene - C60 Bis - Adduct and Cross - Linked Fullerene Interlay er Leading to Highly Efficient Inverted Polymer Solar Cells. Journal of the American Chemical Society 2010, 132 (49), 17381 - 17383. 89. Chou, C. - H.; Kwan, W. L.; Hong, Z.; Chen, L. - M.; Yang, Y., A Metal - Oxide Interconnection Layer for Polymer Tandem Solar C ells with an Inverted Architecture. Advanced Materials 2011, 23 (10), 1282 - 1286. 90. White, M. S.; Olson, D. C.; Shaheen, S. E.; Kopidakis, N.; Ginley, D. S., Inverted bulk - heterojunction organic photovoltaic device using a solution - derived ZnO underlayer. Applied Physics Letters 2006, 89 (14), 143517. 91. Sekine, N.; Chou, C. - H.; Kwan, W. L.; Yang, Y., ZnO nano - ridge structure and its application in inverted polymer solar cell. Organic Electronics 2009, 10 (8), 1473 - 1477. 92. Takanezawa, K.; Tajima, K.; Ha shimoto, K., Efficiency enhancement of polymer photovoltaic devices hybridized with ZnO nanorod arrays by the introduction of a vanadium oxide buffer layer. Applied Physics Letters 2008, 93 (6), 063308. 93. Shin, K. - S.; Lee, K. - H.; Lee, H. H.; Choi, D.; Ki m, S. - W., Enhanced Power Conversion Efficiency of Inverted Organic Solar Cells with a Ga - Doped ZnO Nanostructured Thin Film Prepared Using Aqueous Solution. The Journal of Physical Chemistry C 2010, 114 (37), 15782 - 15785. 94. Hau, S. K.; Cheng, Y. - J.; Yip, H. - L.; Zhang, Y.; Ma, H.; Jen, A. K. Y., Effect of Chemical Modification of Fullerene - Based Self - Assembled Monolayers on the Performance of Inverted Polymer Solar Cells. ACS Applied Materials & Interfaces 2010, 2 (7), 1892 - 1902. 95. Kyaw, A. K. K.; Sun, X . W.; Jiang, C. Y.; Lo, G. Q.; Zhao, D. W.; Kwong, D. L., An inverted organic solar cell employing a sol - gel derived ZnO electron selective layer and thermal evaporated MoO3 hole selective layer. Applied Physics Letters 2008, 93 (22), 221107. 96. Yang, T.; Cai, W.; Qin, D.; Wang, E.; Lan, L.; Gong, X.; Peng, J.; Cao, Y., Solution - Processed Zinc Oxide Thin Film as a Buffer Layer for Polymer Solar Cells with an Inverted Device Structure. The Journal of Physical Chemistry C 2010, 114 (14), 6849 - 6853. 97. Sun, Y.; Seo, J. H.; Takacs, C. J.; Seifter, J.; Heeger, A. J., Inverted Polymer Solar Cells Integrated with a Low - Temperature - Annealed Sol - Gel - Derived ZnO Film as an Electron Transport Layer. Advanced Materials 2011, 23 (14), 1679 - 1683. 98. Sun, Y.; Gong, X.; Hsu, B. B. Y.; Yip, H. - L.; Jen, A. K. Y.; Heeger, A. J., Solution - processed cross - linkable hole selective layer for polymer solar cells in the inverted structure. Applied Physics Letters 2010, 97 (19), 193310. 99. Hsieh, C. - H.; Cheng, Y. - J.; Li, P. - J.; Che n, C. - H.; Dubosc, M.; Liang, R. - M.; Hsu, C. - S., Highly Efficient and Stable Inverted Polymer Solar Cells Integrated with a Cross - Linked Fullerene Material as an Interlayer. Journal of the American Chemical Society 2010, 132 (13), 4887 - 4893. 162 100. Amb, C. M. ; Chen, S.; Graham, K. R.; Subbiah, J.; Small, C. E.; So, F.; Reynolds, J. R., Dithienogermole As a Fused Electron Donor in Bulk Heterojunction Solar Cells. Journal of the American Chemical Society 2011, 133 (26), 10062 - 10065. 101. Liu, M. - Y.; Chang, C. - H. ; Chang, C. - H.; Tsai, K. - H.; Huang, J. - S.; Chou, C. - Y.; Wang, I. - J.; Wang, P. - S.; Lee, C. - Y.; Chao, C. - H.; Yeh, C. - L.; Wu, C. - I.; Lin, C. - F., Morphological evolution of the poly(3 - hexylthiophene)/[6,6] - phenyl - C61 - butyric acid methyl ester, oxidation of the silver electrode, and their influences on the performance of inverted polymer solar cells with a sol gel derived zinc oxide electron selective layer. Thin Solid Films 2010, 518 (17), 4964 - 4969. 102. Zhang, C.; You, H.; Lin, Z.; Hao, Y., Inverted organic p hotovoltaic cells with solution - processed zinc oxide as electron collecting layer. Japanese Journal of Applied Physics 2011, 50 (8R), 082302. 103. Braun, S.; Salaneck, W. R.; Fahlman, M., Energy Level Alignment at Organic/Metal and Organic/Organic Interfac es. Advanced Materials 2009, 21 (14 - 15), 1450 - 1472. 104. Alvarado, S. F.; Seidler, P. F.; Lidzey, D. G.; Bradley, D. D. C., Direct Determination of the Exciton Binding Energy of Conjugated Polymers Using a Scanning Tunneling Microscope. Physical Review Let ters 1998, 81 (5), 1082 - 1085. 105. Liang, Y.; Xu, Z.; Xia, J.; Tsai, S. - T.; Wu, Y.; Li, G.; Ray, C.; Yu, L., For the Bright Future Bulk Heterojunction Polymer Solar Cells with Power Conversion Efficiency of 7.4%. Advanced Materials 2010, 22 (20), E135 - E138 . 106. Jiang, C.; Lunt, R. R.; Duxbury, P. M.; Zhang, P. P., High - performance inverted solar cells with a controlled ZnO buffer layer. RSC Advances 2014, 4 (7), 3604 - 3610. 107. Liu, D.; Kelly, T. L., Perovskite solar cells with a planar heterojunction stru cture prepared using room - temperature solution processing techniques. Nature Photonics 2013, 8 , 133. 108. Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T., Organometal halide perovskites as visible - light sensitizers for photovoltaic cells. Journal of the American Chemical Society 2009, 131 (17), 6050 - 6051. 109. Best Research - Cell Efficiencies by National Renewable Energy Laboratory. http://www.nrel.gov/pv/assets/images/efficiency_char t.jpg 2016 . 110. Baikie, T.; Fang, Y.; Kadro, J. M.; Schreyer, M.; Wei, F.; Mhaisalkar, S. G.; Graetzel, M.; White, T. J., Synthesis and crystal chemistry of the hybrid perovskite (CH3NH3)PbI3 for solid - state sensitised solar cell applications. Journal of Materials Chemistry A 2013, 1 (18), 5628 - 5641. 111. Quarti, C.; Mosconi, E.; Ball, J. M.; D'Innocenzo, V.; Tao, C.; Pathak, S.; Snaith, H. J.; Petrozza, A.; De Angelis, F., Structural and optical properties of methylammonium lead iodide across the tetrago nal to cubic phase transition: implications for perovskite solar cells. Energy & Environmental Science 2016, 9 (1), 155 - 163. 163 112. Brivio, F.; Walker, A. B.; Walsh, A., Structural and electronic properties of hybrid perovskites for high - efficiency thin - film photovoltaics from first - principles. Apl Mater 2013, 1 (4), 042111. 113. Umebayashi, T.; Asai, K.; Kondo, T.; Nakao, A., Electronic structures of lead iodide based low - dimensional crystals. Physical Review B 2003, 67 (15), 155405. 114. Motta, C.; El - Mello uhi, F.; Kais, S.; Tabet, N.; Alharbi, F.; Sanvito, S., Revealing the role of organic cations in hybrid halide perovskite CH3NH3PbI3. Nature Communications 2015, 6 , 7026. 115. Hutter, E. M.; Gélvez - ranks, S. D.; Savenije, T. J., Direct indirect character of the bandgap in methylammonium lead iodide perovskite. Nature Materials 2016, 16 , 115. 116. Shi, D.; Adinolfi, V.; Comin, R.; Yuan, M.; Alarousu, E.; Buin, A.; Chen, Y.; Hoogland, S.; Rothenberger, A.; Katsiev, K.; Losovyj, Y.; Zhang, X.; Dowben, P. A.; Mohammed, O. F.; Sargent, E. H.; Bakr, O. M., Low trap - state density and long carrier diffusion in organolead trihalide perovskite single crystals. Science 2015, 347 (6221), 519 - 522. 117. Dong, Q.; F ang, Y.; Shao, Y.; Mulligan, P.; Qiu, J.; Cao, L.; Huang, J., Electron - hole diffusion lengths > - grown CH3NH3PbI3 single crystals. Science 2015, 347 (6225), 967 - 970. 118. Giorgi, G.; Fujisawa, J. - I.; Segawa, H.; Yamashita, K., Small Photocarrier Effective Masses Featuring Ambipolar Transport in Methylammonium Lead Iodide Perovskite : A Density Functional Analysis. The Journal of Physical Chemistry Letters 2013, 4 (24), 4213 - 4216. 119. Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.; Alcocer, M. J. P.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J., Electron - Hole Dif fusion Lengths Exceeding 1 Micrometer in an Organometal Trihalide Perovskite Absorber. Science 2013, 342 (6156), 341 - 344. 120. Chen, Y.; Peng, J.; Su, D.; Chen, X.; Liang, Z., Efficient and Balanced Charge Transport Revealed in Planar Perovskite Solar Cell s. ACS Applied Materials & Interfaces 2015, 7 (8), 4471 - 4475. 121. Yin, X.; Chen, P.; Que, M.; Xing, Y.; Que, W.; Niu, C.; Shao, J., Highly Efficient Flexible Perovskite Solar Cells Using Solution - Derived NiOx Hole Contacts. ACS Nano 2016, 10 (3), 3630 - 363 6. 122. Qin, P.; Tanaka, S.; Ito, S.; Tetreault, N.; Manabe, K.; Nishino, H.; Nazeeruddin, M. K.; Grätzel, M., Inorganic hole conductor - based lead halide perovskite solar cells with 12.4% conversion efficiency. Nature communications 2014, 5 , 3834. 123. Heo , J. H.; Im, S. H.; Noh, J. H.; Mandal, T. N.; Lim, C. - S.; Chang, J. A.; Lee, Y. H.; Kim, H. - j.; Sarkar, A.; Nazeeruddin, M. K.; Grätzel, M.; Seok, S. I., Efficient inorganic organic hybrid heterojunction solar cells containing perovskite compound and poly meric hole conductors. Nature Photonics 2013, 7 , 486. 164 124. Guo, Y.; Liu, C.; Inoue, K.; Harano, K.; Tanaka, H.; Nakamura, E., Enhancement in the efficiency of an organic inorganic hybrid solar cell with a doped P3HT hole - transporting layer on a void - free p erovskite active layer. Journal of Materials Chemistry A 2014, 2 (34), 13827 - 13830. 125. Eperon, G. E.; Burlakov, V. M.; Docampo, P.; Goriely, A.; Snaith, H. J., Morphological Control for High Performance, Solution - Processed Planar Heterojunction Perovskit e Solar Cells. Advanced Functional Materials 2014, 24 (1), 151 - 157. 126. Zhang, W.; Saliba, M.; Moore, D. T.; Pathak, S. K.; Hörantner, M. T.; Stergiopoulos, T.; Stranks, S. D.; Eperon, G. E.; Al exander - Webber, J. A.; Abate, A.; Sadhanala, A.; Yao, S.; Chen, Y.; Friend, R. H.; Estroff, L. A.; Wiesner, U.; Snaith, H. J., Ultrasmooth organic inorganic perovskite thin - film formation and crystallization for efficient planar heterojunction solar cells. Nature Communications 2015, 6 , 6142. 127. Moore, D. T.; Sai, H.; Tan, K. W.; Smilgies, D. - M.; Zhang, W.; Snaith, H. J.; Wiesner, U.; Estroff, L. A., Crystallization kinetics of organic inorganic trihalide perovskites and the role of the lead anion in crys tal growth. Journal of the American Chemical Society 2015, 137 (6), 2350 - 2358. 128. Liu, D.; Liu, C.; Wu, L.; Li, W.; Chen, F.; Xiao, B.; Zhang, J.; Feng, L., Highly reproducible perovskite solar cells with excellent CH3NH3PbI3 - xClx film morphology fabrica ted via high precursor concentration. RSC Advances 2016, 6 (56), 51279 - 51285. 129. Chang, J.; Zhu, H.; Xiao, J.; Isikgor, F. H.; Lin, Z.; Hao, Y.; Zeng, K.; Xu, Q. - H.; Ouyang, J., Enhancing the planar heterojunction perovskite solar cell performance throug h tuning the precursor ratio. Journal of Materials Chemistry A 2016, 4 (20), 7943 - 7949. 130. Jeon, N. J.; Noh, J. H.; Kim, Y. C.; Yang, W. S.; Ryu, S.; Seok, S. I., Solvent engineering for high - performance inorganic organic hybrid perovskite solar cells. N ature Materials 2014, 13 , 897. 131. Xiao, M.; Huang, F.; Huang, W.; Dkhissi, Y.; Zhu, Y.; Etheridge, J.; Gray - Weale, A.; Bach, U.; Cheng, Y. - B.; Spiccia, L., A Fast Deposition - Crystallization Procedure for Highly Efficient Lead Iodide Perovskite Thin - Film Solar Cells. Angewandte Chemie 2014, 126 (37), 10056 - 10061. 132. Burschka, J.; Pellet, N.; Moon, S. - J.; Humphry - Baker, R.; Gao, P.; Nazeeruddin, M. K.; Grätzel, M., Sequential deposition as a route to high - performance perovskite - sensitized solar cells. Nat ure 2013, 499 , 316. 133. Xiao, Z.; Bi, C.; Shao, Y.; Dong, Q.; Wang, Q.; Yuan, Y.; Wang, C.; Gao, Y.; Huang, J., Efficient, high yield perovskite photovoltaic devices grown by interdiffusion of solution - processed precursor stacking layers. Energy & Environ mental Science 2014, 7 (8), 2619 - 2623. 134. Wu, Y.; Islam, A.; Yang, X.; Qin, C.; Liu, J.; Zhang, K.; Peng, W.; Han, L., Retarding the crystallization of PbI2 for highly reproducible planar - structured perovskite solar cells via sequential deposition. Energ y & Environmental Science 2014, 7 (9), 2934 - 2938. 165 135. Chen, Q.; Zhou, H.; Hong, Z.; Luo, S.; Duan, H. - S.; Wang, H. - H.; Liu, Y.; Li, G.; Yang, Y., Planar Heterojunction Perovskite Solar Cells via Vapor - Assisted Solution Process. Journal of the American Che mical Society 2014, 136 (2), 622 - 625. 136. Yang, W. S.; Noh, J. H.; Jeon, N. J.; Kim, Y. C.; Ryu, S.; Seo, J.; Seok, S. I., High - performance photovoltaic perovskite layers fabricated through intramolecular exchange. Science 2015, 348 (6240), 1234 - 1237. 137 . Xiao, Z.; Dong, Q.; Bi, C.; Shao, Y.; Yuan, Y.; Huang, J., Solvent Annealing of Perovskite - Induced Crystal Growth for Photovoltaic - Device Efficiency Enhancement. Advanced Materials 2014, 26 (37), 6503 - 6509. 138. Chen, W.; Wu, Y.; Yue, Y.; Liu, J.; Zhang, W.; Yang, X.; Chen, H.; Bi, E.; Ashraful, I.; Grätzel, M.; Han, L., Efficient and stable large - area perovskite solar cells with inorganic charge extraction layers. Science 2015, 350 (6263), 944 - 948. 139. Kim, J.; Yun, J. S.; Cho, Y.; Lee, D. S.; Wilkinson , B.; Soufiani, A. M.; Deng, X.; Zheng, J.; Shi, A.; Lim, S.; Chen, S.; Hameiri, Z.; Zhang, M.; Lau, C. F. J.; Huang, S.; Green, M. A.; Ho - Baillie, A. W. Y., Overcoming the Challenges of Large - Area High - Efficiency Perovskite Solar Cells. ACS Energy Letters 2017, 2 (9), 1978 - 1984. 140. Huang, J.; Yuan, Y.; Shao, Y.; Yan, Y., Understanding the physical properties of hybrid perovskites for photovoltaic applications. Nature Reviews Materials 2017, 2 , 17042. 141. De Wolf, S.; Holovsky, J.; Moon, S. - J.; Löper, P. ; Niesen, B.; Ledinsky, M.; Haug, F. - J.; Yum, J. - H.; Ballif, C., Organometallic Halide Perovskites: Sharp Optical Absorption Edge and Its Relation to Photovoltaic Performance. The Journal of Physical Chemistry Letters 2014, 5 (6), 1035 - 1039. 142. Adinolfi, V.; Peng, W.; Walters, G.; Bakr, O. M.; Sargent, E. H., The Electrical and Optical Properties of Organometal Halide Perovskites Relevant to Optoelectronic Performance. Advanced Materials 2018, 30 (1), 1700764. 143. Lin, Q.; Armin, A.; Nagiri, R. C. R.; Bu rn, P. L.; Meredith, P., Electro - optics of perovskite solar cells. Nature Photonics 2014, 9 , 106. 144. Nah, S.; Spokoyny, B.; Stoumpos, C.; Soe, C. M. M.; Kanatzidis, M.; Harel, E., Spatially segregated free - carrier and exciton populations in individual le ad halide perovskite grains. Nature Photonics 2017, 11 , 285. 145. Xing, G.; Mathews, N.; Sun, S.; Lim, S. S.; Lam, Y. M.; Grätzel, M.; Mhaisalkar, S.; Sum, T. C., Long - Range Balanced Electron - and Hole - Transport Lengths in Organic - Inorganic CH3NH3PbI3. Sci ence 2013, 342 (6156), 344 - 347. 146. Adhyaksa, G. W. P.; Veldhuizen, L. W.; Kuang, Y.; Brittman, S.; Schropp, R. E. I.; Garnett, E. C., Carrier Diffusion Lengths in Hybrid Perovskites: Processing, Composition, Aging, and Surface Passivation Effects. Chemis try of Materials 2016, 28 (15), 5259 - 5263. 166 147. Feng, J.; Xiao, B., Effective Masses and Electronic and Optical Properties of Nontoxic MASnX3 (X = Cl, Br, and I) Perovskite Structures as Solar Cell Absorber: A Theoretical Study Using HSE06. The Journal of Physical Chemistry C 2014, 118 (34), 19655 - 19660. 148. Herz, L. M., Charge - Carrier Mobilities in Metal Halide Perovskites: Fundamental Mechanisms and Limits. ACS Energy Letters 2017, 2 (7), 1539 - 1548. 149. Shargaieva, O.; Lang, F.; Rappich, J.; Dittrich, T .; Klaus, M.; Meixner, M.; Genzel, C.; Nickel, N. H., Influence of the Grain Size on the Properties of CH3NH3PbI3 Thin Films. ACS Applied Materials & Interfaces 2017, 9 (44), 38428 - 38435. 150. Watthage, S. C.; Song, Z.; Shrestha, N.; Phillips, A. B.; Liyan age, G. K.; Roland, P. J.; Ellingson, R. J.; Heben, M. J., Enhanced Grain Size, Photoluminescence, and Photoconversion Efficiency with Cadmium Addition during the Two - Step Growth of CH3NH3PbI3. ACS Applied Materials & Interfaces 2017, 9 (3), 2334 - 2341. 151 . Chen, J.; Song, J.; Huang, F.; Li, H.; Liu, S.; Wang, M.; Shen, Y., The Role of Synthesis Parameters on Crystallization and Grain Size in Hybrid Halide Perovskite Solar Cells. The Journal of Physical Chemistry C 2017, 121 (32), 17053 - 17061. 152. Wang, Y. ; Zhang, Y.; Zhang, P.; Zhang, W., High intrinsic carrier mobility and photon absorption in the perovskite CH3NH3PbI3. Physical Chemistry Chemical Physics 2015, 17 (17), 11516 - 11520. 153. Handa, T.; Tex, D. M.; Shimazaki, A.; Wakamiya, A.; Kanemitsu, Y., C harge Injection Mechanism at Heterointerfaces in CH3NH3PbI3 Perovskite Solar Cells Revealed by Simultaneous Time - Resolved Photoluminescence and Photocurrent Measurements. The Journal of Physical Chemistry Letters 2017, 8 (5), 954 - 960. 154. Shi, L. X.; Wang , Z. S.; Huang, Z.; Sha, W. E. I.; Wang, H.; Zhou, Z., The effects of interfacial recombination and injection barrier on the electrical characteristics of perovskite solar cells. AIP Advances 2018, 8 (2), 025312. 155. Jiménez - López, J.; Cambarau, W.; Cabau , L.; Palomares, E., Charge Injection, Carriers Recombination and HOMO Energy Level Relationship in Perovskite Solar Cells. Scientific Reports 2017, 7 (1), 6101. 156. Zhou, H.; Chen, Q.; Li, G.; Luo, S.; Song, T. - b.; Duan, H. - S.; Hong, Z.; You, J.; Liu, Y. ; Yang, Y., Interface engineering of highly efficient perovskite solar cells. Science 2014, 345 (6196), 542 - 546. 157. Li, Y.; Zhao, Y.; Chen, Q.; Yang, Y.; Liu, Y.; Hong, Z.; Liu, Z.; Hsieh, Y. - T.; Meng, L.; Li, Y.; Yang, Y., Multifunctional Fullerene Deri vative for Interface Engineering in Perovskite Solar Cells. Journal of the American Chemical Society 2015, 137 (49), 15540 - 15547. 158. Abate, A.; Saliba, M.; Hollman, D. J.; Stranks, S. D.; Wojciechowski, K.; Avolio, R.; Grancini, G.; Petrozza, A.; Snaith, H. J., Supramolecular Halogen Bond Passivation of Organic Inorganic Halide Perovskite Solar Cells. Nano Letters 2014, 14 (6), 3247 - 3254. 167 159. Noel, N. K.; Abate, A.; Stranks, S. D.; Parrott, E. S.; Burlakov, V. M.; Goriely, A.; Snaith, H. J., Enhanced Pho toluminescence and Solar Cell Performance via Lewis Base Passivation of Organic Inorganic Lead Halide Perovskites. ACS Nano 2014, 8 (10), 9815 - 9821. 160. Xing, G.; Mathews, N.; Lim, S. S.; Yantara, N.; Liu, X.; Sabba, D.; Grätzel, M.; Mhaisalkar, S.; Sum, T. C., Low - temperature solution - processed wavelength - tunable perovskites for lasing. Nature Materials 2014, 13 , 476. 161. Kim, J.; Lee, S. - H.; Lee, J. H.; Hong, K. - H., The Role of Intrinsic Defects in Methylammonium Lead Iodide Perovskite. The Journal of P hysical Chemistry Letters 2014, 5 (8), 1312 - 1317. 162. Yin, W. - J.; Shi, T.; Yan, Y., Unusual defect physics in CH3NH3PbI3 perovskite solar cell absorber. Applied Physics Letters 2014, 104 (6), 063903. 163. Wetzelaer, G. - J. A. H.; Scheepers, M.; Sempere, A. M.; Momblona, C.; Ávila, J.; Bolink, H. J., Trap - Assisted Non - Radiative Recombination in Organic Inorganic Perovskite Solar Cells. Advanced Materials 2015, 27 (11), 1837 - 1841. 164. Stranks, S. D.; Burlakov, V. M.; Leijtens, T.; Ball, J. M.; Goriely, A.; S naith, H. J., Recombination Kinetics in Organic - Inorganic Perovskites: Excitons, Free Charge, and Subgap States. Physical Review Applied 2014, 2 (3), 034007. 165. Yun, J. S.; Ho - Baillie, A.; Huang, S.; Woo, S. H.; Heo, Y.; Seidel, J.; Huang, F.; Cheng, Y. - B.; Green, M. A., Benefit of Grain Boundaries in Organic Inorganic Halide Planar Perovskite Solar Cells. The Journal of Physical Chemistry Letters 2015, 6 (5), 875 - 880. 166. Chu, Z.; Yang, M.; Schulz, P.; Wu, D.; Ma, X.; Seifert, E.; Sun, L.; Li, X.; Zhu, K.; Lai, K., Impact of grain boundaries on efficiency and stability of organic - inorganic trihalide perovskites. Nature Communications 2017, 8 (1), 2230. 167. Li, J. - J.; Ma, J. - Y.; Ge, Q. - Q.; Hu, J. - S.; Wang, D.; Wan, L. - J., Microscopic Investigation of Gra in Boundaries in Organolead Halide Perovskite Solar Cells. ACS Applied Materials & Interfaces 2015, 7 (51), 28518 - 28523. 168. deQuilettes, D. W.; Jariwala, S.; Burke, S.; Ziffer, M. E.; Wang, J. T. W.; Snaith, H. J.; Ginger, D. S., Tracking Photoexcited Ca rriers in Hybrid Perovskite Semiconductors: Trap - Dominated Spatial Heterogeneity and Diffusion. ACS Nano 2017, 11 (11), 11488 - 11496. 169. Mamun, A. A.; Ava, T. T.; Jeong, H. J.; Jeong, M. S.; Namkoong, G., A deconvoluted PL approach to probe the charge car rier dynamics of the grain interior and grain boundary of a perovskite film for perovskite solar cell applications. Physical Chemistry Chemical Physics 2017, 19 (13), 9143 - 9148. 170. MacDonald, G. A.; Yang, M.; Berweger, S.; Killgore, J. P.; Kabos, P.; Ber ry, J. J.; Zhu, K.; DelRio, F. W., Methylammonium lead iodide grain boundaries exhibit depth - dependent electrical properties. Energy & Environmental Science 2016, 9 (12), 3642 - 3649. 168 171. Tvingstedt, K.; Gil - Escrig, L.; Momblona, C.; Rieder, P.; Kiermasch, D.; Sessolo, M.; Baumann, A.; Bolink, H. J.; Dyakonov, V., Removing Leakage and Surface Recombination in Planar Perovskite Solar Cells. ACS Energy Letters 2017, 2 (2), 424 - 430. 172. Hou, Y.; Chen, W.; Baran, D.; Stubhan, T.; Luechinger, N. A.; Hartmeier, B .; Richter, M.; Min, J.; Chen, S.; Quiroz, C. O. R.; Li, N.; Zhang, H.; Heumueller, T.; Matt, G. J.; Osvet, A.; Forberich, K.; Zhang, Z. - G.; Li, Y.; Winter, B.; Schweizer, P.; Spiecker, E.; Brabec, C. J., Overcoming the Interface Losses in Planar Heterojun ction Perovskite - Based Solar Cells. Advanced Materials 2016, 28 (25), 5112 - 5120. 173. Correa - Baena, J. - P.; Tress, W.; Domanski, K.; Anaraki, E. H.; Turren - Cruz, S. - H.; Roose, B.; Boix, P. P.; Gratzel, M.; Saliba, M.; Abate, A.; Hagfeldt, A., Identifying an d suppressing interfacial recombination to achieve high open - circuit voltage in perovskite solar cells. Energy & Environmental Science 2017, 10 (5), 1207 - 1212. 174. Tan, H.; Jain, A.; Voznyy, O.; Lan, X.; García de Arquer, F. P.; Fan, J. Z.; Quintero - Bermu dez, R.; Yuan, M.; Zhang, B.; Zhao, Y.; Fan, F.; Li, P.; Quan, L. N.; Zhao, Y.; Lu, Z. - H.; Yang, Z.; Hoogland, S.; Sargent, E. H., Efficient and stable solution - processed planar perovskite solar cells via contact passivation. Science 2017, 355 (6326), 722 - 726. 175. Cao, J.; Wu, B.; Chen, R.; Wu, Y.; Hui, Y.; Mao, B. - W.; Zheng, N., Efficient, Hysteresis - Free, and Stable Perovskite Solar Cells with ZnO as Electron - Transport Layer: Effect of Surface Passivation. Advanced Materials , 1705596. 176. Peng, J.; Wu, Y.; Ye, W.; Jacobs, D. A.; Shen, H.; Fu, X.; Wan, Y.; Duong, T.; Wu, N.; Barugkin, C.; Nguyen, H. T.; Zhong, D.; Li, J.; Lu, T.; Liu, Y.; Lockrey, M. N.; Weber, K. J.; Catchpole, K. R.; White, T. P., Interface passivation using ultrathin polymer - fullerene films for high - efficiency perovskite solar cells with negligible hysteresis. Energy & Environmental Science 2017, 10 (8), 1792 - 1800. 177. Yang, G.; Wang, C.; Lei, H.; Zheng, X.; Qin, P.; Xiong, L.; Zhao, X.; Yan, Y.; Fang, G., Interface engineering in plan ar perovskite solar cells: energy level alignment, perovskite morphology control and high performance achievement. Journal of Materials Chemistry A 2017, 5 (4), 1658 - 1666. 178. Snaith, H. J.; Abate, A.; Ball, J. M.; Eperon, G. E.; Leijtens, T.; Noel, N. K. ; Stranks, S. D.; Wang, J. T. - W.; Wojciechowski, K.; Zhang, W., Anomalous Hysteresis in Perovskite Solar Cells. The Journal of Physical Chemistry Letters 2014, 5 (9), 1511 - 1515. 179. .; Barnes, P. R. F., Evidence for ion migration in hybrid perovskite solar cells with minimal hysteresis. Nature Communications 2016, 7 , 13831. 180. transport in hybr id lead iodide perovskite solar cells. Nature Communications 2015, 6 , 7497. 181. Levine, I.; Nayak, P. K.; Wang, J. T. - W.; Sakai, N.; Van Reenen, S.; Brenner, T. M.; Mukhopadhyay, S.; Snaith, H. J.; Hodes, G.; Cahen, D., Interface - Dependent Ion 169 Migration/A ccumulation Controls Hysteresis in MAPbI3 Solar Cells. The Journal of Physical Chemistry C 2016, 120 (30), 16399 - 16411. 182. Tao, C.; Neutzner, S.; Colella, L.; Marras, S.; Srimath Kandada, A. R.; Gandini, M.; Bastiani, M. D.; Pace, G.; Manna, L.; Caironi, M.; Bertarelli, C.; Petrozza, A., 17.6% stabilized efficiency in low - temperature processed planar perovskite solar cells. Energy & Environmental Science 2015, 8 (8), 2365 - 2370. 183. Shao, Y.; Xiao, Z.; Bi, C.; Yuan, Y.; Huang, J., Origin and elimination o f photocurrent hysteresis by fullerene passivation in CH3NH3PbI3 planar heterojunction solar cells. Nature Communications 2014, 5 , 5784. 184. Kim, H. - S.; Park, N. - G., Parameters Affecting I V Hysteresis of CH3NH3PbI3 Perovskite Solar Cells: Effects of Pero vskite Crystal Size and Mesoporous TiO2 Layer. The Journal of Physical Chemistry Letters 2014, 5 (17), 2927 - 2934. 185. Tress, W.; Marinova, N.; Moehl, T.; Zakeeruddin, S. M.; Nazeeruddin, M. K.; Gratzel, M., Understanding the rate - dependent J - V hysteresis, slow time component, and aging in CH3NH3PbI3 perovskite solar cells: the role of a compensated electric field. Energy & Environmental Science 2015, 8 (3), 995 - 1004. 186. De Bastiani, M.; Dell'Erba, G.; Gandini, M.; D'Innocenzo, V.; Neutzner, S.; Kandada, A. R. S.; Grancini, G.; Binda, M.; Prato, M.; Ball, J. M.; Caironi, M.; Petrozza, A., Solar Cells: Ion Migration and the Role of Preconditioning Cycles in the Stabilization of the J V Characteristics of Inverted Hybrid Perovskite Solar Cells (Adv. Energy M ater. 2/2016). Advanced Energy Materials 2016, 6 (2) , 1501453 . 187. Frost, J. M.; Butler, K. T.; Walsh, A., Molecular ferroelectric contributions to anomalous hysteresis in hybrid perovskite solar cells. Apl Mater 2014, 2 (8), 081506. 188. Sherkar, T. S.; Jan Anton Koster, L., Can ferroelectric polarization explain the high performance of hybrid halide perovskite solar cells? Physical Chemistry Chemical Physics 2016, 18 (1), 331 - 338. 189. Chen, B.; Yang, M.; Priya, S.; Zhu, K., Origin of J V Hysteresis in P erovskite Solar Cells. The Journal of Physical Chemistry Letters 2016, 7 (5), 905 - 917. 190. Nazeeruddin, M. K.; Zakeeruddin, S. M.; Rothlisberger, U.; Graetzel, M., Ionic polarization - induced current voltage hysteresis in CH3NH3PbX3 perovskite solar cells. Nature Communications 2016, 7 , 10334. 191. Yuan, Y.; Chae, J.; Shao, Y.; Wang, Q.; Xiao, Z.; Centrone, A.; Huang, J., Photovoltaic Switching Mechanism in Lateral Structure Hybrid Perovskite Solar Cells. Advanced Energy Materials 2015, 5 (15), 1500615. 170 192. Li, C.; Guerrero, A.; Zhong, Y.; Gräser, A.; Luna, C. A. M.; Köhler, J.; Bisquert, J.; Hildner, R.; Huettner, S., Real - Time Observation of Iodide Ion Migration in Methylammonium Lead Halide Perovskites. Small 2017, 13 (42), 1701711. 193. Shao, Y.; Fang, Y .; Li, T.; Wang, Q.; Dong, Q.; Deng, Y.; Yuan, Y.; Wei, H.; Wang, M.; Gruverman, A.; Shield, J.; Huang, J., Grain boundary dominated ion migration in polycrystalline organic - inorganic halide perovskite films. Energy & Environmental Science 2016, 9 (5), 175 2 - 1759. 194. Yun, J. S.; Seidel, J.; Kim, J.; Soufiani, A. M.; Huang, S.; Lau, J.; Jeon, N. J.; Seok, S. I.; Green, M. A.; Ho - Baillie, A., Critical Role of Grain Boundaries for Ion Migration in Formamidinium and Methylammonium Lead Halide Perovskite Solar Cells. Advanced Energy Materials 2016, 6 (13), 1600330. 195. Wu, F.; Yue, X.; Song, Q.; Zhu, L., Replacing PbI2 by MAPbI3 to Realize Large Grain Size and Reduced Hysteresis for Highly Efficient Perovskite Solar Cells. Solar RRL 2018, 2 (1), 1700147. 196. K e, W.; Xiao, C.; Wang, C.; Saparov, B.; Duan, H. - S.; Zhao, D.; Xiao, Z.; Schulz, P.; Harvey, S. P.; Liao, W.; Meng, W.; Yu, Y.; Cimaroli, A. J.; Jiang, C. - S.; Zhu, K.; Al - Jassim, M.; Fang, G.; Mitzi, D. B.; Yan, Y., Employing Lead Thiocyanate Additive to R educe the Hysteresis and Boost the Fill Factor of Planar Perovskite Solar Cells. Advanced Materials 2016, 28 (26), 5214 - 5221. 197. Yang, D.; Zhou, X.; Yang, R.; Yang, Z.; Yu, W.; Wang, X.; Li, C.; Liu, S.; Chang, R. P. H., Surface optimization to eliminate hysteresis for record efficiency planar perovskite solar cells. Energy & Environmental Science 2016, 9 (10), 3071 - 3078. 198. Petrus, M. L.; Schlipf, J.; Li, C.; Gujar, T. P.; Giesbrecht, N.; Müller - Buschbaum, P.; Thelakkat, M.; Bein, T.; Hüttner, S.; Doca mpo, P., Capturing the Sun: A Review of the Challenges and Perspectives of Perovskite Solar Cells. Advanced Energy Materials 2017, 7 (16), 1700264. 199. Yang, S.; Fu, W.; Zhang, Z.; Chen, H.; Li, C. - Z., Recent advances in perovskite solar cells: efficiency , stability and lead - free perovskite. Journal of Materials Chemistry A 2017, 5 (23), 11462 - 11482. 200. Correa - Baena, J. - P.; Saliba, M.; Buonassisi, T.; Grätzel, M.; Abate, A.; Tress, W.; Hagfeldt, A., Promises and challenges of perovskite solar cells. Scie nce 2017, 358 (6364), 739 - 744. 201. Han, Y.; Meyer, S.; Dkhissi, Y.; Weber, K.; Pringle, J. M.; Bach, U.; Spiccia, L.; Cheng, Y. - B., Degradation observations of encapsulated planar CH3NH3PbI3 perovskite solar cells at high temperatures and humidity. Journa l of Materials Chemistry A 2015, 3 (15), 8139 - 8147. 202. K.; Ng, A. M. C., Encapsulation of Perovskite Solar Cells for High Humidity Conditions. ChemSusChem 2016, 9 (18), 2597 - 2603. 171 203. Bush, K. A.; Palmstrom, A. F.; Y u, Z. J.; Boccard, M.; Cheacharoen, R.; Mailoa, J. P.; McMeekin, D. P.; Hoye, R. L. Z.; Bailie, C. D.; Leijtens, T.; Peters, I. M.; Minichetti, M. C.; Rolston, N.; Prasanna, R.; Sofia, S.; Harwood, D.; Ma, W.; Moghadam, F.; Snaith, H. J.; Buonassisi, T.; H olman, Z. C.; Bent, S. F.; McGehee, M. D., 23.6% - efficient monolithic perovskite/silicon tandem solar cells with improved stability. Nature Energy 2017, 2 , 17009. 204. Sun, Y.; Peng, J.; Chen, Y.; Yao, Y.; Liang, Z., Triple - cation mixed - halide perovskites: towards efficient, annealing - free and air - stable solar cells enabled by Pb(SCN)2 additive. Scientific Reports 2017, 7 , 46193. 205. Saliba, M.; Matsui, T.; Seo, J. - Y.; Domanski, K.; Correa - Baena, J. - P.; Nazeeruddin, M. K.; Zakeeruddin, S. M.; Tress, W.; Ab ate, A.; Hagfeldt, A.; Gratzel, M., Cesium - containing triple cation perovskite solar cells: improved stability, reproducibility and high efficiency. Energy & Environmental Science 2016, 9 (6), 1989 - 1997. 206. Domanski, K.; Correa - Baena, J. - P.; Mine, N.; Na zeeruddin, M. K.; Abate, A.; Saliba, M.; Tress, W.; Hagfeldt, A.; Grätzel, M., Not All That Glitters Is Gold: Metal - Migration - Induced Degradation in Perovskite Solar Cells. ACS Nano 2016, 10 (6), 6306 - 6314. 207. Abate, A.; Paek, S.; Giordano, F.; Correa - Ba ena, J. - P.; Saliba, M.; Gao, P.; Matsui, T.; Ko, J.; Zakeeruddin, S. M.; Dahmen, K. H.; Hagfeldt, A.; Gratzel, M.; Nazeeruddin, M. K., Silolothiophene - linked triphenylamines as stable hole transporting materials for high efficiency perovskite solar cells. Energy & Environmental Science 2015, 8 (10), 2946 - 2953. 208. Gregg, B. A.; Hanna, M. C., Comparing organic to inorganic photovoltaic cells: Theory, experiment, and simulation. Journal of Applied Physics 2003, 93 (6), 3605 - 3614. 209. Clarke, T. M.; Durrant, J. R., Charge Photogeneration in Organic Solar Cells. Chemical Reviews 2010, 110 (11), 6736 - 6767. 210. Richter, T. V.; Braun, C. H.; Link, S.; Scheuble, M.; Crossland, E. J. W.; Stelzl, F.; Würfel, U.; Ludwigs, S., Regioregular Polythiophenes with Alkylth iophene Side Chains. Macromolecules 2012, 45 (14), 5782 - 5788. 211. Deibel, C.; Mack, D.; Gorenflot, J.; Schöll, A.; Krause, S.; Reinert, F.; Rauh, D.; Dyakonov, V., Energetics of excited states in the conjugated polymer poly(3 - hexylthiophene). Physical Rev iew B 2010, 81 (8), 085202. 212. Tamai, Y.; Ohkita, H.; Benten, H.; Ito, S., Exciton Diffusion in Conjugated Polymers: From Fundamental Understanding to Improvement in Photovoltaic Conversion Efficiency. The Journal of Physical Chemistry Letters 2015, 6 (1 7), 3417 - 3428. 213. Dimitrakopoulos, C. D.; Mascaro, D. J., Organic thin - film transistors: A review of recent advances. IBM Journal of Research and Development 2001, 45 (1), 11 - 27. 214. Yan, L.; Gao, Y., Interfaces in organic semiconductor devices. Thin So lid Films 2002, 417 (1), 101 - 106. 172 215. Cahen, D.; Kahn, A., Electron Energetics at Surfaces and Interfaces: Concepts and Experiments. Advanced Materials 2003, 15 (4), 271 - 277. 216. Veenstra, S. C.; Jonkman, H. T., Energy - level alignment at metal organic an d organic organic interfaces. Journal of Polymer Science Part B: Polymer Physics 2003, 41 (21), 2549 - 2560. 217. Cravino, A.; Schilinsky, P.; Brabec, C. J., Characterization of Organic Solar Cells: the Importance of Device Layout. Advanced Functional Materi als 2007, 17 (18), 3906 - 3910. 218. Hu, Z.; Zhang, J.; Hao, Z.; Zhao, Y., Influence of doped PEDOT:PSS on the performance of polymer solar cells. Solar Energy Materials and Solar Cells 2011, 95 (10), 2763 - 2767. 219. Konios, D.; Kakavelakis, G.; Petridis, C. ; Savva, K.; Stratakis, E.; Kymakis, E., Highly efficient organic photovoltaic devices utilizing work - function tuned graphene oxide derivatives as the anode and cathode charge extraction layers. Journal of Materials Chemistry A 2016, 4 (5), 1612 - 1623. 220. Stratakis, E.; Savva, K.; Konios, D.; Petridis, C.; Kymakis, E., Improving the efficiency of organic photovoltaics by tuning the work function of graphene oxide hole transporting layers. Nanoscale 2014, 6 (12), 6925 - 6931. 221. intensities in ultrathin organic double - heterostructure photovoltaic diodes. Applied Physics Letters 2000, 76 (19), 2650 - 2652. 222. Lee, J.; Park, S.; Lee, Y.; Kim, H.; Shin, D.; Jeong, J.; Jeong, K.; Cho, S. W.; Lee, H.; Yi, Y., Electron transport mechanism of bathocuproine exciton blocking layer in organic photovoltaics. Physical Chemistry Chemical Physics 2016, 18 (7), 5444 - 5452. 223. Zhang, F. J.; Zhao, D. W.; Zhu o, Z. L.; Wang, H.; Xu, Z.; Wang, Y. S., Inverted small molecule organic solar cells with Ca modified ITO as cathode and MoO3 modified Ag as anode. Solar Energy Materials and Solar Cells 2010, 94 (12), 2416 - 2421. 224. Wang, M.; Li, Y.; Huang, H.; Peterson, E. D.; Nie, W.; Zhou, W.; Zeng, W.; Huang, W.; Fang, G.; Sun, N.; Zhao, X.; Carroll, D. L., Thickness dependence of the MoO3 blocking layers on ZnO nanorod - inverted organic photovoltaic devices. Applied Physics Letters 2011, 98 (10), 103305. 225. Mang, A. ; Reimann, K.; Rübenacke, S., Band gaps, crystal - field splitting, spin - orbit coupling, and exciton binding energies in ZnO under hydrostatic pressure. Solid State Communications 1995, 94 (4), 251 - 254. 226. Özgür, Ü.; Alivov, Y. I.; Liu, C.; Teke, A.; Reshc Cho, S. - J.; Morkoç, H., A comprehensive review of ZnO materials and devices. Journal of Applied Physics 2005, 98 (4), 041301. 227. Look, D. C., Recent advances in ZnO materials and devices. Materials Science and Engine ering: B 2001, 80 (1), 383 - 387. 173 228. MacLeod, B. A.; Tremolet de Villers, B. J.; Schulz, P.; Ndione, P. F.; Kim, H.; Giordano, A. J.; Zhu, K.; Marder, S. R.; Graham, S.; Berry, J. J.; Kahn, A.; Olson, D. C., Stability of inverted organic solar cells with Z nO contact layers deposited from precursor solutions. Energy & Environmental Science 2015, 8 (2), 592 - 601. 229. Chang, Y. - M.; Leu, C. - Y., Conjugated polyelectrolyte and zinc oxide stacked structure as an interlayer in highly efficient and stable organic ph otovoltaic cells. Journal of Materials Chemistry A 2013, 1 (21), 6446 - 6451. 230. Ohyama, M.; Kouzuka, H.; Yoko, T., Sol - gel preparation of ZnO films with extremely preferred orientation along (002) plane from zinc acetate solution. Thin solid films 1997, 3 06 (1), 78 - 85. 231. Klahr, B. M.; Martinson, A. B. F.; Hamann, T. W., Photoelectrochemical Investigation of Ultrathin Film Iron Oxide Solar Cells Prepared by Atomic Layer Deposition. Langmuir 2011, 27 (1), 461 - 468. 232. Binnig, G.; Quate, C. F.; Gerber, C. , Atomic Force Microscope. Physical Review Letters 1986, 56 (9), 930 - 933. 233. Jalili, N.; Laxminarayana, K., A review of atomic force microscopy imaging systems: application to molecular metrology and biological sciences. Mechatronics 2004, 14 (8), 907 - 94 5. 234. Almora, O.; Zarazua, I.; Mas - Marza, E.; Mora - Sero, I.; Bisquert, J.; Garcia - Belmonte, G., Capacitive Dark Currents, Hysteresis, and Electrode Polarization in Lead Halide Perovskite Solar Cells. The Journal of Physical Chemistry Letters 2015, 6 (9), 1645 - 1652. 235. Guerrero, A.; Garcia - Belmonte, G.; Mora - Sero, I.; Bisquert, J.; Kang, Y. S.; Jacobsson, T. J.; Correa - Baena, J. - P.; Hagfeldt, A., Properties of Contact and Bulk Impedances in Hybrid Lead Halide Perovskite Solar Cells Including Inductive Lo op Elements. The Journal of Physical Chemistry C 2016, 120 (15), 8023 - 8032. 236. Bisquert, J.; Bertoluzzi, L.; Mora - Sero, I.; Garcia - Belmonte, G., Theory of Impedance and Capacitance Spectroscopy of Solar Cells with Dielectric Relaxation, Drift - Diffusion T ransport, and Recombination. The Journal of Physical Chemistry C 2014, 118 (33), 18983 - 18991. 237. Bisquert, J., Chemical capacitance of nanostructured semiconductors: its origin and significance for nanocomposite solar cells. Physical Chemistry Chemical P hysics 2003, 5 (24), 5360 - 5364. 238. Bassiri - Gharb, N.; Bastani, Y.; Bernal, A., Chemical solution growth of ferroelectric oxide thin films and nanostructures. Chemical Society Reviews 2014, 43 (7), 2125 - 2140. 239. Schwartz, R. W.; Schneller, T.; Waser, R. , Chemical solution deposition of electronic oxide films. Comptes Rendus Chimie 2004, 7 (5), 433 - 461. 174 240. Zuo, L.; Dong, S.; De Marco, N.; Hsieh, Y. - T.; Bae, S. - H.; Sun, P.; Yang, Y., Morphology Evolution of High Efficiency Perovskite Solar Cells via Vapo r Induced Intermediate Phases. Journal of the American Chemical Society 2016, 138 (48), 15710 - 15716. 241. Rong, Y.; Venkatesan, S.; Guo, R.; Wang, Y.; Bao, J.; Li, W.; Fan, Z.; Yao, Y., Critical kinetic control of non - stoichiometric intermediate phase tran sformation for efficient perovskite solar cells. Nanoscale 2016, 8 (26), 12892 - 12899. 242. Kang, R.; Kim, J. - E.; Yeo, J. - S.; Lee, S.; Jeon, Y. - J.; Kim, D. - Y., Optimized Organometal Halide Perovskite Planar Hybrid Solar Cells via Control of Solvent Evaporat ion Rate. The Journal of Physical Chemistry C 2014, 118 (46), 26513 - 26520. 243. Choi, J. J.; Yang, X.; Norman, Z. M.; Billinge, S. J. L.; Owen, J. S., Structure of Methylammonium Lead Iodide Within Mesoporous Titanium Dioxide: Active Material in High - Perfo rmance Perovskite Solar Cells. Nano Letters 2014, 14 (1), 127 - 133. 244. Zarazua, I.; Han, G.; Boix, P. P.; Mhaisalkar, S.; Fabregat - Santiago, F.; Mora - Seró, I.; Bisquert, J.; Garcia - Belmonte, G., Surface Recombination and Collection Efficiency in Perovskit e Solar Cells from Impedance Analysis. The Journal of Physical Chemistry Letters 2016, 7 (24), 5105 - 5113. 245. Pockett, A.; Eperon, G. E.; Sakai, N.; Snaith, H. J.; Peter, L. M.; Cameron, P. J., Microseconds, milliseconds and seconds: deconvoluting the dyn amic behaviour of planar perovskite solar cells. Physical Chemistry Chemical Physics 2017, 19 (8), 5959 - 5970. 246. Contreras - Bernal, L.; Salado, M.; Todinova, A.; Calio, L.; Ahmad, S.; Idígoras, J.; Anta, J. A., Origin and Whereabouts of Recombination in P erovskite Solar Cells. The Journal of Physical Chemistry C 2017, 121 (18), 9705 - 9713. 247. Correa - Baena, J. - P.; Turren - Cruz, S. - H.; Tress, W.; Hagfeldt, A.; Aranda, C.; Shooshtari, L.; Bisquert, J.; Guerrero, A., Changes from Bulk to Surface Recombination Mechanisms between Pristine and Cycled Perovskite Solar Cells. ACS Energy Letters 2017, 2 (3), 681 - 688. 248. Richardson, G.; O'Kane, S. E. J.; Niemann, R. G.; Peltola, T. A.; Foster, J. M.; Cameron, P. J.; Walker, A. B., Can slow - moving ions explain hyster esis in the current - voltage curves of perovskite solar cells? Energy & Environmental Science 2016, 9 (4), 1476 - 1485. 249. Walsh, A.; Scanlon, D. O.; Chen, S.; Gong, X. G.; Wei, S. H., Self - Regulation Mechanism for Charged Point Defects in Hybrid Halide Per ovskites. Angewandte Chemie International Edition 2015, 54 (6), 1791 - 1794. 250. Marin - Beloqui, J. M.; Lanzetta, L.; Palomares, E., Decreasing Charge Losses in Perovskite Solar Cells Through mp - TiO2/MAPI Interface Engineering. Chemistry of Materials 2016, 2 8 (1), 207 - 213. 251. Wang, L.; Moghe, D.; Hafezian, S.; Chen, P.; Young, M.; Elinski, M.; Martinu, L.; Kéna - Cohen, S.; Lunt, R. R., Alkali Metal Halide Salts as Interface Additives to Fabricate Hysteresis - 175 Free Hybrid Perovskite - Based Photovoltaic Devices. ACS Applied Materials & Interfaces 2016, 8 (35), 23086 - 23094. 252. Sun, J.; Pimcharoen, K.; Wagner, S. R.; Duxbury, P. M.; Zhang, P., Nanoscale imaging of dense fiber morphology and local electrical response in conductive regioregular poly(3 - hexylthiophene ). Organic Electronics 2014, 15 (2), 441 - 448. 253. Tian, W.; Cui, R.; Leng, J.; Liu, J.; Li, Y.; Zhao, C.; Zhang, J.; Deng, W.; Lian, T.; Jin, S., Limiting Perovskite Solar Cell Performance by Heterogeneous Carrier Extraction. Angewandte Chemie Internation al Edition 2016, 55 (42), 13067 - 13071. 254. Juarez - Perez, E. J.; Sanchez, R. S.; Badia, L.; Garcia - Belmonte, G.; Kang, Y. S.; Mora - Sero, I.; Bisquert, J., Photoinduced Giant Dielectric Constant in Lead Halide Perovskite Solar Cells. The Journal of Physical Chemistry Letters 2014, 5 (13), 2390 - 2394. 255. Kim, J.; Kim, G.; Kim, T. K.; Kwon, S.; Back, H.; Lee, J.; Lee, S. H.; Kang, H.; Lee, K., Efficient planar - heterojunction perovskite solar cells achieved via interfacial modification of a sol - gel ZnO electro n collection layer. Journal of Materials Chemistry A 2014, 2 (41), 17291 - 17296. 256. Sandberg, O. J.; Sundqvist, A.; Nyman, M.; Österbacka, R., Relating Charge Transport, Contact Properties, and Recombination to Open - Circuit Voltage in Sandwich - Type Thin - F ilm Solar Cells. Physical Review Applied 2016, 5 (4), 044005. 257. Zhang, B.; Zhang, M. J.; Pang, S. P.; Huang, C. S.; Zhou, Z. M.; Wang, D.; Wang, N.; Cui, G. L., Carrier Transport in CH3NH3PbI3 Films with Different Thickness for Perovskite Solar Cells. A dvanced Materials Interfaces 2016, 3 (17), 1600327. 258. Khoram, P.; Brittman, S.; Dzik, W. I.; Reek, J. N. H.; Garnett, E. C., Growth and Characterization of PDMS - Stamped Halide Perovskite Single Microcrystals. The Journal of Physical Chemistry C 2016, 12 0 (12), 6475 - 6481. 259. Dingley, D. J.; Randle, V., Microtexture determination by electron back - scatter diffraction. Journal of Materials Science 1992, 27 (17), 4545 - 4566. 260. Kronik, L.; Cahen, D.; Schock, H. W., Effects of Sodium on Polycrystalline Cu(I n,Ga)Se2 and Its Solar Cell Performance. Advanced Materials 1998, 10 (1), 31 - 36. 261. Fuertes Marrón, D.; Sadewasser, S.; Meeder, A.; Glatzel, T.; Lux - Steiner, M. C., Electrical activity at grain boundaries of Cu ( In,Ga)Se2 thin films. Physical Review B 200 5, 71 (3), 033306. 262. Persson, C.; Zunger, A., Compositionally induced valence - band offset at the grain boundary of polycrystalline chalcopyrites creates a hole barrier. Applied Physics Letters 2005, 87 (21), 211904. 263. Yan, Y.; Jiang, C. S.; Noufi, R. ; Wei, S. - H.; Moutinho, H. R.; Al - Jassim, M. M., Electrically Benign Behavior of Grain Boundaries in Polycrystalline CuInSe 2 Films. Physical Review Letters 2007, 99 (23), 235504. 176 264. Haruyama, J.; Sodeyama, K.; Han, L.; Tateyama, Y., Termination Dependenc e of Tetragonal CH3NH3PbI3 Surfaces for Perovskite Solar Cells. The Journal of Physical Chemistry Letters 2014, 5 (16), 2903 - 2909. 265. Shi, T.; Yin, W. - J.; Hong, F.; Zhu, K.; Yan, Y., Unipolar self - doping behavior in perovskite CH3NH3PbBr3. Applied Physic s Letters 2015, 106 (10), 103902. 266. Meyer, J.; Shu, A.; Kröger, M.; Kahn, A., Effect of contamination on the electronic structure and hole - injection properties of MoO3/organic semiconductor interfaces. Applied Physics Letters 2010, 96 (13), 133308. 267. Bouzidi, K.; Chegaar, M.; Bouhemadou, A., Solar cells parameters evaluation considering the series and shunt resistance. Solar Energy Materials and Solar Cells 2007, 91 (18), 1647 - 1651. 268. Nunes, P.; Fortunato, E.; Martins, R., Influence of the post - tre atment on the properties of ZnO thin films. Thin Solid Films 2001, 383 (1), 277 - 280. 269. Ishii, H.; Hayashi, N.; Ito, E.; Washizu, Y.; Sugi, K.; Kimura, Y.; Niwano, M.; Ouchi, Y.; Seki, K., Kelvin probe study of band bending at organic semiconductor/metal interfaces: examination of Fermi level alignment. physica status solidi (a) 2004, 201 (6), 1075 - 1094. 270. Lange, I.; Blakesley, J. C.; Frisch, J.; Vollmer, A.; Koch, N.; Neher, D., Band Bending in Conjugated Polymer Layers. Physical Review Letters 2011, 106 (21), 216402. 271. Kim, J.; Kim, G.; Choi, Y.; Lee, J.; Park, S. H.; Lee, K., Light - soaking issue in polymer solar cells: Photoinduced energy level alignment at the sol - gel processed metal oxide and indium tin oxide interface. Journal of Applied Physic s 2012, 111 (11), 114511. 272. Verbakel, F.; Meskers, S. C. J.; Janssen, R. A. J., Electronic memory effects in diodes from a zinc oxide nanoparticle - polystyrene hybrid material. Applied Physics Letters 2006, 89 (10), 102103. 273. Ke, C.; Gang, C.; Shujie, W.; Linsong, L.; Shuxi, D.; Xingtang, Z.; Bingsuo, Z.; Zuliang, D., Surface states dominative Au Schottky contact on vertical aligned ZnO nanorod arrays synthesized by low - temperature growth. New Journal of Physics 2007, 9 (7 ), 214. 274. Natsume, Y.; Sakata, H., Zinc oxide films prepared by sol - gel spin - coating. Thin Solid Films 2000, 372 (1), 30 - 36. 275. Reese, M. O.; White, M. S.; Rumbles, G.; Ginley, D. S.; Shaheen, S. E., Optimal negative electrodes for poly(3 - hexylthiophe ne): [6,6] - phenyl C61 - butyric acid methyl ester bulk heterojunction photovoltaic devices. Applied Physics Letters 2008, 92 (5), 053307. 276. Son, D. - Y.; Lee, J. - W.; Choi, Y. J.; Jang, I. - H.; Lee, S.; Yoo, P. J.; Shin, H.; Ahn, N.; Choi, M.; Kim, D.; Park, N. - G., Self - formed grain boundary healing layer for highly efficient CH3 NH3 PbI3 perovskite solar cells. Nature Energy 2016, 1 , 16081. 177 277. Yuan, Y.; Huang, J., Ion Migration in Organometal Trihalide Perovskite and Its Impact on Photovoltaic Efficiency an d Stability. Accounts of Chemical Research 2016, 49 (2), 286 - 293.