No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

A thesis submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of PhD. August 9, 2017 / Nano-structured carbon lms, long studied due to the promise of exceptional quantum transport properties, present a signi cant problem in condensed matter due to the disorder which inherently forms in these materials. This work addresses the role of structural disorder in low dimensional carbon systems. The in uence of structural disorder on resonant transmission is studied in diamond-like carbon superlattices. Having established a model for disorder, this model for the structural changes is then applied to interpret experimental measurements of diamond-like carbon superlattices. The role of phonons on resonant transmission under a high frequency gate potential was also studied. This model for structural disorder in heterogeneous carbon lms was then applied to disordered superconductors close to the Anderson-Mott transition using the inhomogeneous Bogoliubov-de Gennes theory. This analysis is then used in support of experimental work to understand the superconductor-insulator transition in boron doped nano-crystalline diamond lms. Coherent quantum transport e ects were demonstrated in structurally-disordered diamondlike carbon (DLC) superlattices through distinct current modulation (step-like features) with negative differential resistance in the current-voltage (I-V) measurements. A model for these structurally disordered superlattices was developed using tight-binding calculations within the Landauer-B uttiker formalism assuming a random variation of the hopping integral following a Gaussian distribution. Calculations of the I-V characteristics for different con gurations of superlattices compliment the interpretation of the measured I-V characteristics and illustrate that while these DLC superlattice structures do not behave like conventional superlattices, the present model can be used to tailor the properties of future devices. Furthermore this tandem theoretical and experimental analysis establishes the validity of the model for structural disorder. The same model for the variation of disorder was then applied to interpret the electronic transport properties of disordered graphene-like carbon thin films. The influence of disorder on the activation energy in few layer graphitic lms was modelled and compared with experimental observations through collaboration. The lms, grown by laser ablation, allowed the speci c e ects of structural disorder in the sp2 - C phase to be probed. Defects acted as effective barriers resulting in localization of charge carriers. Electron transmission spectra, calculated with a tight-binding model, accounted for the change of localization length as a result of disorder in the sp2 - C phase. This theoretical study showed that the localization length of the thin graphitic lms can be tuned with the level of disorder and was shown to be consistent with experimental studies. The in uence of nitrogen incorporation on resonant transmission in DLC superlattices was then studied theoretically. This study illuminated the speci c role of the nitrogen potential in relation to the Fermi level (EF ) in nitrogen incorporated amorphous carbon (a- CN) superlattice structures. In a-CN systems, the variation of conductivity with nitrogen percentage has been found to be strongly non-linear due to the change of disorder level. The e ect of correlated carbon and nitrogen disorder was investigated in conjunction with the nitrogen potential through analysis of transmission spectra, calculated using a tight binding model, which showed two broad peaks related to these species. It was shown that the characteristic transmission time through nitrogen centres can be controlled through a combination of the nitrogen potential and correlated disorder. In particular, by controlling the arrangement of the nitrogen sites within the sp2 - C clusters as well as their energetic position relative to EF , a crossover of the pronounced transmission peaks of nitrogen and carbon sites can be achieved. Furthermore, it was shown that nitrogen incorporated as a potential barrier can also enhance the transmission in the a-CN superlattice structures. The strong non-linear variation of resistance and the characteristic time of the structures can explain the transport features observed experimentally in a-CN fi lms. This analysis was then partnered with measurements performed on nitrogen-incorporated carbon superlattices (N-DLC QSL) by Neeraj Dwivedi (National University of Singapore). The electrical characteristics of these nitrogen incorporated superlattice devices revealed prominent negative di erential resistance (NDR) behavior. The interpretation of these measurements was supported by 1D tight binding calculations of disordered superlattice structures (chains), which included signi cant bond alternation in sp3-hybridized regions. This analysis showed improved resonant transmission, which can be ascribed to nitrogendriven structural modi cation of the N-DLC QSL structures, especially the increased sp2-C clustering that provides additional conduction paths throughout the network. In order to determine the in uence of additional factors on coherent quantum states in molecular systems as an extension to the analysis on superlattices, a theoretical study of the electron-phonon interaction in double barrier structures under the in uence of a timedependent gate potential was undertaken. The Floquet theory was employed along with expansion in a polaron eigenbasis to render a multi-dimensional single body problem. An essentially exact solution was found using the Riccati matrix technique. It was demonstrated that optimal transmission can be achieved by varying the frequency of the gate potential. In addition, it was shown that the gate potential can be used to control the energy of the resonant states very precisely while maintaining optimal transmission. Having gained a deep understanding of the structural changes induced in carbon systems through the incorporation of nitrogen, a similar structural model was then applied to study the changes induced in diamond and nanocrystalline fi lms by boron incorpora- tion. Boron doped diamond provides an interesting superconductor with ongoing debate surrounding the nature of the impurity band and the effect on the superconducting phase transition of structural changes induced by boron incorporation. The in uence of disorder, both structural (non-diagonal) and on-site (diagonal), was studied through the inhomogeneous Bogoliubov-de Gennes (BdG) theory in narrow-band disordered superconductors with a view towards understanding superconductivity in boron doped diamond (BDD) and boron-doped nanocrystalline diamond (B-NCD) lms. We employed the attractive Hubbard model within the mean eld approximation, including a short range Coulomb interaction between holes in the narrow acceptor band. We studied substitutional boron incorporation in a triangular lattice, with disorder in the form of random potential uctuations at the boron sites. The role of structural disorder was investigated through non-uniform variation of the tight-binding coupling parameter where, following experimental ndings in BDD and B-NCD lms, we incorporated the concurrent increase in structural disorder with increasing boron concentration. Stark differences between the ffects of structural and on-site disorder were demonstrated and showed that structural disorder has a much greater e ect on the density of states, mean pairing amplitude and super uid density than on-site potential disorder. We showed that structural disorder can increase the mean pairing amplitude while the spectral gap in the density of states decreases, with states eventually appearing within the spectral gap for high levels of disorder. This study illustrated how the effects of structural disorder can explain some of the features found in superconducting BDD and B-NCD lms, such as a tendency towards saturation of the critical temperature (Tc) with boron doping and deviations from the expected Bardeen-Cooper-Shrie er (BCS) theory in the temperature dependence of the pairing amplitude and spectral gap. The variation of the super uid density considering only structural disorder was markedly different from the variation with on-site disorder only and revealed that structural disorder is far more detrimental to superconductivity and accounts for the relatively low Tc of BDD and B-NCD in comparison to the Tc predicted using the conventional BCS theory. This theoretical work was then used to interpret features in the measured transport properties of B-NCD lms with di erent doping concentrations and microstructures. The temperature dependence of a distinct local maximum in eld dependent magnetoresistance measurements showed suppression of the density of states as the system breaks up into superconducting regions separated by grain boundaries. Differential resistance measurements at different temperatures and magnetic fi elds showed a transition from a local minimum at zero applied current, indicative of persisting superconducting regions, to a local maximum. A power law dependence over a certain current range in the measured I-V characteristics at di erent magnetic elds suggests a Berezinski-Kosterlitz-Thouless (BKT) transition. In addition, features in the magnetoresistance clearly indicate additional phases. Together with features in current-voltage measurements, these signatures show the coexistence of superconductivity and additional competing phases close to the Anderson-Mott transition. / LG2018

銅銦鎵硒薄膜太陽能電池是一種清潔、環保的發電技術。 最近，銅銦鎵硒太陽能電池實現了20.9%的光電轉換效率，超出了多晶硅太陽能電池所保持的20.4%的薄膜太陽能電池的最高紀錄。 多種技術改進促成了這項薄膜太陽能電池的新紀錄。 其中一種重要改進是將1 到2 微米厚的硫化鎘硫化鋅混合窗口層替換成薄層硫化鎘和摻鋁氧化鋅透明導電層。 / 基於本實驗室在生長高質量銅銦鎵硒吸收層的先進技術，本工作重點研究了位於吸收層和透明窗口層之間的緩衝層和高阻窗口層。 這兩層的常規結構是由化學水浴法生長的硫化鎘層和本征氧化鋅層組成。 本論文的第一部分是關於這種常規結構的參數優化。 經過優化，本實驗室實現了在小型組件（總面積60 平方釐米）上15.6%的最高轉換效率。 / 本論文的第二部分關於用化學水浴法生長緩衝層。 我們發展了一種新型生長制備，用於避免氣泡和孔洞在吸收層表面的形成。 表面形貌測試結果顯示，使用此種設備生長的緩衝層能均勻的覆蓋銅銦鎵硒吸收層的表面。 其它硫化鎘的生長參數也根據新設備的特點進行了優化。 優化結果顯示，在空間電荷區的復合對電池轉換效率影響較大，而這種復合損失可以經過調整緩衝層與吸收能之間能帶結構得到減少。 我們研究了另外一種用化學水浴法生長的緩衝層：硫化鋅。 硫化鋅是一種無毒的寬禁帶材料，在短波部分有較少的光吸收。因此，它是一種很好的硫化鎘替代物。 我們研究了在不同生長溫度下的生長動力學機制。 最優的生長溫度是95 攝氏度。 經過生長結束後的退火過程，硫化鋅的禁帶寬度由3.61eV 下降到3.2eV。 再經過在氧氣環境中的退火，禁帶寬度可由3.2eV 繼續下降到2.9eV。 在單結電池中，硫化鋅的最優厚度在43 納米到62 納米之間。 在此厚度範圍中，具有硫化鋅緩衝層的電池實現了相對於具有硫化鎘緩衝層的電池更高的轉換效率。硫化鋅電池實現了與硫化鎘電池相近的開路電壓。 此項改進主要是由於在高溫條件下生長的硫化鋅與銅銦鎵硒層形成了更合適的能帶結構。 / 本論文的第三部分是關於用共濺射的方法生長鋅鎂氧化物緩衝層。 實驗結果顯示，鋅鎂氧化物的晶體結構和禁帶寬度與鎂含量相關。 當鎂含量小於0.4 時，鋅鎂氧化物具有(002)從優方位的纖鋅礦結構。 晶體質量隨鎂含量的增加而降低，同時，鋅鎂氧化物的禁帶寬度隨鎂含量的增加線性增加。 對於濺射方法生長的緩衝層，吸收層的表面鈍化對提高轉化效率非常重要。 / 本論文的最後一部分是關於高阻窗口層的研究。 相比於由本征氧化鋅構成的高阻窗口層，由鋅鎂氧化物構成的高阻窗口層能使電池有更優的穩定性。對於單結電池，本層的最優厚度是50 納米。對於小型組件，最優厚度在100 納米左右。 關於鎂的最優組分，結果仍爭議，但可以確定的是由較高濺射功率（大於2.2 瓦每平方釐米）產生的濺射損傷是應當盡量避免的。關於光照產生的亞穩定性的研究表明，亞穩定性強度與濺射環境中的氧氣含量正相關。 相對於無氧氣摻雜的電池，通過將1%的氧氣摻入氬氣濺射環境中，電池效率提高了0.5 個百分點。 / Cu(In,Ga)Se2 (CIGS)-based thin film solar cells have been regarded as a promising technology for cheap and environmentally friendly electricity generation. CIGS based solar cell has achieved 20.9% conversion effciency, while the offcial record for multicrystalline Silicon solar cell is 20.4%. A series of improvements have lead to this record for thin film based solar cell. An important improvement originated from the replacement of 1- to 2-um-thick doped (Cd,Zn)S layer by a thin, undoped CdS and a transparent conductive oxide(TCO). / Based on our techniques on growing high quality CIGS absorber layer, this work focuses on further optimization of buffer layer and high resistance window layer located between the CIGS absorber and the TCO window layer. The standard buffer structure includes a chemical-bath-deposited CdS layer and an intrinsic ZnO layer. The first part of this thesis is about optimization of this standard structure carried out in our laboratory. The best conversion effciency achieved on mini-module with total area of 60 cm² is 15.6 %. / The second part is about the fabrication of alternative buffer layers by chemical bath deposition. New deposition equipment has been invented to eliminate stationary bubbles and uncovered pinholes on absorber surface in the deposition of CBD CdS. Surface morphology studies shown that the buffer layer grown by this equipment has uniform coverage on the CIGS surface. Other deposition parameters in the chemical bath deposition of CdS buffer layer have been systematically studied employing this new equipment. Our results suggest that the detrimental effect of recombination in SCR region can be mitigated by proper band alignment in the buffer/absorber interface. / Another buffer layer grown by CBD method is ZnS. Because the wider bandgap and less light absorption in short wavelength range, ZnS is a good candidate to replace the toxic CdS buffer layer. The growth kinetics under different deposition temperature have been studied. The optimal temperature profile has been achieved by setting temperature at 95°C. The results of post annealing after deposition indicate that the bandgap energy of CBD ZnS decreases from 3.61 eV to 3.2 eV by annealing in vacuum. A further decrease from 3.2 eV to 2.9 eV could be caused by annealing with oxygen gas. The optimum thickness of ZnS used in single solar cells is between 43nm and 62nm. In this range, devices with CBD-ZnS buffer layer have achieved higher conversion effciency than CBD-CdS buffer layer solar cell. The open circuit voltage for ZnS-buffer devices has approached the value with CdS-buffer. The improvement is mainly due to proper band alignment of ZnS/CIGS interface achieved under high deposition temperature of CBD process. / The third part of this thesis is to study how to deposit (Zn,Mg)O buffer layers by co-sputtering method. It was found that the crystalline structure and optical bandgap of sputtered (Zn,Mg)O varies with Mg concentration. (Zn,Mg)O thin films with Mg concentration less than 0.4 have preferential orientation with a wurtzite phase (002). The crystal quality decreases with increasing Mg concentration and the band gap of the (Zn,Mg)O films has a linear relationship with the Mg concentration in this range. An interesting finding to emerge from this study is that oxygen passivation of absorber surface is critical to improve device performance with (Zn,Mg)O buffer layer deposited by sputtering method. / The last chapter assesses the effect of replacing high resistance window layer with (Zn,Mg)O in devices with CBD-ZnS buffer layer. Compared to devices with i-ZnO (high-resistance window) HRW layer, better device stability has been confirmed on solar cells with (Zn,Mg)O HRW layer. For single cells, the optimum thickness of HRW layer is about 50 nm, and the optimum thickness for mini-modules is around 100nm. Although no conclusion can be drawn with the optimum Mg concentration, the sputtering damage caused by sputtering power density higher than 2.2 W/cm² should be avoided. It was also shown that the metastability effect activated by illumination has positive correlation with the number of energetic oxygen ions in sputtering process. Compared to devices without oxygen doping, a higher effciency (increase of 0.5 % unit) has been achieved by the oxygen/argon doping ratio of 1 %. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Zhu, Jiakuan = 基於銅銦鎵硒薄膜太陽能電池的緩衝層結構研究 / 朱家寬. / Thesis (Ph.D.) Chinese University of Hong Kong, 2014. / Includes bibliographical references (leaves 121-134). / Abstracts also in Chinese. / Zhu, Jiakuan = Ji yu tong yin jia xi bo mo tai yang neng dian chi de huan chong ceng jie gou yan jiu / Zhu Jiakuan.

by Cheung-Yin Tang. / Title also in Chinese. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1991. / Bibliography: leaves 83-84. / ACKNOWLEDGEMENTS --- p.i / TABLE OF CONTENTS --- p.ii / ABSTRACT --- p.iv / Chapter Chapter 1 - --- Introduction --- p.1 / Chapter 1.1 --- Structure --- p.2 / Chapter 1.1.1 --- Physical Structure --- p.2 / Chapter 1.1.2 --- Electronic Structure --- p.3 / Chapter 1.2 --- Hydrogenation --- p.9 / Chapter 1.2.1 --- Hydrogenation during film formation --- p.10 / Chapter 1.2.2 --- Posthydrogenation --- p.10 / Chapter 1.3 --- Doping of a-Si --- p.11 / Chapter 1.4 --- Previous Results and Applications --- p.13 / Chapter 1.4.1 --- Results --- p.13 / Chapter 1.4.2 --- Applications --- p.24 / Chapter Chapter 2 - --- Experimental Set-up and Techniques --- p.25 / Chapter 2.1 --- Sample Preparation --- p.25 / Chapter 2.1.1 --- Substrate cleaning procedure --- p.25 / Chapter 2.1.2 --- Deposition Method --- p.26 / Chapter 2.1.3 --- Annealing Method --- p.30 / Chapter 2.1.4 --- Hydrogenation Method --- p.31 / Chapter 2.1.5 --- Doping Method --- p.33 / Chapter 2.2 --- Measurements --- p.34 / Chapter 2.2.1 --- Dark Conductivity --- p.34 / Chapter 2.2.2 --- Room Temperature Photo-conductivity --- p.39 / Chapter 2.2.3 --- ESR (Electron Spin Resonance) --- p.39 / Chapter Chapter 3 - --- Results and Discussions --- p.41 / Chapter 3.1 --- Doping effect and posthydrogenation --- p.42 / Chapter 3.2 --- Annealing of the doped films --- p.44 / Chapter 3.3 --- Implantation at different dose levels --- p.46 / Chapter Chapter 4 - --- Conclusions --- p.82 / REFERENCES --- p.83 / APPENDIX --- p.85