This dissertation presents a series of studies related to the slurry mean residence time analysis and the pad-wafer contact characterization in Chemical Mechanical Planarization (CMP). The purpose of these studies is to further understand the fundamentals of CMP and to explore solutions to some of CMP's challenges. Mean residence time (MRT) is a widely used term that is mostly seen in classical chemical engineering reactor analysis. In a CMP process, the wafer-pad interface can be treated as a closed system reactor, and classical reactor theory can be applied to the slurry flow through the region. Slurry MRT represents the average time it takes for fresh incoming slurry to replace the existing slurry in the region bound between the pad and the wafer. Understanding the parameters that have an impact on MRT, and therefore removal rate, is critical to maintain tight specifications in the CMP process. In this dissertation, we proposed a novel slurry injection system (SIS) which efficiently introduced fresh slurry into the pad-wafer interface to reduce MRT. Results indicated that SIS exhibited lower slurry MRT and dispersion numbers but higher removal rates than the standard pad center slurry application by blocking the spent slurry and residual rinse water from re-entering the pad-wafer interface during polishing. Another study in this dissertation dealt with the effect of pad groove width on slurry MRT in the pad-wafer interface as well as slurry utilization efficiency (η). Three concentrically grooved pads with different groove widths were tested at different polishing pressures to experimentally determine the corresponding MRT using the residence time distribution (RTD) technique. Results showed that MRT and η increased significantly when the groove width increased from 300 to 600μm. On the other hand, when the groove width increased further to 900μm, MRT continued to increase while n remained constant. Results also indicated that MRT was reduced at a higher polishing pressure while η did not change significantly with pressure for all three pads. In the last study of this dissertation, the effect of pad surface micro-texture on removal rate during tungsten CMP was investigated. Two different conditioner discs ("Disc A" and "Disc B") were employed to generate different pad surface micro-textures during polishing. Results showed that "Disc B" generated consistently lower removal rates and coefficients of friction than "Disc A". To fundamentally elucidate the cause(s) of such differences, pad surface contact area and topography were analyzed using laser confocal microscopy. The comparison of the pad surface micro-texture analysis on pad surfaces conditioned by both discs indicated that "Disc A" generated a surface having a smaller abruptness (λ) and more solid contact area which resulted in a higher removal rate. In contrast, "Disc B" generated many large near-contact areas as a result of fractured and collapsed pore walls.

Nanoparticles (NPs) have unique electronic, optical and chemical properties due to the extreme small size. Engineered nanoparticles (ENPs) are intentionally produced for desired applications, with specific properties related to shape, size, surface properties and chemistry. Nano-sized silica (SiO2), alumina (Al2O3) and ceria (CeO2) are three important ENPs with large production and wide applications. One of the principal uses of these ENPs is in chemical and mechanical planarization (CMP), a key process applied to polish wafers when fabricating integrated circuits in semiconductor manufacturing, in which SiO2, Al2O3 and CeO2 NPs are used as abrasive particles in CMP slurries. CMP generates large amounts of waste effluents containing high levels of ENPs. Some ENPs have been proven to be able to cause toxicity to microorganisms and higher life forms, including humans. Therefore, there are concerns about the potential risks that ENPs may pose to the natural environment and human health. In addition, III-V materials like indium arsenide (InAs) and gallium arsenide (GaAs) are increasingly used in electronic and photovoltaic devices. Besides ENPs, the waste streams from III-V manufacturing also contain dissolved and particulate materials removed from III-V films during CMP. Arsenic is one of the most notorious contaminants that has been widely studied, while only very limited ecotoxicity information is available for gallium and indium. Finally, since ENPs have high surface area, it is very likely they will interact with the soluble species (such as arsenic ions) in CMP wastewater. Therefore, it is of great importance to understand whether the interactions between these materials could alter their fate and toxicity. The objective of this work is to investigate the potential environmental and health risks from the ENPs and III-V materials used in semiconductor manufacturing. To this end, the physical, chemical and toxicological characterization of ENPs used in CMP was performed (Chapter 3). Furthermore, the fate and transport of the most used ENP, SiO2, in porous media was studied (Chapter 4). In addition, acute toxicity of As(III), As(V), In(III) and Ga(III) species was evaluated using different bioassays (Chapter 5). Finally, the cytotoxicity of ENPs used in CMP slurries to human lung bronchial epithelial cells was evaluated using an impedance based real time cell analysis (RTCA) assay (Chapter 6). In Chapter 3, four model slurries containing ENPs including colloidal silica (c-SiO2), fumed silica (f-SiO2) cerium oxide (CeO2) and aluminum oxide (Al2O3) were characterized for their physical, chemical and toxicological properties. Ecotoxicity of these slurries to the marine bacterium, Aliivibrio fischeri, was evaluated by measuring its bioluminescence activity as a function of the ENP concentration dosed. The results showed that f-SiO2 and CeO2 were not toxic at concentrations up to 700 and 1000 mg/L, respectively. On the other hand, c-SiO2 and Al2O3 were inhibitory only at very high concentrations (>600 mg/L). At about 1300 mg/L, c-SiO2 and Al2O3 led to 37.6% and 28.4% decrease of cell activity after 30 min exposure, respectively. The inhibitory effect from c-SiO2 was related to additives in the slurry. In summary, the results indicate that these slurries are not likely to cause acute toxicity at environmentally relevant concentrations. The potential risks from ENPs are dependent on their fate and transport in the environment. In Chapter 4, the transport and abatement of SiO2 NPs was studied through laboratory scale column experiments. Synthetic fluorescent core-shell SiO2 NPs (83 nm) were used to facilitate NP traceability. Three widely used filtering materials, i.e., sand, anthracite and granular activated carbon (GAC), were used as porous media. Sand showed very poor capacity for the filtration of SiO2 NPs due to its limited surface area and high concentration of negative surface charge. In addition, the stability and transport of SiO2 NP was strongly dependent on the ionic strength of the solution. High ionic strength led to NP agglomeration and facilitated SiO2 NP retention, while low ionic strength resulted in release of captured NPs from the sand bed. Compared to sand, anthracite and GAC showed higher efficiency for SiO2 NP capture. The superior capacity of GAC was primarily due to its porous structure and high surface area. A process model was developed to simulate NP capture in the packed bed columns and determine fundamental attraction parameters. This model provided an excellent fit to the experimental data. Taken together the results obtained indicate that GAC is an interesting material for SiO2 NPs filtration. With the increasing usage of III-V materials, there are concerns about the ecological threats posed by III-V ions released during semiconductor manufacturing and from disposal of decommissioned electronic devices. In Chapter 5, the acute toxicity of As(III), As(V), In(III) and Ga(III) species was evaluated using different bioassays, including three microbial assays, testing for methanogenic activity, O2 uptake and bioluminescence inhibition of marine bacterium A. fischeri. Acute toxicity to the freshwater crustacean Daphnia magna was also tested. The results showed that In(III) and Ga(III) were generally not toxic or only mildly toxic in all assays, while both As(III) and As(V) showed strong inhibitory effects on different microbial activities (methanogenic and bioluminescence). The toxicity of these ions was strongly dependent on the bioassay target. For In(III) and Ga(III), D. magna was the most sensitive organism with 50% lethal concentrations (LC50) of 57.4 and 237.0 mg/L, respectively. On the other hand, As(III) and As(V) were particularly toxic to methanogens. The 50% inhibitory concentrations (IC50) of both species were about 1.5mg/L. Mixed aerobic heterotrophic culture was highly resistant to all four ions and O2 uptake by the aerobes was not affected in the tested concentrations. Overall, the results indicate that the ecotoxicity of In(III) and Ga(III) is much lower than that of the As species. This finding is important in filling the knowledge gap regarding the ecotoxicology of In and Ga. Besides ecotoxicity, ENPs and III-V materials in CMP effluents could also pose a threat to human health. In Chapter 6, the cytotoxicity of CMP slurries to human bronchial epithelial cells (16HBE14o-) was assessed using a novel impedance based real time cell analyzer (RTCA). Cell death and detachment was observed in assays supplied with high concentrations of c-SiO2 and f-SiO2 NPs (≥250 mg/L). On the other hand, CeO2 and Al2O3 slurries were not inhibitory at concentrations up to 1250 mg/L. In addition, since CMP wastewater generated during the planarization of III-V films contains a mixture of ENPs and soluble III-V species, it is important to understand whether the interactions between these materials could alter their fate and toxicity. As(III) toxicity to human lung cells in the presence and absence of CeO2 NPs was evaluated using the RTCA assay. Exposure to As(III) (0.5 mg/L) for 48 h resulted in 81.3% inhibition of cell viability and proliferation, while cell inhibition decreased to only 13.0% when As(III) was dosed together with sub-toxic levels of CeO2 NPs (250 mg/L). This detoxification effect was mainly due to As(III) adsorption onto CeO2 NPs. When the NPs were added, the soluble arsenic concentration was reduced significantly from 0.5 mg/L to 0.03 mg/L. This work demonstrates that adsorption of As(III) on CeO2 NPs can lower As(III) concentration in the solution and reduce its bioavailability and subsequently result in As(III) detoxification. In conclusion, this dissertation indicates that the ENPs (SiO2, CeO2 and Al2O3) used in semiconductor industry are not expected to cause acute toxicity to the natural environment and human health under environmentally relevant concentration (<1 mg/L). Among the soluble III-V species, In(III) and Ga(III) showed no or mild acute inhibitory effects in different bioassays even at comparatively high concentration. However arsenic species are highly toxic to various important microbial populations in the environment and human cells. The results showed that arsenic could induce toxic effects under current discharge limit set for semiconductor industry. Finally, we demonstrated that the adsorption of As(III) on CeO2 NPs can lower the concentration of soluble As(III) and subsequently resulted in As(III) detoxification.

Run-by-Run (RbR) control is an online supervisory control strategy designed for the batch manufacturing industry. The objective of RbR control is to minimize process drift, shift and variability between machine runs, thereby reducing costs. The most widely used RbR controllers use the Exponentially Weighted Moving Average (EWMA) filter. However, the linear nature of the EWMA filter makes these RbR controllers inefficient for processes with features at multiple frequencies (also known as multiscale processes). Recent developments in wavelet theory have enhanced the ability to analyze events in multiscale processes. New RbR control strategies have started to emerge that incorporate wavelet analysis. These controllers, developed at the University of South Florida, seem to be robust in dealing with multiscale processes. The objective of this research is to integrate the wavelet based, multiscale analysis approach with the existing double EWMA RbR control strategy for controlling a multiple input multiple output (MIMO) process. The new controller (WRbR controller) is applied on a Chemical Mechanical Planerization process having four inputs and two outputs. A continuous drift and mean shift are introduced in the process, which is then controlled using both the existing double EWMA and the new wavelet based RbR controllers. The results indicate that the wavelet based controller is better in terms of the average square deviation and the standard deviation in the process outputs. Moreover, the observed decrease in the magnitude of the average absolute input deviation indicates a smoother process operation.

Due to the recent rapid growth in personal mobile communication devices (smartphones, PDA's, tablets, etc.), the wireless market is always looking for new ways to further miniaturize the RF front-ends while reducing the cost and power consumption. For many years, wireless transceivers and subsystems have been relying on high quality factor (Q) passives (e.g., quartz crystal, ceramics) to implement oscillators, filters, and other key RF front-end circuitry elements. However, these off-chip discrete components occupy large chip area and require power-demanding interfacing circuits. As a result, a great deal of research effort has been devoted to the development of micromechanical resonators that are much more amenable to direct integration with integrated circuit (IC). Over the past few years, vibrating RF MEMS (Micro-Electrical-Mechanical-System) resonator technology has emerged as a viable solution, most notably, the film bulk acoustic resonator (FBAR) and surface acoustic wave (SAW) resonator, which have already been successfully implemented into commercial products. Undoubtedly, micromechanical resonators such as FBAR's can perform as well as if not better than its bulky conventional counterparts and facilitate the miniaturization and power reduction of conventional RF systems. However, in some cases when multi-frequency functionality on a single-chip is needed, FBAR simply won't deliver. To address this dilemma, contour-mode MEMS resonators have been developed and regarded as the most viable on-chip high-Q alternative. Unlike FBAR, contour-mode resonators use lateral dimensions to define its resonating frequencies, thus allowing for single-chip multi-frequency functionality. However, there is still room for improvement with respect to lowering the motional resistance of these devices to allow matching to 50 Ω electronics, while retaining low power consumption, small size, and simpler manufacturing process. This dissertation presents the design, fabrication, characterization and experimental analysis of two types of micro-mechanical resonators. Piezoelectrically- and electrostatically-transduced micromechanical resonators will both be shown. Both types of resonator will be fabricated in the same micro-fabrication run, which makes the comparison between the two much more impartial. The impacts of substrate's resistivity over the device performances will also be studied. Among the most significant results, this dissertation also presents several ideas that are enabled by the use of silicon-on-insulator (SOI) wafer. A novel single-mask fabrication process that can produce capacitive resonator with nano-meter gap is demonstrated. The concept of dual-transduced micro-mechanical resonator is introduced by combining both piezoelectric and capacitive based resonators. Finally, frequency tuning of MEMS resonator are explored and detailed in this work as well.

The production of recombinant glycoproteins from mammalian cell cultures requires robust processes that can achieve high protein yield while ensuring the efficacy of these proteins as human therapeutics. We describe two approaches currently being developed in our group to genetically engineer cell lines with desirable characteristics for recombinant protein production. To enhance the degree of sialylation in the glycoprotein product, we propose to increase intracellular sialic acid availability by overexpressing the CMP-sialic acid transporters. We are also interested in engineering mammalian cells that can proliferate at reduced cultivation temperatures. Low temperature cultivation of mammalian cells has been shown to enhance glycoprotein production but reduces cell growth. It is hypothesized that a mutant cell line that can proliferate at low temperatures may be coupled with low temperature cultivation to improve recombinant protein production. / Singapore-MIT Alliance (SMA)

As semiconductor technology advances into the nano-scale era and more functional blocks are added into systems on chip (SoC), the interface between circuit design and manufacturing is becoming blurred. An increasing number of features, traditionally ignored by designers are influencing both circuit performance and yield. As a result, design tools need to incorporate new factors. One important source of circuit performance degradation comes from deterministic within-die variation from lithography imperfections and Cu interconnect chemical mechanical polishing (CMP). To determine how these within-die variations impact circuit performance, a new analysis tool is required. Thus a methodology has been proposed to involve layout-dependent within-die variations in static timing analysis. The methodology combines a set of scripts and commercial tools to analyze a full chip. The tool has been applied to analyze delay of ISCAS85 benchmark circuits in the presence of imperfect lithography and CMP variation. Also, this thesis presents a methodology to generate test sets to diagnose the sources of within-die variation. Specifically, a delay fault diagnosis algorithm is developed to link failing signatures to physical mechanisms and to distinguish among different sources of within-die variation. The algorithm relies on layout-dependent timing analysis, path enumeration, test pattern generation, and correlation of pass/fail signatures to diagnose lithography-caused delay faults. The effectiveness in diagnosis is evaluated for ISCAS85 benchmark circuits.

ABSTRACT In this study, two chemical mechanical polishing (CMP) wastewaters were treated by crossflow electro-microfiltration. Also studied are the effects of operation parameters on their treatment efficiencies. In the semiconductor industry, presently, CMP has become the key technique to provide global planarization on interlevel dielectrics (ILD) and metal layers of wafers. In general, the post-CMP cleaning process will produce a great quantity of CMP wastewater. Normally, CMP wastewater consists of abrasives of high concentration and stability, chemicals (e.g., oxidant and surfactant), and a tremendous mass of de-ionized water. Because of the negatively charged suspended solids in CMP wastewater, crossflow electro-microfiltration was used to treat this type of wastewater. By applying an electric field to the system, the negatively charged suspended solids were expelled from the membrane surface moving toward the anode. Not only reducing the cake formation on the membrane, enhancement of the filtration rate and permeate flux have also been found when an external electric field is applied to the filtration system. In this investigation, CMP wastewaters obtained from wafer fabs A and B were first characterized by various standard methods. In CMP wastewater A, the suspended solids were found to have a high negative zeta potential, about ¡V78 mV. Its electrical conductivity was determined to be 127.2 £gS/cm. Before testing, each CMP wastewater was pre-filtered using a filter paper of 1.2 £gm in pore size. An experimental design based on the Taguchi method was employed. The L9 orthogonal arrays were utilized to investigate the effects of four experimental factors ( i.e., electric field strength, crossflow velocity, transmembrane pressure, and membrane pore size) on the filtration rate and permeate quality in the crossflow electro-microfiltration system. When the electric field strength applied was lower than the critical electric field strength, increases of the electric field strength, transmembrane pressure, and membrane pore size were found to be beneficial to the filtration rate. The experimental results were further subjected to the analysis of variance and regular analysis. For both CMP wastewaters A and B, the electric field strength and membrane pore size were determined to be very significant parameters. In this filtration system, the optimal treatment efficiency could be achieved by using a higher electric field strength, lower crossflow velocity, higher transmembrane pressure, and larger membrane pore size. The quality of permeate thus obtained was even better than the tap water quality standards. Therefore, the permeate might be worth recycling for various purposes.

ABSTRACT In this study, two chemical mechanical polishing (CMP) wastewaters were treated by crossflow electro-microfiltration. Also studied are the effects of operation parameters on their treatment efficiencies. In the semiconductor industry, presently, CMP has become the key technique to provide global planarization on interlevel dielectrics (ILD) and metal layers of wafers. In general, the post-CMP cleaning process will produce a great quantity of CMP wastewater. Normally, CMP wastewater consists of abrasives of high concentration and stability, chemicals (e.g., oxidant and surfactant), and a tremendous mass of de-ionized water. Because of the negatively charged suspended solids in CMP wastewater, crossflow electro-microfiltration was used to treat this type of wastewater. By applying an electric field to the system, the negatively charged suspended solids were expelled from the membrane surface moving toward the anode. Not only reducing the cake formation on the membrane, enhancement of the filtration rate and permeate flux have also been found when an external electric field is applied to the filtration system. In this investigation, CMP wastewaters obtained from wafer fabs A and B were first characterized by various standard methods. In CMP wastewater A, the suspended solids were found to have a high negative zeta potential, about ¡V78 mV. Its electrical conductivity was determined to be 127.2 £gS/cm. Before testing, each CMP wastewater was pre-filtered using a filter paper of 1.2 £gm in pore size. An experimental design based on the Taguchi method was employed. The L9 orthogonal arrays were utilized to investigate the effects of four experimental factors ( i.e., electric field strength, crossflow velocity, transmembrane pressure, and membrane pore size) on the filtration rate and permeate quality in the crossflow electro-microfiltration system. When the electric field strength applied was lower than the critical electric field strength, increases of the electric field strength, transmembrane pressure, and membrane pore size were found to be beneficial to the filtration rate. The experimental results were further subjected to the analysis of variance and regular analysis. For both CMP wastewaters A and B, the electric field strength and membrane pore size were determined to be very significant parameters. In this filtration system, the optimal treatment efficiency could be achieved by using a higher electric field strength, lower crossflow velocity, higher transmembrane pressure, and larger membrane pore size. The quality of permeate thus obtained was even better than the tap water quality standards. Therefore, the permeate might be worth recycling for various purposes.

This dissertation presents a series of studies relating to kinetics and kinematics of inter-layer dielectric and metal chemical mechanical planarization processes. These are also evaluated with the purposes of minimizing environmental and cost of ownership impact.The first study is performed to obtain the real-time substrate temperature during the polishing process and is specifically intended to understand the temperature distribution across the polishing wafer during the chemical mechanical planarization process. Later, this technique is implemented to study the effect of slurry injection position for optimum slurry usage. It is known that the performance of chemical mechanical planarization depends significantly on the polishing pad and the kinematics involved in the process. Variations in pad material and pad grooving type as well as pressure and sliding velocity can affect polishing performance. One study in this dissertation investigates thermoset and thermoplastic pad materials with different grooving methods and patterns. The study is conducted on multiple pressure and sliding velocity variations to understand the characteristic of each pad. The analysis method elaborated in this study can be applied generically.A subsequent study focuses in a slurry characterization technique. Slurry, a critical component in chemical mechanical planarization, is typically a water-based dispersion of fine abrasive particles with various additives to control material removal rate and microscratches. Simultaneous turbidity and low angle light scattering methods under well-defined mixing conditions are shown to quantify the stability of abrasive particle from aggregations. Further contribution of this dissertation involves studies related to the spectral analysis of raw shear force and down force data obtained during chemical mechanical planarization. These studies implemented Fast Fourier Transforms to convert force data from time to frequency domain. A study is performed to detect the presence of larger, defect-causing particles during polishing. In a further application on diamond disc conditioning work is performed to achieve optimum break-in time and an optimum conditioning duty cycle. Studies on spectral analysis are also extended to planarization of shallow trench isolation pattern wafers to monitor the polishing progress in real-time.

In chip-multiprocessor (CMP) designs, when the number of core increases, the size of on-chip communication fabric and data storage grows accordingly and therefore the chip power challenge is exacerbated. This thesis work considers the power management for networks-on-chip (NoC) and the last level cache, which constitute the uncore in CMP designs. NoC is regarded as a scalable approach to cope with the increasing demand for on-chip communication bandwidth. The last level cache is shared among all cores. The focus of this work is on the control techniques for uncore dynamic voltage and frequency scaling. A realistic but not well-studied scenario is investigated. That is, the entire uncore shares a single voltage/frequency domain, as opposed to separated domains in most of previous works. One appealing advantage here is that data packets no longer experience the interfacing overhead across different voltage/frequency domains. The classic PI (Proportional and Integral) control method is adopted due to its simplicity, flexibility and low implementation overhead. This thesis research outcome includes three parts. First, stability of the PI control is analyzed. Second, a model-assisted PI control scheme is proposed and studied. The model assist is to address the problem that no universally good reference point exists for the control. Third, the windup issue for the PI control is investigated. Full architecture simulations are performed on public benchmark suites to validate the proposed techniques. The result show 76% energy reduction with less than 6% performance degradation compared to constantly high voltage/frequency for uncore.