Academic literature on the topic 'Cu-Si systems'

The processing window is important for the semisolid processability of alloys. Applications of semi-solid metal (SSM) processing, especially aluminium alloys have been expanding for their excellent mechanical properties. However, the alloys well suited and commercially used for SSM processing today are limited in types. The main purpose of this Ph.D. project is to understand what makes an alloy suitable for SSM processing on both aspects of thermodynamics and kinetics. This research started with a fundamental study of binary alloys based on Al-Si, Al-Cu and Al-Mg systems (wt%): Al-1Si, Al-5Si, Al-12Si and Al-17Si; Al-1Cu, Al-2Cu and Al-5Cu; Al-0.5Mg, Al-3Mg and Al-5.5Mg. These are representative of Si, Cu and Mg contents in commercial alloys used for SSM processing. The Single-Pan Scanning Calorimeter (SPSC) and Differential Scanning Calorimeter (DSC) were used to investigate the liquid fraction changes during heating and cooling of these binary alloys. Thermo-Calc and DICTRA (DIffusion-Controlled TRAnsformations) software have been used to predict the fraction liquid versus temperature taking into account both thermodynamics and kinetics. Comparison of the predictions with experimental data revealed that the simulation results show the same pattern with experimental results in the fraction liquid-temperature relationship. However, the SPSC results are closer to the prediction than DSC curves are, even with the relatively large sample size associated with SPSC. This is potentially a significant result as predicting the liquid fraction versus temperature for the heating of a billet for semi-solid processing remains one of the challenges. The results also suggest that the fraction liquid sensitivity to time should be identified as a critical parameter of the process window for semi-solid processing in addition to the fraction liquid sensitivity to temperature. For microstructure investigation, microanalysis techniques, including Scanning Electron Microscopy (SEM) and micro-indentation testing, have been used on polished sections, and compared to theoretical predictions. In addition, some parts of this project are in cooperation with General Research Institute for Nonferrous Metals (GRINM), which aims to design and develop high performance semi-solid alloys. Thermodynamic analysis (both predictions and experiments) were carried out on thixoformed 319s (2.95Cu, 6.10Si, 0.37Mg, wt%) and 201 (4.80Cu, 0.7Ag, wt%) aluminium alloys. SEM techniques and Transmission Electron Microscopy (TEM) were used for the microstructural characterisation. The results showed that the DSC curves were sensitive to microsegregation in SSM alloys and resulted in a lower liquid fraction than the cast alloys calculated through the integration method from the DSC results. Al2Cu phase in SSM alloys 319s and 201 can be dissolved into matrix up to 0.4 % before melting temperature under 3K/min heating rate when compared with 10K/min heating rate. The DSC scan rate should be carefully selected as higher heating rate can inhibit dissolution of the intermetallic phases during heating leading to less accurate liquid fractions predictions.

A skyrmion is a vortex-like spin pattern which has been observed in so-called B20 magnetic systems such as MnSi, (Mn,Fe)Si and a few other metallic magnets as well as in insulating Cu₂OSeO₃. We conduct a comprehensive study of muon spin relaxation (μSR) on bulk single crystals of MnSi and (Mn,Fe)Si, a MBE thin film of MnSi, and a ceramic specimen of Cu₂OSeO₃ in this work. The generic second-order like phase transition indicated by 1/T₁ peaks at T_c in bulk systems is discussed in light of the Brazovskii-type first-order phase transition due to the presence of the DM interaction. We also discuss the different temperature dependences of μ⁺ spin-lattice relaxation rate 1/T₁ in bulk pure systems MnSi and Cu₂OSeO₃ and their commonalities in the paramagnetic state and the ordered state due to the DM interaction. Furthermore, we highlight the enhanced 1/T₁ in the skyrmion crystal (SkX) phase compared to neighboring conical phases due to an abundance of low-energy magnetic fluctuations/excitations. This abundance is corroborated by the reduced static order parameter in the SkX phase of MnSi compared to neighboring conical phases, deduced by combining μSR experiments and magnetic field simulations. The intermediate (IM) region above T_c, where the modification of magnetic transition by the DM interaction starts to appear in MnSi, exhibit multi-time scale spin fluctuations, topologically non-trivial Hall resistivity and non-Fermi-liquid exponent of longitudinal resistivity in single-crystal Mn₀.₉Fe₀.₁Si and the MnSi MBE thin film, similar to the magnetically disordered phase of pure MnSi under hydrostatic pressure. These three defining features indicate a fluctuating skyrmion liquid in this magnetically ordered state, stabilized by pressure, disorder or reduced dimensionality. Moreover, the magnetic transition is strongly first order in the MnSi MBE thin film sample, different from the Brazovskii-type weakly first order transition in bulk samples, suggesting the importance of reduced dimensionality in modifying the nature of magnetic phase transitions in B20 systems.

This work falls in general area of the electro-thermo-mechanical driven mass transport in Cu-Si systems, which often finds relevance while accessing reliability issues pertaining to thin film interconnects in microelectronic devices. In such a system, the major driving forces are electric potential gradient (and hence electromigration), current crowding induced temperature gradient (and hence thermomigration) and coefficient of thermal expansion (CTE) mismatch induced stresses. Herein, a coupling between different driving forces, such as electromigration and thermomigration, may also occur, which can subdue or accelerate the mass transport in Cu. In addition, due to decrease in the thickness of interconnects to a few nanometers, the contribution of diffusion through the Cu-Si interface in overall mass transport cannot be neglected due to an increase in the interface area to volume ratio. Therefore, it can be inferred that electromigration, thermomigration and thermal stress induced failures of Cu-Si systems should be sensitive to the property of the interface, making it imperative to investigate the role of the interlayer placed in between Cu and Si on mass transport. Accordingly, this work focuses on studying the role of the coupling between the aforementioned major driving forces, especially electric potential gradient and temperature gradient, and the interlayer on the mass transport behavior in Cu-Si system. Firstly, the effect of current crowding induced temperature gradient on the electric current induced mass transport in Cu films was studied. This effect was studied using samples fabricated according to the standard Blech configuration, wherein long Cu thin film was deposited on Si substrate with a very thin W interlayer. In these tests, regular mass transport at the cathode, termed as forward mass transport, was observed along with an anomalous mass depletion at the anode, termed as backward mass transport, especially when currents of very high current density (>106 A/m2) was passed. The anomalous backward mass transport behavior is explained by illuminating the coupling between the temperature gradient induced mass transport (i.e., thermomigration) and the electric current induced mass transport (i.e., electromigration) at the anode. Herein, temperature gradient was estimated using finite element analysis, performed using COMSOL Multiphysics, using the full-length scale model. The kinetics of the anomalous backward mass transport at the anode was also studied by varying current density. The anomalous mass transport, which has origins in the establishment of very high temperature gradients at the anode, became more pronounced with increase in the higher current density. In addition to the temperature gradient, the temperature of the sample also increased with an increase in the current density, and since the kinetics of electromigration as well as thermomigration induced mass transport are diffusion controlled, an increase in the current density further exacerbates the net mass transport, irrespective of whether it is regular forward or anomalous backward mass transport. Subsequent to establishment of the existence of significant thermomigration-electromigration coupling in samples fabricated using Blech configuration, systemic experiments were performed to understand the role of the thermomigration-electromigration coupling induced mass transport on the so-called Blech length effect1. Herein, experiments were performed by passing current through a sample wherein a long Cu film on Si substrate with W interlayer was segmented into multiple stripes with length varying from 10 m to 200 m. Contrary to the Blech length effect, these samples showed enhancement in the regular mass transport at the 1 Blech length effect is understood as elimination of electromigration (i.e., material depletion at the cathode and material accumulation in form of whiskers or hillocks at anode) when the product of the current density and the sample length is smaller than a critical value. cathode with a decrease in the stripe length. We term this behavior as inverse Blech length effect. These results imply that thermomigration, besides electromigration, should also be considered while understanding the role of electric current on reliability of Cu-Si systems having bends, e.g., modern 3-D Cu interconnects fabricated using dual damascene process, etc., as the thermomigration-electromigration coupling violates the conventional wisdoms of mass depletion only at the cathode, existence of Blech length effect, etc. Finally, the role of interlayers, such as W, Ta, and Ti, in the mass transport in Cu in Cu-Si system due to the electromigration-thermomigration coupling and CTE mismatch induced thermal stresses was studied. A significant role of the interlayer was observed in the electromigration-thermomigration coupling induced mass transport, wherein a strongly bonded interface, such as Cu-Ti, did not show an inverse Blech length effect, whereas a weakly bonded interface, such as Cu-Ta, Cu-W and Cu-TiO2, showed the aforementioned inverse Blech length effect. The interface structure was characterized using transmission electron microscope, and the obtained information, along with the finite element analysis, was used to explain the observed results. Similarly, the experiments performed by cycling the temperature of the Cu-Si samples between -50 to 150 oC revealed a significant role of the interlayer on the extent as well as the nature of plastic deformation in Cu. These experiments were performed by depositing Cu islands on Si substrate with Ni, W or no interlayer and by measuring the extent of sliding of Cu film. In addition to sliding, a few Cu grains also protruded to accommodate the CTE mismatch induced stresses. In summary, the mass transport in Cu-Si system can be tuned by understanding the role of the sample geometry, coupling between driving forces and the interlayer.

碩士<br>國立交通大學<br>電子研究所<br>83<br>Thermal stability of the Cu/SiO2/Si system was investigated with respect to the dielectric degradation and Cu ion migration in the Cu-gate MOS capacitor. We used the rapid thermal annealing (RTA) and the technique of bias-temperature stress in conjunction with the dielectric breakdown field (Ebd) and SiO2/ Si interface state density (Dit) measurements to characterize the thermal stability of the Cu/SiO2/Si system. We found that the Ebd degradation started to occur after Cu/SiO2/Si structure was annealed with 60 sec RTA at a temperature as low as 300℃; and the dielectric strength deteriorated progressively with the increase of annealing temperature. The dielectric degradation is presumably due to Cu dissolution in SiO2 layer in the form of positive ion. The mobility of Cu ion in the SiO2 layer was evaluated using the data obtained from the bias- temperature stress. The Cu ion concentration in the SiO2 layer of Cu-gate MOS capacitor resulting from RTA anneal was also evaluated using a simple extractation scheme. It is also concluded that Cu is a fast diffusion specises in SiO2 and may diffuse into Si substrate once it arrives at the SiO2/Si interface.

Two methods for refining metallurgical grade silicon to solar grade silicon have been investigated. The first method involved the reduction of impurities from metallurgical grade silicon by high temperature vacuum refining. The concentrations of analyzed elements were reduced several times. The main steps in the second refining method include alloying with copper, solidification, grinding and heavy media separation. A metallographic study of the Si-Cu alloy showed the presence of only two microconstituents, mainly pure silicon dendrites and the Cu3Si intermetallic. SEM analysis showed a distinct boundary between the silicon and the Cu3Si phases, with a large concentration of microcracks along the boundary, which allowed for efficient separation. After alloying and grinding, a heavy media liquid was used to separate the light silicon phase from the heavier Cu3Si phase. Cu3Si residues together with the remaining impurities were found to be located at the surface of the pure silicon particles, and should be efficiently removed by acid leaching. Thirty elements were analyzed by the Inductively Coupled Plasma Mass Spectrometry (ICP) chemical analysis technique. ICP revealed a several times higher impurity level in the Cu3Si intermetallic than in the pure silicon; furthermore, the amounts of 22 elements in the refined silicon were reduced below the detection limit where the concentrations of 7 elements were below 1ppmw and 6 elements were below 2ppmw. The results showed that the suggested method is efficient in removing impurities from metallurgical grade silicon with great potential for further development.