Recently, refractory materials have been proposed as a strong alternative to poly-silicon and aluminum alloys as metallization systems for Very Large Scale Integrated (VLSI) circuits because of their improved performance at smaller Integrated Circuit (IC) feature size and higher interconnect current densities. However, processing and reliability problems associated with the use of refractory materials have limited their widespread acceptance. The hot-wall low pressure chemical vapor deposition (LPCVD) of Molybdenum and Tungsten from their respective hexacarbonyl sources has been studied as a potential remedy to such problems, in addition to providing the potential for higher throughput and better step coverage. Using deposition chemistries based on carbonyl sources, Mo and W deposits have been characterized with respect to their electrical, mechanical, structural, and chemical properties as well as their compatibility with conventional IC processing. Excellent film step coverage and uniformity were obtained by low temperature (300-350 C) deposition at pressures of 400-600 mTorr. As-deposited films were observed to be amorphous, with a resistivity of 250 and 350 microohm-cm for Mo and W respectively. On annealing at high temperatures in a reducing or inert atmosphere, the films crystallize with attendant reduction in resistivity to 9.3 and 12 microohm-cm for Mo and W, respectively. The average grain size also increases as a function of time and temperature to a maximum of 2500-3000 A. The metals and their silicides that are deposited, using silane as silicon source, are integratable to form desired metal-silicide gate contact structures. Thus, use of the low resistivity of the elemental metal coupled with the oxidation resistance of its silicide manifests the quality and economy of the process. MOS capacitors with Mo and W as the gate material have been fabricated on n-type (100) silicon. A work function of 4.7 +/- 0.1 eV was measured by means of MOS capacitance-voltage techniques. The experimental results further indicate that the characteristics of W-gate MOS devices related to the charges in SiO₂ are comparable to those of poly-silicon; while, the resistivity is about two orders of magnitude lower than poly-silicon. It is therefore concluded that hot-wall low pressure chemical vapor deposition of Mo and W from their respective carbonyl sources is a viable technique for the deposition of reliable, high performance refractory metal/silicide contact and interconnect structures on very large scale integrated circuits.

The displacement damage induced by primary recoils of fusion neutrons in tungsten and alloys has been studied with self-ion irradiations, followed by damage characterization with electron microscopy. Tungsten and alloys (≤ 5 wt.% Re, Ta, V) were implanted with 2 MeV W+ ions over a dose range of 3.3×1017 - 2.5×1019 W+m-2 at temperatures ranging from 300 to 750°C. Dislocation loops of b = ½<111> (> 60%) and b = <100> were identified, and that ½<111> loops were found more thermally stable. Among loops that were large enough for nature determination, at least 50% were found to be of interstitial type, with larger fractions in high-temperature and high-dose conditions. The diameter of loops did not exceed 20 nm, with the majority being ≤ 5 nm. The loop number density varied between 1022 and 1023 m-3. The effects of ion dose, irradiation temperature, composition and grain orientation on damage microstructure were investigated. In-situ irradiations (150 keV W+ ions) were carried out as a complement to the bulk implantations. Qualitative trends in loop size, geometry and nature with irradiation dose and temperature were similar to bulk irradiated specimens. Also, the dynamics of defects and their effects on the damage evolution were explored. In-situ annealing of irradiated thin foils was performed to investigate the thermal stability of radiation damage in tungsten. The majority of microstructure transformations were completed within 15 min of annealing. However, extended durations did favour the increase of loop size and the fraction of ½<111> loops.

Many technological hurdles remain before a viable commercial magnetic fusion energy reactor can be constructed, including the development of plasma-facing components with long lifetimes that can survive the harsh environment inside a reactor. One such component, the divertor, which maintains the purity of the plasma by removing fusion byproducts from the reactor, must be able to accommodate very large incident heat fluxes of at least 10 MW/m^2 during normal operation. Modular helium-cooled tungsten divertors are one of the leading divertor designs for future commercial fusion reactors, and a number of different candidates have been proposed including the modular He-cooled divertor concept with pin array (HEMP), the modular He-cooled divertor concept with multiple-jet-cooling (HEMJ), and the helium-cooled flat plate (HCFP). These three designs typically operate with helium coolant inlet temperatures of 600 °C and inlet pressures of 10 MPa. Performing experiments at these conditions to evaluate the thermal performance of each design is both challenging and expensive. An alternative, more economical approach for evaluating different designs exploits dynamic similarity. Here, geometrically similar mockups of a single divertor module are tested using coolants at lower temperatures and pressures. Dynamically similar experiments were performed on an HEMP-like divertor with helium and argon at inlet temperatures close to room temperature, inlet pressures below 1.4 MPa, and incident heat fluxes up to 2 MW/m^2. The results are used to predict the maximum heat flux that the divertor can accommodate, and the pumping power as a fraction of incident thermal power, for a given maximum tungsten temperature. A new nondimensional parameter, the thermal conductivity ratio, is introduced in the Nusselt number correlations which accounts for variations in the amount of conduction heat transfer through the walls of the divertor module. Numerical simulations of the HCFP divertor are performed to investigate how the thermal conductivity ratio affects predictions for the maximum heat flux obtained in previous studies. Finally, a helium loop is constructed and used to perform dynamically similar experiments on an HEMJ module at inlet temperatures as high as 300 °C, inlet pressures of 10 MPa, and incident heat fluxes as great as 4.9 MW/m^2. The correlations generated from this work can be used in system codes to determine optimal designs and operating conditions for a variety of fusion reactor designs.

Thesis (PhD)--Stellenbosch University, 2000. / ENGLISH ABSTRACT: Please refer to fulltext for abstract / AFRIKAANSE OPSOMMING: Sien asb volteks vir opsomming

In this paper, a study of tungsten metallization in advanced BEOL interconnects is presented. A mature 10 nm process is used for comparison between the tungsten and conventional copper metallization. Wafers were processed together till M1 dual-damascene etch then separated for different metallization. Tungsten metal line of 24 nm width is showing a 1.6X wire resistance comparing to the copper metal line. Comparable opens/shorts yield were obtained on a 0.8 M comb serpentine, Kelvin-via and 4K via chains. Similar physical profile were also achieved. This study has demonstrated the feasibility of replacing the copper by tungsten at BEOL using the conventional tungsten metallization tools and processes. This could be a cost- effective solution for the low-power products.

The influence of mechanical alloying (MA) milling time, temperature, sintering method and microstructure on the mechanical properties of a tungsten carbide-cobalt (WC-Co) hardmetal, based on 10wt% Co, has been established. The effects of high-energy milling for 30, 60, 180 and 300 min and the interrelation between milling time and powder properties, and the resultant effects on the mechanical properties of the consolidated WC-10Co material, has been obtained for a horizontally designed ball mill. Nanostructured WC-10Co powder was synthesised after 60 min cyclic milling at room temperature with an average WC domain size of 21 nm. In direct comparison, a WC-10Co composition MA at -30°C for 60 min produced an average WC domain size of 26 nm with a higher lattice strain. WC domain size showed a slight increase with milling time, measured at 27 nm after 300 min ball milling. Extended ball milling (300 min) reduced the mean particle size from 0.148 μm for 60 min milling to 0.117 μm. Thermal analysis showed that the onset temperature of the WC-Co eutectic was related to particle size with increased milling time reducing the onset temperature from 1344°C after 60 min milling to 1312°C after 300 min milling. Onset temperature was further reduced by the addition of vanadium carbide (VC), reducing the onset temperature to 1283°C after 300 min milling. Powder contamination increased with increased milling time with Fe content measured at ~ 3wt% after 300 min ball milling. Milling at -30°C reduced Fe contamination to an almost undetectable level. Increased ball milling time resulted in decreased levels of green density with the powders milled for 30 and 300 min achieving 62.5% and 59.5% TD, respectively. Relative density increased for the powder milled at -30°C compared to the RT milled powder due to its flattened, slightly rounded morphology. A large difference in VC starting particle size compared to WC and Co led to non-uniform dispersion of the inhibitor during milling. Densification and hardness reached optimum levels for the 60 min milled powder for both pressureless sintering and sinter-HIP. Both properties decreased with increased milling time, regardless of the sintering method. Low temperature milling resulted in a higher hardness value of 1390 HV30 compared to 1326 HV30 for the 60 min, RT milled material after pressureless sintering. Densification levels of the doped materials were restricted to < 90% TD for both sintering methods due to inhomogeneity in the microstructures. Palmqvist fracture toughness (WK) of the RT milled powders increased with increased milling time and increasing WC grain size for both sintering methods. WK reached 11.6 MN.m3/2 with 300 min milling after pressureless sintering but reached 16.1 MN.m32 for the same material after sinter-HIP due to the effect of mean WC grain size and binder phase mean free path. The -30°C milled powder exhibited higher fracture toughness for both sintering methods than the 60 min, RT milled material. Spark plasma sintering (SPS) showed that the onset of densification was dependent upon particle size with the powder from 300 min milling showing an onset temperature of ~ 800°C compared to ~ 1000°C for the 60 min milled powder. The low temperature milled powder showed an onset temperature of ~ 980°C, which suggested that low temperature milling provided enhanced densification kinetics.

As part of the effort to increase the reliability of the MMT Observatory (MMTO) 6.5m Primary Mirror Coating System, the specified filament has changed from a configuration in which the aluminum charge is hand wound around a tungsten filament to a configuration in which the aluminum is integrally wound with the tungsten at the time of filament manufacture. In the MMTO configuration, this filament consists of the three strands of tungsten wire and one strand of aluminum wire. In preparation of a full system test utilizing two hundred filaments fired simultaneously, an extensive testing program was undertaken to characterize these filaments using a four filament configuration in the MMTO small coating chamber (0.5m) and then a forty filament configuration in the University of Arizona Steward Observatory coating chamber (2m). The testing using the smaller coating chambers has shown these filaments provide very predicable coatings from test to test, and with the proper heating profile, these filaments greatly reduce the likelihood of aluminum drips. The initial filament design was modified during the course of testing by shortening the unwound filament length to closer match the aluminum load required in the MMTO coating chamber. This change increased the aluminum deposition rates without increasing the power delivered of the filament power supplies (commercial welders). Filament power levels measured at the vacuum chamber feedthroughs, currents, and deposition rates from multiple coating tests, including a full system test, are presented.

Advances in microlithography, dry etching, scaling of devices, ion-implantation, process control, and computer aid design brought the integrated circuit technology into the era of VLSI circuits. Those circuits are characterized by high packing density, improved performance, complex circuits, and large chip sizes. Interconnects and their spacing dominate the chip area of VLSI circuits and they degrade the circuit performance through the unacceptable high time delays. Multilayer metallization enables shorter interconnects, ease of design and yet higher packing density for VLSI circuits. It was shown in this dissertation that, tungsten films deposited in a cold-wall LPCVD reactor offer viable solution to the problems of VLSI multilayer interconnects. Experiments showed that LPCVD tungsten films have good uniformity, high purity, low resistivity, low stress-good adherence and are readily patterned into high resolution lines. Moreover, a multilayer interconnect system consisting of three layers of tungsten metallization followed by a fourth layer of aluminum metallization has been designed, fabricated and tested. The interlevel dielectric used to separate the metal layers was CVD phosphorus doped silicon dioxide. Low ohmic contacts were achieved for heavily doped silicon. Also, low resistance tungsten-tungsten intermetallic contacts were obtained. In addition to excellent step coverage, high electromigration resistance of interconnects was realized. Finally, CMOS devices and logic gates were successfully fabricated and tested using tungsten multilayer metallization schemes.

This work was primarily motivated by the industrial need for control of problems associated with the Gas Tungsten Arc Welding (GTAW) of small bore titanium and austenitic stainless steel tubes. These include: pore creation and entrapment in the weld zone, and variability of the fusion zone geometry. The primary aim of this study was the development of a low current orbital plasma welding capability using a structured approach which could lead to defect minimisation. The methodology should also have the potential to be used in a number of different conditions, extending the use of plasma welding in both melt-in and keyhole modes for the orbital welding of small bore tubes. The project originally involved the modification of a totally enclosed orbital GTAW welding head for low current welding operations. It was established that for the current range required for small bore and small to medium thickness tubes, the use of a solid copper torch was sufficient to provide the required heat absorption. A stable arc was produced even for very low current values (down to 7A) while arc voltages were within the operating range of a standard GTA welding power source. Procedural (i.e. off line) control was adopted for identification and optimisation of welding parameters. Since no procedure was available for the proposed welds it was necessary to generate the parameters required for the production of consistent weld profiles. Simultaneously, an expert system has been developed for the determination of optimum process parameters based on empirical models, developed using statistical techniques. Parameter combinations were selected based on physical as well as statistical relevance, providing a measure of confidence when predicting the required weld bead output characteristics. The approach also indicates the influence of the major input parameters on weld bead geometry and defect formation, such as undercut. Two quality acceptance criteria were employed during this investigation, weld bead dimensional accuracy, and the type and seriousness of defects present (penetration / burn-through, porosity and undercut). Off line programming was utilised to control heat build up and to ensure welds were obtained with the desired geometry and minimal defect levels. The end result was the development of a prototype system for low current orbital plasma welding (in both melt-in and keyhole mode) of small bore tubes in a totally enclosed head. Tolerant procedures for low current orbital melt-in and particularly keyhole welding have been generated and a systematic methodology for the prediction and optimisation of welding procedures based on predetermined criteria has been developed.

The transformations occurring in the sub-stoichiometric Ti(C,N) – W - Al system processed by high energy ball mill were investigated. The milling parameters included the milling time and the temperature comprising milling at subzero temperature and above 25°C. Two sub-stoichiometric Ti(C,N) stocks were selected, the Ti(C0.5N0.05) containing more interstitial elements than the Ti(C0.5N0.5)0.6.The transformation stages and mechanisms of alloying are discussed with respect to the changes in crystal structures of the powder constituents. The milling atmosphere had an effect on the lattice strain of milled products, and hence on the kinetics of solid state dissolution between the powder constituents, but it did not affect the fracturing process. The release of the stored crystallite lattice strain energy was the major determinant in mechanical alloying, with particle size reduction playing a necessary, but less significant role. The strain energy and the fine particle size contributed to the increased chemical reactivity with oxygen of the powders milled for shorter times. The affinity of the powders with oxygen decreased after W dissolution in Ti(C,N), and the subsequent decrease in lattice strains. The annealing behaviour of Ti(C0.5N0.05) - 40wt% W and Ti(C0.5N0.5)0.6 - 40wt% W mechanically alloyed powders were investigated using XRD, TEM, SEM and DTA techniques. It was observed that the reaction start and finish temperatures between constituents were lower in the system that had higher residual lattice strains after milling. The compositions of the intermetallic compounds and the solid solutions formed were dependent on the milling conditions and the annealing temperature. Thermal alloying was observed during annealing of Ti(C0.5N0.05) - 40wt% W mechanically alloyed products, whereas de-mixing of W-rich phases from the metastable solid solution occurred during annealing of the Ti(C0.5N0.5)0.6 - 40wt% W milled powders. The effects of Al addition and milling at subzero temperatures on the transformation of Ti(C0.5N0.05)-W powder mixtures were investigated. Addition of Al powder improved the kinetics of solid solution between powder constituents. The effect of Al was ascribed to the increase of lattice strain during short milling time followed of relaxation at longer time, and to the fast diffusion of atoms. Also, it was noticed that the high viscosity of the process control agent could inhibit the alloying process. Multiple three-component compounds could be formed. Aluminium preferably reacted with tungsten. The W(Al,C) and W(Al,Ti) formed were stable, thus solubility of W in Ti(C0.5N0.05) in the presence of Al was limited. The evolution of the morphologies of Ti(C,N)-W mixtures show that fracturing of hard particles dominated in the early stage of milling in the absence of Al, whereas with Al, plastic deformation of particles and cold welding of Ti(C,N) and W particles by the softer Al prevailed at the same time. Longer milling time improved the homogeneity and the formation of nanostructured binder pools in the sintered products. Lower oxygen contents in sintered PcBN were achieved by mechanically alloying Ti(C,N), W and Al in the high energy ball milling stage. Low level of Co in the infiltration layer was also achieved when sintering PcBN with this type of binder. A link was established between the addition of Al at the attrition milling stage and high oxygen content in the sintered PcBN, thus should be avoided. The pressure and temperature applied during sintering or annealing had a strong effect on the compositions and the crystal structures of the phases formed in the mechanically alloyed binder. The lattice strains of the binder and the PcBN were higher in the sintered materials prepared with the Ti(C0.5N0.5)0.6-W binder than in those made using the Ti(C0.5N0.05)-W alloys.