Дисертації з теми "Thermal interdiffusion"

Multicomponent metal alloys play an essential role in many technologies and their properties must be optimized by rational selection of the alloy’s components and its fractional composition of each. High-throughput materials synthesis allows us to prepare Composition Spread Alloy Films (CSAFs), sample libraries that contains all possible compositions of a binary or ternary alloy. In our lab, a Rotatable Shadow Mask (RSM) – CSAF deposition tool has been developed for the creation of CSAFs. Such CSAFs can be prepared with composition gradients and/or thickness gradients in arbitrarily controlled directions and on a variety of substrates. Once prepared, the CSAF libraries can be characterized thoroughly using a variety of highthroughput spectroscopic methods. Their bulk composition is mapped across the library using Energy Dispersive X-ray spectroscopy (EDX). The near-surface compositions are mapped across composition space using X-ray Photoemission Spectroscopy (XPS). Finally, the electronic structure can be mapped using UV photoemission spectroscopy (UPS) and valence band XPS. Once characterized, these CSAFs are being used for high-throughput studies of alloy catalysis and thermal properties of the alloys and of alloy-substrate interfaces. First of all, PdzCu1-z CSAF was prepared to show that alloy nanoparticles (aNPs) and thin films can adopt phases that differ from those of the corresponding bulk alloy. The mapping of XPS-derived core level binding energy shifts across PdzCu1-z SCSNaP library shows a promising result that the FCC phase can be dimensionally stabilized over the composition range where B2 phase exists in the bulk. This observation can potentially improve the performance of PdzCu1-z NP catalysts in H2 separation. Secondly, the relationship between catalyst activity-electronic structure-composition has been investigated. A high throughput characterization of electronic structure (valence band energy) of binary PdxAg1-x and ternary PdxCuyAu1-x-y CSAFs were performed by XPS. This XPS-derived valence band center is compared with UPS-derived data across PdxCuyAu1-x-y CSAFs. In addition, H2-D2 exchange reaction was studied on PdxAg1-x CASF. A higher HD formation rate is experimentally observed but cannot be predicted by the Langmuir-Hinshelwood model when the surface coverage is saturated. However, the proposed H2-D2 exchange mechanism (breakthrough model) involved with surface and subsurface hydrogen reaction is investigated to produce a same reaction order as Langmuir-Hinshelwood mechanism, which cannot explain the experimental observation. Furthermore, the thermal interface conductance (G) was studied as a function of metal alloy composition. A high-throughput approach to preparation, characterization, and measurement of G was also demonstrated to study the thermal property of alloyed materials. Our result in studying the G across the AuxY1-x (Y = Pd and Cu) CSAFs-dielectric interfaces has shown a linear relationship with alloy composition, which monotonically increases with decreasing Au (at. %). Lastly, the effect of interdiffusion in metal films on G at metal-dielectric interface was also examined. The XPS depth profiling was designed to experimentally determine the temperature effect on compositional profiles in the Au-Cu system, and how to further influence G. This study provides fundamental understanding of stability of adhesion layer of Cu and the effect of interdiffusion in Cu-Au alloy on the heat dissipation. All in all, the key value to these CSAF libraries is that they enable measurement of important alloy properties across entire binary or ternary alloy composition spaces, without the need to prepare and characterize numerous discrete composition samples.

Superalloys are widely applied as materials for components in the hot section of gas turbines. As superalloys have a limited oxidation life, the application of a coating is vital. The most commonly applied coatings in stationary gas turbines are MCrAlY coatings. Since the turbine components are exposed to high cyclic thermal stresses, MCrAlY coatings must also show a high thermal fatigue resistance. In this thesis, the effect of Cobalt and Rhenium on microstructure, oxidation and thermal fatigue of NiCoCrAlY coatings is presented. Additionally the condition of the coatings after testing in an industrial gas turbine is shown. The influence of Cobalt and Rhenium on coating microstructure was investigated by thermodynamic modelling and by metallography. It could be shown that both elements reduce the γ`-phase fraction and increase the β-phase fraction owing to an expansion of the γ+β field in the phase diagram. Modelling showed that Rhenium promotes the formation of α-Cr, which could be explained by a shift of the α-Cr solvus to higher temperatures and lower Cr concentrations. In the real coatings Re causes the precipitation of TCP-phase. The oxide scale growth rate is increased by Cobalt and Rhenium and it appears that Yttrium plays a significant role for that effect. Coating consumption due to simultaneous oxidation and interdiffusion could be decreased by the application of Cobalt and Rhenium. In thermal fatigue testing Rhenium reduces the time to crack initiation and increases crack propagation rate, although it could be shown that Rhenium increases the creep resistance of the coating. The effect could be explained by the influence of Rhenium on the microstructure, which increases creep resistance, but also reduces the ductility of the coating.

A haute température, l’interdiffusion entre un superalliage et son revêtement protecteur (ß-NiAl ou ß- NiPtAl) dégrade à la fois la protection contre l’oxydation, par modification de la composition chimique du revêtement, et la microstructure du superalliage (3ième et 4ième générations) par formation de Zones de Réaction Secondaires (SRZ). Le but de cette étude a donc été (1) de développer des barrières de diffusion (BD) constituées d’une dense précipitation de phases a-W après traitement sous vide (BD simple) ou chromisation en phase vapeur (BD enrichie en chrome) (2) de mettre au point une méthode pour en étudier l’efficacité. Des mesures de concentration chimique (à partir de cartographies spectrales EDS), couplées à des ajustements des comportements en oxydation cyclique en utilisant le modèle « p-kp », et le développement d’un modèle « p-kp-ß » ont permis de montrer l’efficacité de la BD selon sa composition et la durée de vieillissement. Pour des longues durées de vieillissement, l’efficacité de la BD se réduit par la dissolution des précipités d’a-W dans les phases y’ et y formées à cause de la dégradation des propriétés protectrices du revêtement ß NiPtAl (augmentation de l’écaillage de l’oxyde formé et de la cinétique d’oxydation). Plusieurs causes probables de cette dégradation ont pu être déterminées, soit dues aux procédés (pollution au soufre) soit liées à la mise en place de la BD : augmentation de la transformation martensitique, enrichissement en tungstène et présence de précipités d’alpha chrome. Enfin, il a été montré que si l’initiation des SRZ est modifiée par l’ajout de la BD, leur cinétique de propagation ne l’est pas et est essentiellement dépendante de la composition de l’alliage. Un modèle de propagation des SRZ décrivant les évolutions chimiques locales de part et d’autres de l’interface « SRZ / superalliage » a été proposé. L’ajout de chrome à la BD permet d’inhiber la formation des SRZ, une couche riche en phases TCP remplace alors la SRZ
At high temperature, interdiffusion between a superalloy and its protective coating (ß-NiAl or ß- NiPtAl) degrades the oxidation protection by modifying the chemical composition of the coating. It also degrades the 3rd et 4th generation superalloy microstructure due to the formation of Secondary Reaction Zones (SRZ). As a consequence, the aim of this study was (1) to develop diffusion barriers (DB) composed of a dense precipitation of a-W phases after a thermal treatment under vacuum (simple DB) or a vapour phase chromisation (Cr enriched DB), (2) to develop a method for quantifying the DB efficiency. Chemical concentration measurements (with EDS spectral maps) coupled with the « p-kp » modelling of the cyclic oxidation kinetics, and the development of the model « p-kp-ß » have permitted to study DB efficiency as a function of its composition and its high temperature ageing. For long ageing duration, the efficiency of the DB is reduced. Indeed, it is shown that the DB degrades the protection character of the ß-NiPtAl by increasing the oxide scale spallation and of its growth kinetic. This, in turns, accelerates the ß to y’ and y phases transformation and then increases the a-W precipitates dissolution. Some likely causes of this degradation have been determined, either due to the process (sulphur pollution) or intrinsic of the DB addition (increase of the martensitic transformation, enrichment in tungsten and a-Cr formation in the coating). Finally, it has been proved that DB addition modifies the SRZ initiation but not their propagation kinetic, which only depends on the superalloy local composition. A SRZ propagation model which describes local chemical evolutions on both sides of the « SRZ / superalloy » interface was proposed. The addition of chromium to the DB permits to inhibit the SRZ formation. In this case, a layer rich in TCP platelets replaces the SRZ

Ce travail concerne la caracterisation structurale et l'etude de l'interdiffusion dans les superreseaux contraints cdte/cdznte et cdte/hgte, elabores par epitaxie par jets moleculaires, au moyen de la double diffraction de rayons x. Les parametres structuraux (composition, epaisseur, etat de deformation) sont obtenus de facon precise en utilisant conjointement l'experience et la simulation en theorie cinematique. Les coefficients d'interdiffusion des couples cd/zn (400c) et cd/hg (200c) sont mesures par diffraction x. Ils permettent de calculer les profils de concentration des superreseaux en fonction de la temperature et du temps de recuit. Par extrapolation, nous precisons les conditions de croissance pour obtenir des interfaces raides, qui sont necessaires pour des dispositifs optoelectroniques performants

Afin de poursuivre la miniaturisation des dispositifs CMOS, l'empilement HfO2/Métal a remplacé l'empilement SiO2/polySi. Cependant, la diffusion incontrôlée des espèces chimiques dans ces nouveaux empilements fabriqués avec un fort budget thermique compromet l'obtention des travaux de sortie (EWF) et des épaisseurs d'oxyde équivalent (EOT) définis par l'ITRS. Une solution consiste à utiliser une intégration à plus bas budget thermique. Avec cette nouvelle approche, l'objectif de ce travail de thèse était de comprendre les paramètres physiques permettant d'obtenir une EOT<1nm et des EWF permettant une co-intégration nMOS et pMOS pour des nœuds futurs CMOS 20-14 nm. En nous appuyant sur différents méthodes d'analyse physico-chimique (STEM EDX, TOF-SIMS et XPS), la distribution spatiale des éléments et leurs liaisons chimiques au sein d'empilements de taille nanométrique ont été discutées et, sur la base de considérations thermodynamiques, corrélées aux valeurs mesurées de l'EOT et EWF. Nous avons démontré pour la première fois un écart de ~0. 8eV entre une électrode TiAlNx déficitaire et riche en azote, déposée sur HfO2. Ces résultats ont été obtenus après avoir identifié les mécanismes qui contrôlent l'EWF et l'EOT dans des empilements plus simples TiN/Ti, Al et TiAl. Les grilles HfO2/TiAlNx ne sont cependant pas stables thermiquement. Nous avons alors proposé deux systèmes métalliques plus simples et plus stables utilisant des alliages TaNix et NiTix obtenus par interdiffusion dans les empilements HfO2/Ta/Ni et de HfO2/Ni/Ti. Ces structures de grilles à base de Ni apparaissent prometteuses pour une co-intégration CMOS à bas budget thermique
To continue CMOS scaling, the HfO2/metal gate stack replaced the historical SiO2/PolySi gate stack. But the uncontrolled interdiffusion and reactivities of the new gate materials integrated with the classical high thermal budget approach appear to be a roadblock to reach the effective work function (EWF) and equivalent oxide thickness (EOT) ITRS targets. One solution consisted in implementing an approach with a lower thermal budget. Using this new approach, the aim of this thesis work was to understand the physical mechanisms, which enable to reach an EOT<1nm and an EWF relevant for nMOS and pMOS co-integration as required for the next 20-14nm CMOS nodes. Using spatially resolved TEM/EDX analyses and macroscopic TOF-SIMS and XPS techniques, elemental distributions and chemical bonds across nanometric-sized stacks were discussed and, based on thermodynamic considerations, correlated with the measured EWF and EOT. We showed for the first time that the modulation of nitrogen during TiAlN deposition on HfO2 results in a ~0. 8eV EWF shift between the N-poor and N-rich HfO2/TiAlNx electrodes. The TiAlN complex system was understood after the identification of the EWF and EOT modulation mechanisms in the simple gate stacks TiN/Ti, Al or TiAl. Although TiAlNx electrodes define the best compromise for a variable EWF with a sub-nm EOT, it exhibits a low thermal stability. Therefore, we investigated two simpler metallic and stable systems using TaNix and NiTix alloys resulting from thermally assisted Ni-Ta and Ni-Ti interdiffusion in HfO2/Ta/Ni and HfO2/Ni/Ti stacks, respectively. These Ni-based electrodes are shown to be promising for a low thermal budget CMOS co-integration

Les systèmes barrière thermique actuels connaissent une importante dispersion de durées de vie liée principalement aux ondulations de surface du revêtement métallique β-(Ni,Pt)Al provoquant l’écaillage du dépôt céramique. Les revêtements γ-γ’ riches en platine sont étudiés en tant qu’alternative au système actuel. Ce travail de thèse s’est intéressé à l’élaboration des revêtements γ-γ’ riches en platine sur un superalliage à base de nickel, l’AM1 à partir de procédés conventionnels : dépôt électrolytique de platine et aluminisation courte. Les mécanismes de dégradation par oxydation cyclique à 1100°C ont été étudiés sur des systèmes revêtement/AM1 et sur des systèmes barrière thermique. Pour comparaison, trois types de revêtement ont été élaborés : γ-γ’ Pt seul, γ-γ’ Pt+Al et β-(Ni,Pt)Al. Ces essais ont mis en évidence une meilleure tenue à l’oxydation cyclique des systèmes revêtus γ-γ’ Pt+Al comparée aux deux autres systèmes revêtus. L’importance de l’ajout d’aluminium dès l’élaboration sur la tenue à l’oxydation cyclique a été soulignée. La modélisation p-kp a mis en avant une augmentation de la proportion d’écaillage au cours du temps du fait de la dégradation de l’interface métal/oxyde et une augmentation du kp du fait de la formation d’un oxyde à croissance plus rapide. Outre l’oxydation, les phénomènes d’interdiffusion lors des tous premiers instants à haute température ont été étudiés à partir de matériaux modèles (Ni13Al et Ni11Al10Cr) et de revêtements de Pt et/ou de Pt-Ir. Ces essais ont mis en avant la rapide formation de la phase α-NiPtAl, les transformations de phases et les chemins de diffusion à 1100°C dans les systèmes Ni-Al-Pt et Ni-Al-Cr-Pt. L’effet du chrome et de l’iridium sur les cinétiques de diffusion a été évalué. La modélisation de l’interdiffusion a mis en évidence les interactions chimiques entre les espèces et une sursaturation en lacunes dans la zone d’interdiffusion prouvant que l’effet Kirkendall est responsable de la formation des pores.

An in depth study of the performance of thermally interdiffused concentration gradient polymer photovoltaic devices is carried out with particular attention to the effect of the thickness and the thermal treatments on the power conversion efficiency, short circuit current, open circuit voltage and other key electrical properties. Bilayer films of sequentially spin-cast donor and acceptor materials are exposed to various heat treatments in order to induce the interdiffusion. The depth profiles show concentration gradients in the donor and acceptor as a result of interdiffusion and these devices show an order of magnitude increase in the device performance compared to the bilayer devices. Dual spin-cast poly (3-octylthiophene-2,5-diyl) (P3OT)- [6,6] phenyl C61 butyric acid methyl ester (PCBM) and poly (3-hexylthiophene-2,5-diyl) (P3HT)-PCBM interdiffused devices are studied in detail by varying the thickness of the donor and acceptor layers as well as the annealing conditions for initial polymer layer and the time and temperature of the interdiffusion process. Auger spectroscopy and X-ray photoelectron spectroscopy along with ion beam milling are used to investigate the concentration gradient formed as a result of the interdiffusion. The sulfur signal present in the P3OT and P3HT backbone is detected to identify the concentration profiles in the P3OT-PCBM and P3HT-PCBM devices. The interdiffusion conditions and thickness of the active layers have been optimized to obtain the highest power conversion efficiency. The best device performance of the P3OT-PCBM interdiffused devices is achieved when the interdiffusion is carried out at 150°C for 20 minutes and the P3OT thickness is maintained at 70 nm and the PCBM thickness at 40-50 nm. The highest efficiency achieved for P3OT-PCBM interdiffused devices is 1.0% under AM1.5G solar simulated spectrum. In order to further increase the efficiency, P3OT is replaced by (P3HT) which has higher hole mobility. P3HT- PCBM based concentration gradient devices show improved device performance over P3OT-PCBM devices. Power conversion efficiency of the order of ~3.0% is obtained for P3HT-PCBM interdiffused devices when the interdiffusion is carried out at 150°C for 20 minutes. For both P3OT:PCBM and P3HT:PCBM devices, the optimum performance occurs when the concentration gradient extends across the entire film and is correlated with an increase in the short circuit current density and fill factor as well as a decrease in the series resistance. The results demonstrate that an interdiffused bilayer fabrication approach is a novel and efficient approach for fabrication of polymer solar cell devices. In addition, porphyrin derivative 5, 10, 15, 20-Tetraphenyl-21H, 23H-porphine zinc (ZnTPP) is studied as a new donor material for organic solar cells. ZnTPP: PCBM blend devices are investigated in detail by varying the weight ratio of the donor and acceptor materials in blend devices. The devices with ZnTPP: PCBM in 1:9 ratios showed the best device performance and the efficiency of the order of 0.2% is achieved under AM1.5G solar simulated conditions. Trimetallic Nitride Tempelated (TNT) endohedral fullerenes are also examined in this thesis as the novel acceptor materials. Bulk heterojunction or blend devices are fabricated with P3HT as the donor material and several TNT endohedral fullerenes as the acceptor material. Y3N@C₈₀PCBH based devices which are annealed both before and after the electrode deposition show improvement in the device performance compared to devices that are only annealed before the electrode deposition. The highest power conversion efficiency achieved for TNT endohedral fullerene devices is only 0.06%, suggesting that substantial additional work must be done to optimize the compatibility of the donor and acceptor as well as the device fabrication parameters.
Ph. D.

In this thesis, the interface between the electron donor polymer and the electron acceptor fullerene in organic photovoltaic devices is studied. Starting from a bilayer system of donor and acceptor materials, the proximity of polymer and fullerene throughout the bulk of the devices is improved by inducing an interdiffusion of the two materials by heating the devices in the vicinity of the glass transition temperature of the polymer. In this manner, a concentration gradient of polymer and fullerene throughout the bulk is created. The proximity of a fullerene within 10 nm of any photoexcitation in the polymer ensures that the efficient charge separation occurs. Measurements of the absorption, photoluminescence, and photocurrent spectra as well as I-V characteristics are used to study the interdiffusion and its influence on the efficiency of the photovoltaic devices. In addition, the film morphology is studied on a microscopic level with transmission electron microscopy and with Auger spectroscopy combined with ion beam milling to create a depth profile of the polymer concentration in the film. Initial studies to induce an interdiffusion were done on poly(2-methoxy-5-(2'-ethylhexyloxy)-1,4-phenylenevinylene) (MEH-PPV) as the electron donor polymer and the buckminsterfullerene C60 as the electron acceptor. Interdiffused devices show an order of magnitude photoluminescence quenching with concomitant increase in the photocurrents by an order of magnitude. Variation of the polymer layer thickness shows that the photocurrents increase with decreasing thickness down to 70 nm due to charge transport limitation. The choice of layer thickness in organic photovoltaic devices is critical for optimization of the efficiency. The interdiffusion process is also monitored in situ and a permanent increase in photocurrents is observed during the heat treatment. Transmission electron microscopy (TEM) studies on cross sections of the film reveal that C60 interdiffuses into the MEH-PPV bulk in the form of >10 nm clusters. This clustering of C60 is a result of its tendency to crystallize and the low miscibility of C60 in MEH-PPV, leading to strong phase separation. To improve the interdiffusion process, the donor polymer is replaced by poly(3-octylthiophene-2,5-diyl) (P3OT), which has a better miscibility with C60. Again, the photocurrents of the interdiffused devices are improved significantly. A monochromatic power conversion efficiency of 1.5 % is obtained for illumination of 3.8 mW/cm2 at 470 nm. The polymer concentration in unheated and interdiffused films is studied with Auger spectroscopy in combination with ion beam milling. The concentration profile shows a distinct interface between P3OT and C60 in unheated films and a slow rise of the P3OT concentration throughout a large cross-section of the interdiffused film. TEM studies on P3OT/C60 films show that C60 still has some tendency to form clusters. The results of this thesis demonstrate that thermally-controlled interdiffusion is a viable approach for fabrication of efficient photovoltaic devices through nanoscale control of composition and morphology. These results are also used to draw conclusions about the influence of film morphology on the photovoltaic device efficiency and to identify important issues related to materials choice for the interdiffusion process. Prospective variations in materials choice are suggested to achieve better film morphologies.
Ph. D.

An in-depth study of concentration gradients in thermally-interdiffused polymer – fullerene photovoltaic devices, with a focus on thickness and heat treatments, is presented in this thesis. Device performance is improved from the bilayer by the creation of a concentration gradient of the donor and acceptor materials throughout the active layer of the device. Concentration gradients are expected to improve device performance by optimizing the charge transfer, transport and collection processes. This is achieved through heat-induced interdiffusion of the two materials at temperatures above the glass transition temperature of the polymer. Investigation of the poly(3-octylthiophene) (P3OT) – C₆₀ system show a three-fold improvement in the external quantum efficiencies (EQE) as compared with bilayer devices. Auger spectroscopy, combined with argon-ion beam milling, serves to record the concentration depth profile and identify concentration gradients in the device through detection of the sulfur in the P3OT backbone. Concentration gradients are optimized to yield the best devices through a thickness variation study conducted on the P3OT – C₆₀ system for fixed thermal interdiffusion conditions at 118 °C for 5 minutes. An optimum thickness of 40 to 60 nm is obtained for the two materials that yields the ideal morphology of a concentration gradient as recorded by Auger spectroscopy. For such devices, the concentration gradient is seen to extend through the device, ending in a thin layer of pure material at each electrode. A monochromatic power conversion efficiency of 2.05% is obtained for 5.3 mW/cm²⁺ illumination at 470 nm. A brief study is also presented to optimize the concentration gradient profile through variations of the thermal parameters. The dependence of the concentration gradient on the interdiffusion time and temperature is investigated. The merits of heat treatment on the crystallinity of P3OT and the overall device performance are also discussed. It is shown in some case that devices with annealed P3OT layers show almost twice the EQE as non-annealed P3OT layer devices. Potential alternatives for C₆₀ in interdiffused devices with P3OT have been presented. [6,6]-phenyl C₆₁-butyric acid methyl ester (PCBM), a well-investigated acceptor for blend devices, is studied as an acceptor in concentration gradient devices. A method for spin-coating uniform bilayers of P3OT and PCBM, without solution damage to either layer, is presented. A thermal variation study of the interdiffusion conditions on this system indicated higher interdiffusion temperatures and times are preferred for P3OT – PCBM systems. For interdiffusion at 150 °C for ten minutes, EQE values approaching 35 % at 500 nm are obtained. Auger spectroscopy studies on this system yielded the same conclusions about the concentration gradient device morphology that gives optimum device output. 1:1 and 1:2 blends of P3OT – PCBM are also studied. The influence of various thermal treatments on these devices is described. The endohedral fullerene Sc₃N@C₈₀ is introduced as a new acceptor material. The endohedral fullerene consists of Sc₃N cluster enclosed in a C₈₀ cage. An order of magnitude increase is seen in device performance upon sublimation of these molecules on a P3OT layer confirming its effectiveness as an acceptor. Preliminary studies done on this system indicated the need for greater thermal treatment to produce optimum concentration gradients. An in depth study for varying temperatures and times is presented. The best device performance was seen for interdiffusion at 160 °C for 25 minutes. The endohedral fullerene devices also show a long-term deterioration and so best result are presented from a set of devices fabricated within the same time period. The study of these three donor-acceptor systems confirms that the conclusions on the thickness dependence and device performance study conducted for the P3OT – C₆₀ system extend to other acceptors. A model of EQE for varying thicknesses based on absorption in the interdiffused concentration gradient regions is also presented. This model effectively highlights the influence of P3OT layer thickness on the trends observed in the EQE. It did not, however, reproduce the experimental thickness variation results for varying C₆₀ thicknesses. Incorporation of the effects of the electric field intensity distribution is expected to correct for this. Suggestions have been given on how this might be achieved.
Ph. D.

Advances in wide-bandgap semiconductor devices have increased the allowable operating temperature of power electronic systems. High-temperature devices can benefit applications such as renewable energy, electric vehicles, and space-based power electronics that currently require bulky cooling systems for silicon power devices. Cooling systems can typically be reduced in size or removed by adopting wide-bandgap semiconductor devices, such as silicon carbide. However, to do this, semiconductor device packaging with high reliability at high temperatures is necessary. Transient liquid phase (TLP) die-attach has shown in literature to be a promising bonding technique for this packaging need. In this work TLP has been comprehensively investigated and characterized to assess its viability for high-temperature power electronics applications. The reliability and durability of TLP die-attach was extensively investigated utilizing electrical resistivity measurement as an indicator of material diffusion in gold-indium TLP samples. Criteria of ensuring diffusive stability were also developed. Samples were fabricated by material deposition on glass substrates with variant Au–In compositions but identical barrier layers. They were stressed with thermal cycling to simulate their operating conditions then characterized and compared. Excess indium content in the die-attach was shown to have poor reliability due to material diffusion through barrier layers while samples containing suitable indium content proved reliable throughout the thermal cycling process. This was confirmed by electrical resistivity measurement, EDS, FIB, and SEM characterization. Thermal and mechanical characterization of TLP die-attached samples was also performed to gain a newfound understanding of the relationship between TLP design parameters and die-attach properties. Samples with a SiC diode chip TLP bonded to a copper metalized silicon nitride substrate were made using several different values of fabrication parameters such as gold and indium thickness, Au–In ratio, and bonding pressure. The TLP bonds were then characterized for die-attach voiding, shear strength, and thermal impedance. It was found that TLP die-attach offers high average shear force strength of 22.0 kgf and a low average thermal impedance of 0.35 K/W from the device junction to the substrate. The influence of various fabrication parameters on the bond characteristics were also compared, providing information necessary for implementing TLP die-attach into power electronic modules for high-temperature applications. The outcome of the investigation on TLP bonding techniques was incorporated into a new power module design utilizing TLP bonding. A full half-bridge inverter power module for low-power space applications has been designed and analyzed with extensive finite element thermo-mechanical modeling. In summary, TLP die-attach has investigated to confirm its reliability and to understand how to design effective TLP bonds, this information has been used to design a new high-temperature power electronic module.
Ph.D.
Doctorate
Electrical Engineering and Computer Science
Engineering and Computer Science
Electrical Engineering

Identification des types de recombinaison entre 2 et 300k dans les couches epaisses de gainas et gainp epitaxiees sur leur support respectif inp et gaas. Etude de l'origine de la luminescence et variation en fonction de l'epaisseur du taux de capture des porteurs de la barriere dans les puits quantiques ingaas/inp. Determination du coefficient d'interdiffusion de al et ga aux interfaces dans les puits quantiques gaas/gaalas

An investigation into the effects of thermal interdiffusion on the characteristics of quantum dot infrared photodetectors (QDIPs) yields results useful for the creation of a two-colour QDIP. For high temperature rapid thermal annealing, quantum dot interdiffusion is induced, resulting in a large wavelength redshift in the photodetector spectral response, at the cost of a small degradation in device performance. The evaluation of a two-colour QDIP fabricated using selective suppression of interdiffusion during thermal annealing shows uniform performance in the two different detector pixels. This has implications as a useful process for the future fabrication of multi-colour QDIPs.

碩士
國立臺灣科技大學
化學工程研究所
84
At present the primary material for the metalization of the ULSI(ultra-large scale integration) is aluminum. The goal of achieving high reliability and yield for ULSI fabrication can only be realized with the advanced annealing technology. We applied two different annealing processes on our multi-layer system in which Al films were sputtering deposited and TiN films were deposited with the reactive sputtering. The effects of the annealing process on the TiN diffusion barrier, surface morphology, and interdiffusion and reactions between interfaces, was executed by x-ray diffraction (XRD), scanning electron microscopy（SEM）, optical microscopy, Auger electron spectroscopy and Rurtherford backscattering spectroscopy. Experimental results revealed that aluminum melted locally at the unstable area and forming holes on the surface of samples. AlN and Al3Ti formed at the interface between Al and TiN. The formation of titaniumsilicide between Ti and Si was also obsvered. It is believed that atoms will diffuse through the barrier film via defects in the barrier films, such as grain boundaries or voids. Furnace annealing makes those defects combine into voids and the Al atoms can be more easily to diffuse through the barrier during annealing at the temperature of 550℃. On the other hand, in-situ annealing produced fewer defects associated with the TiN barrier film and therefore could avoid the surface reactions and interdiffusion with Al and enhanced the barrier quality.

碩士
國立交通大學
電子物理系所
105
ZnTe quantum dots were grown by Stranski-Krastnov mode on ZnMgSe of different Mg composition by Molecular Beam Epitaxy. The coverage thickness of quantum dots were 4.0, 5.0, 6.0, and 7.0 MLs. ZnTe quantum dots grown on ZnSe was investigated for reference. The physical characteristics were studied by rapid thermal annealing (RTA), photoluminescence (PL) spectroscopy, temperature dependent PL, and time resolved PL (TRPL). 　　Different coverage of ZnTe quantum dots were annealed at temperature from 350°C to 550°C for 30 seconds by RTA. Two mechanisms, ripening and inter-diffusion, were observed in PL spectra of annealed samples. For the samples of annealing temperature larger than 470°C, a red shift of PL peak position was observed in quantum dots of smaller coverage. It is the signature of ripening of quantum dots. However, for the quantum dots with larger coverage, there was a blue shift of PL peak position with increasing annealing temperature. It results from the inter-diffusion of Mg from the barrier layers to the quantum dots. For the quantum dots with moderate coverages, two mechanisms balance and it results in no shift of PL peak position for different annealing temperatures. 　　Three different activation energies were found from the Arrhenius fitting of the temperature-dependent PL. The activation energies also vary with ripening and inter-diffusion mechanisms, and the effect of ripening and inter-diffusion on the band structure could be understood. By using two recombination lifetimes to fit the TRPL spectra, the effect of annealing mechanisms on the carrier lifetime was studied. Current study showed that the dot density and size can be controlled by different coverage thickness of quantum dots and the ripening and inter-diffusion mechanisms of rapid thermal annealing.

Superalloys are widely applied as materials for components in the hot section of gas turbines. As superalloys have a limited oxidation life, the application of a coating is vital. The most commonly applied coatings in stationary gas turbines are MCrAlY coatings. Since the turbine components are exposed to high cyclic thermal stresses, MCrAlY coatings must also show a high thermal fatigue resistance. In this thesis, the effect of Cobalt and Rhenium on microstructure, oxidation and thermal fatigue of NiCoCrAlY coatings is presented. Additionally the condition of the coatings after testing in an industrial gas turbine is shown. The influence of Cobalt and Rhenium on coating microstructure was investigated by thermodynamic modelling and by metallography. It could be shown that both elements reduce the γ`-phase fraction and increase the β-phase fraction owing to an expansion of the γ+β field in the phase diagram. Modelling showed that Rhenium promotes the formation of α-Cr, which could be explained by a shift of the α-Cr solvus to higher temperatures and lower Cr concentrations. In the real coatings Re causes the precipitation of TCP-phase. The oxide scale growth rate is increased by Cobalt and Rhenium and it appears that Yttrium plays a significant role for that effect. Coating consumption due to simultaneous oxidation and interdiffusion could be decreased by the application of Cobalt and Rhenium. In thermal fatigue testing Rhenium reduces the time to crack initiation and increases crack propagation rate, although it could be shown that Rhenium increases the creep resistance of the coating. The effect could be explained by the influence of Rhenium on the microstructure, which increases creep resistance, but also reduces the ductility of the coating.

碩士
國立中興大學
材料工程學研究所
86
Indium and its alloys exhibit good service life and reliability, and are well suited for applications such as solder interconnections in electronic packaging. In this study, two types of Au-In microjoints, i.e. Au/In/Au in which In foil was used and Au/In, were prepared by either solid state interdiffusion (SSID) or solid-liquid interdiffusion (SLID) bondings for a single lap tensile test. Deposition of the Au and In thin films was carried out by thermal evaporation on a polyethylene terephthalate (PET) sstrate. It was found that the shear strength of the Au/In microjoints is higher than that of which using In foil, i.e. Au/In/Au. In addition, it is observed that the fracture mode of the Au-In microjoints depends on the types of In used. Failure of the Au/In microjoints appeared to be along the joint-substrate interfaces, whereas it occurred within the In foil for the other type of specimens. Examination of the Au/In microjoints by glancing angle X-ray diffraction reveals the presence of intermetallics In2, Au10In3, and Au9In4, in small amount, in addition to the two major constitutent phases, Au7In3 and Au. On the other hand, only intermetallic AuIn2 and pure In were obtained in the Au/In/Au microjoints, where the thickness of In is much higher than that of Au.