High-temperature electrical- and morphological-stability of interconnect is critical for electronic systems based on wide band gap (WBG) semiconductors. In this context, the thermal stability of both Ag and Cu films with Ta and TaN films as diffusion barriers and/or surface-capping layers at high temperatures up to 800 oC is investigated in this thesis. The investigation of un-capped Ag films with either Ta or TaN diffusion barrier layers shows electrical stability upon annealing up to 600 °C. Degradation occurs above 600 °C mainly as a result of void formation and Ag agglomeration. Sandwiching Ag films between Ta and/or TaN layers is found to electrically and morphologically stabilize the Ag metallization up to 800 °C. The barrier layer plays a key role; the β-to-α phase transition in the underlying Ta barrier layer is identified as the major cause for the morphological instability of the film above 600 °C. This phase transition can be avoided using a stacked Ta/TaN barrier. Furthermore, no observable Ta diffusion in Ag films is found. Copper films with a Ta diffusion barrier show clearly different behaviors. In the Cu/Ta sample, Ta starts to diffuse up to the surface via fast-diffusing grain boundaries (GBs) after annealing at 500 °C. The activation energy for the GB diffusion is 1.0+0.3 eV. Un-capped Cu is electrically stable up to 800 °C. An appreciable increase in sheet resistance occurs above 600 °C for the asymmetric combinations Ta/Cu/TaN and TaN/Cu/Ta. This degradation is closely related to a substantial diffusion of Ta across the Cu film and on to the TaN layer, where Ta1+xN forms. The symmetrical combinations Ta/Cu/Ta and TaN/Cu/TaN show only small changes in sheet resistance even after annealing at 800 °C. No Ta diffusion can be found in the Ta/Cu/Ta and TaN/Cu/TaN stacks. Finally, the influence of barrier and cap, their interfaces to Cu and Ta diffusion and segregation in the Cu GBs on electromigration is studied. Our preliminary results with the TaN/Cu/Ta and TaN/Cu/TaN structures report a 2-fold higher activation energy and a 10-fold longer lifetime for the former, thus confirming an important role of the interface between Cu and the cap and/or barrier.

<p>Electromigration in thin films is a well known failure mode for scaled microelectronics. While our understanding of electromigration physics has improved immensely in the last few decades, there are still some gaps in literature. In particular, the influence of interfaces on the mass transport rate is not well understood. Through reliability studies conducted on passivated metals films, marked improvement in electromigration lifetimes was observed. Specifically, some choices of materials for passivation appear to perform better than others. Qualitatively this improvement in electromigration performance is attributed to surface adhesion. However, a theoretical connection is largely missing in the literature. Lane et al. through in-situ electromigration experiments and separate interfacial debond experiments on sandwich specimens showed that a correlation exists between the void growth rate and the debond energy. However, a fundamental understanding of the relation between the two is missing. In this study we explore the connection between interfacial adhesion and void growth in a current driven system. Several experiments with varying test conditions are carried out on Blech-like test structures with different capping layers. The influence of these capping layers is captured through direct void growth measurements. Comparison of activation energy associated with electromigration was made against existing literature. It was found to be consistent with values reported for surface/interface dominated diffusion mechanisms. Further, an extension is proposed to the phase growth relations derived in existing literature to include the effect of surface adhesion. Interfacial adhesion energy ratios are extracted from the electromigration experiments for two of the test structures (Cu-Ta and Cu-SiNx) tested in this study. This ratio is compared to values reported in literature for the two interfaces and they show good agreement with experimental data.<br></p>

Basic information about nail behaviour, under passive and especially iontophoretic condition, lacks in the literature. Thus, this thesis aims to fill gaps in the nail understanding by studying the potential and feasibility of the application of iontophoresis to human nail. The iontophoretic and passive delivery of Sodium Fluorescein (SF) and Nile Blue Chloride (NBC) were studied, in vitro, in order to determine their pathways as well as their depth and uniformity of penetration into the nail. The permselective properties of the nail were investigated by characterizing the contribution of electroosmosis, using mannitol as a marker, and by studying the flux of two inorganic cations, sodium and lithium, during in vitro experiments. Finally, the feasibility of transungual iontophoresis and the extraction of sodium and chloride ions from the body through the nail plate were performed on a group of human volunteers. Iontophoresis led the fluorescent markers slightly deeper into the nail plate than passive diffusion. The delivery of the bianion and of the cation was not different. Both compounds mainly penetrated the nail via the transcellular pathway. Electroosmosis resulted only in a slight enhancement of the mannitol fluxes compared to passive diffusion and the fluxes presented high variability, especially at pH 7.4 and when the current was applied in the anode-to-cathode direction. The delivery of the two inorganic cations was significantly higher at pH 7.0 than at pH 4.0 and supported that nails hold a negative charge at physiological pH. Ions were easily extractable through the nail plate during in vivo iontophoresis and all volunteers' feedbacks supported iontophoresis as an acceptable technique. This thesis demonstrated the feasibility and potential of in vivo transungual iontophoresis.

This work was on the effect of electric current on the processing of ceramics. The focus was on electromigration/electrochemistry and plasma effects. While there is no solid evidence that there is plasma in Spark Plasma Sintering, (SPS), newer techniques e.g. flash, use different conditions so there is an interest in understanding the conditions under which a plasma forms. The minimum arcing voltage was found from literature to be from 10-15V for materials of interest. This is above that found in SPS (10V). However, due to the many contact points in a powder compact much higher voltages (50V) were required in practical experiments. Optical spectroscopy was used to verify the formation of a plasma, and emission peaks from the powder compact material were visible implying they were vaporised and formed the plasma. Electromigration was exploited to alter the oxidation of zirconium diboride, by passing current through the oxide layer (120μm zirconia base grown at 1200°C) oxygen could be pumped either away or toward the diboride bulk. Small cubes (3mm) of diboride had platinum foil electrodes applied on both sides and oxidation was performed at 1400°C for 5hr. Without a field the oxide grew to 360μm, by applying 10V and 100mA the oxide grew to 150μm under the +ve electrode but 1400μm under the -ve electrode. Electrochemical reduction was believed to have occurred due to the electrical properties of the material changing during oxidation and visible blackening of the oxide. Combining the techniques from both earlier works, a contactless flash sintering setup was developed. This used two plasma arcs as electrodes to heat and pass current through the sample. Various materials, currents and times were used, but the best result was with SiC:B4C which was sintered in 3s with 6A, the microstructure showed sharp grains, no segregation and limited grain growth ( initially 0.7μm SiC and 0.5μm B4C, this grew to 1.1μm and 1.4μm). This was the first recorded case of contactless flash sintering and the technique has the potential to sinter ceramics in a continuous manner.

An investigation has been carried out to determine the fundamental reliability unit of copper dual-damascene metallization. Electromigration experiments have been carried out on straight via-to-via interconnects in the lower metal (M1) and the upper metal (M2), and in a simple interconnect tree structure consisting of straight via-to-via line with an extra via in the middle of the line (a "dotted-I"). Multiple failure mechanisms have been observed during electromigration testing of via-to-via Cu interconnects. The failure times of the M2 test structures are significantly longer than that of identical M1 structures. It is proposed that this asymmetry is the result of a difference in the location of void formation and growth, which is believed to be related to the ease of electromigration-induced void nucleation and growth at the Cu/Si₃N₄ interface. However, voids were also detected in the vias instead of in the Cu lines for some cases of early failure of the test lines. These early failures are suspected to be related to the integrity and reliability of the Cu via. Different magnitudes and directions of electrical current were applied independently in two segments of the interconnect tree structure. As with Al-based interconnects, the reliability of a segment in this tree strongly depends on the stress conditions of the connected segment. Beyond this, there are important differences in the results obtained under similar test conditions for Al-based and Cu-based interconnect trees. These differences are thought to be associated with variations in the architectural schemes of the two metallizations. The absence of a conducting electromigration-resistant overlayer in Cu technology allows smaller voids to cause failure in Cu compared to Al. Moreover, the Si₃N₄ overlayer that serves as an interlevel diffusion barrier provides sites for easy nucleation of voids and also provides a high diffusivity path for electromigration. The results reported here suggest that while segments are not the fundamental reliability unit for circuit-level reliability assessments for Al or Cu, vias, rather than trees, might be the appropriate fundamental units for the assessment of Cu reliability. / Singapore-MIT Alliance (SMA)

Corrosion is a serious industrial concern. According to a cost of corrosion study released in 2002, the direct cost of corrosion is approximately $276 billion dollars in the United States approximately 3.1% of their Gross Domestic Product. Key influences on the severity of corrosion include: metal and electrolyte composition, temperature, turbulent flow, and location of attack. In this work, mechanistic models of localized and flow influenced corrosion were constructed and these influences on corrosion were simulated.<p>A rigourous description of mass transport is paramount for accurate corrosion modelling. A new moderately dilute mass transport model was developed. A customized hybrid differencing scheme was used to discretize the model. The scheme calculated an appropriate upwind parameter based upon the Peclet number. Charge density effects were modelled using an algebraic charge density correction. Activity coefficients were calculated using Pitzers equations. This transport model was computationally efficient and yielded accurate simulation results relative to experimental data. Use of the hybrid differencing scheme with the mass transport equation resulted in simulation results which were up to 87% more accurate (relative to experimental data) than other conventional differencing schemes. In addition, when the charge density correction was used during the solution of the electromigration-diffusion equation, rather than solving the charge density term separately, a sixfold increase in the simulation time to real time was seen (for equal time steps in both simulation strategies). Furthermore, the charge density correction is algebraic, and thus, can be applied at larger time steps that would cause the solution of the charge density term to not converge.<p>The validated mass transport model was then applied to simulate crevice corrosion initiation of passive alloys. The cathodic reactions assumed to occur were crevice-external oxygen reduction and crevice-internal hydrogen ion reduction. Dissolution of each metal in the alloy occurred at anodic sites. The predicted transient and spatial pH profile for type 304 stainless steel was in good agreement with the independent experimental data of others. Furthermore, the pH predictions of the new model for 304 stainless steel more closely matched experimental results than previous models.<p>The mass transport model was also applied to model flow influenced CO2 corrosion. The CO2 corrosion model accounted for iron dissolution, H+, H2CO3, and water reduction, and FeCO3 film formation. The model accurately predicted experimental transient corrosion rate data.<p>Finally, a comprehensive model of crevice corrosion under the influence of flow was developed. The mass transport model was modified to account for convection. Electrode potential and current density in solution was calculated using a rigourous electrode-coupling algorithm. It was predicted that as the crevice gap to depth ratio increased, the extent of fluid penetration also increased, thereby causing crevice washout. However, for crevices with small crevice gaps, external flow increased the cathodic limiting current while fluid penetration did not occur, thereby increasing the propensity for crevice corrosion.

Corrosion is a serious industrial concern. According to a cost of corrosion study released in 2002, the direct cost of corrosion is approximately $276 billion dollars in the United States approximately 3.1% of their Gross Domestic Product. Key influences on the severity of corrosion include: metal and electrolyte composition, temperature, turbulent flow, and location of attack. In this work, mechanistic models of localized and flow influenced corrosion were constructed and these influences on corrosion were simulated.<p>A rigourous description of mass transport is paramount for accurate corrosion modelling. A new moderately dilute mass transport model was developed. A customized hybrid differencing scheme was used to discretize the model. The scheme calculated an appropriate upwind parameter based upon the Peclet number. Charge density effects were modelled using an algebraic charge density correction. Activity coefficients were calculated using Pitzers equations. This transport model was computationally efficient and yielded accurate simulation results relative to experimental data. Use of the hybrid differencing scheme with the mass transport equation resulted in simulation results which were up to 87% more accurate (relative to experimental data) than other conventional differencing schemes. In addition, when the charge density correction was used during the solution of the electromigration-diffusion equation, rather than solving the charge density term separately, a sixfold increase in the simulation time to real time was seen (for equal time steps in both simulation strategies). Furthermore, the charge density correction is algebraic, and thus, can be applied at larger time steps that would cause the solution of the charge density term to not converge.<p>The validated mass transport model was then applied to simulate crevice corrosion initiation of passive alloys. The cathodic reactions assumed to occur were crevice-external oxygen reduction and crevice-internal hydrogen ion reduction. Dissolution of each metal in the alloy occurred at anodic sites. The predicted transient and spatial pH profile for type 304 stainless steel was in good agreement with the independent experimental data of others. Furthermore, the pH predictions of the new model for 304 stainless steel more closely matched experimental results than previous models.<p>The mass transport model was also applied to model flow influenced CO2 corrosion. The CO2 corrosion model accounted for iron dissolution, H+, H2CO3, and water reduction, and FeCO3 film formation. The model accurately predicted experimental transient corrosion rate data.<p>Finally, a comprehensive model of crevice corrosion under the influence of flow was developed. The mass transport model was modified to account for convection. Electrode potential and current density in solution was calculated using a rigourous electrode-coupling algorithm. It was predicted that as the crevice gap to depth ratio increased, the extent of fluid penetration also increased, thereby causing crevice washout. However, for crevices with small crevice gaps, external flow increased the cathodic limiting current while fluid penetration did not occur, thereby increasing the propensity for crevice corrosion.

The electromigration (EM) and chip-package interaction (CPI) reliability of flip chip packages with Cu pillar structures was investigated. First the EM-related characteristics of Cu pillars with solder tips were studied and compared with standard controlled collapse chip connection (C4) Pb-free solder joints. The simulation results revealed a significant reduction in the current crowding effect when C4 solder joints was replaced by Cu pillar structures. As a result, the current-induced Joule heating and local temperature gradients were reduced in the Cu pillar structure. This was followed by a study of the impact of the Cu pillar bumps on the mechanical reliability of low-k dielectrics. The CPI-induced crack driving force for delamination in the low-k interconnect structure was evaluated using a 3D sub-modeling technique. The energy release rate was found to increase significantly for packages with Cu pillar bumps compared with those with C4 Pb-free solder joints only. Structural optimization of Cu pillar bumps to improve the mechanical stability of packages with low-k chips was discussed. / text

abstract: For decades, microelectronics manufacturing has been concerned with failures related to electromigration phenomena in conductors experiencing high current densities. The influence of interconnect microstructure on device failures related to electromigration in BGA and flip chip solder interconnects has become a significant interest with reduced individual solder interconnect volumes. A survey indicates that x-ray computed micro-tomography (µXCT) is an emerging, novel means for characterizing the microstructures' role in governing electromigration failures. This work details the design and construction of a lab-scale µXCT system to characterize electromigration in the Sn-0.7Cu lead-free solder system by leveraging in situ imaging. In order to enhance the attenuation contrast observed in multi-phase material systems, a modeling approach has been developed to predict settings for the controllable imaging parameters which yield relatively high detection rates over the range of x-ray energies for which maximum attenuation contrast is expected in the polychromatic x-ray imaging system. In order to develop this predictive tool, a model has been constructed for the Bremsstrahlung spectrum of an x-ray tube, and calculations for the detector's efficiency over the relevant range of x-ray energies have been made, and the product of emitted and detected spectra has been used to calculate the effective x-ray imaging spectrum. An approach has also been established for filtering `zinger' noise in x-ray radiographs, which has proven problematic at high x-ray energies used for solder imaging. The performance of this filter has been compared with a known existing method and the results indicate a significant increase in the accuracy of zinger filtered radiographs. The obtained results indicate the conception of a powerful means for the study of failure causing processes in solder systems used as interconnects in microelectronic packaging devices. These results include the volumetric quantification of parameters which are indicative of both electromigration tolerance of solders and the dominant mechanisms for atomic migration in response to current stressing. This work is aimed to further the community's understanding of failure-causing electromigration processes in industrially relevant material systems for microelectronic interconnect applications and to advance the capability of available characterization techniques for their interrogation. / Dissertation/Thesis / Doctoral Dissertation Materials Science and Engineering 2015

Electromigration (EM) and thermomigration (TM) reliability of Pb-free solder joints are emerging as critical concerns in advanced packages. In this study, EM and TM phenomena in Sn-2.5Ag solder joints with thick Cu or thin Ni under-bump metallurgy (UBM) were investigated. A series of EM tests were performed to obtain activation energy (Q) and current density exponent (n), and to understand failure mechanisms. Joule heating was also taken into account. Q and n values were determined as follows: for Cu UBM solders, Q = 1.0 eV and n = 1.5; for Ni UBM solders, Q = 0.9 and n = 2.2. Important factors limiting EM reliability of Pb-free solder joints were found to be UBM dissolution with extensive intermetallic compound (IMC) growth and current crowding. IMC growth without current stressing was found to follow the parabolic growth law whereas linear growth law was observed for Cu₆Sn₅ and Ni₃Sn₄ under high current stressing. For Cu UBM solders, the apparent activation energy for IMC growth was consistent with the activation energy for EM, which supports that EM failure was closely related to IMC growth. In contrast, for Ni UBM solders the apparent activation energy was higher than the EM activation energy. It was suggested that the EM failure in the Ni UBM solders could be associated with more than one mass transport mechanism. The current crowding effect was analyzed with different thicknesses of Ni UBM. It was found that the maximum current density in solder could represent the current density term in Black's equation better than the average current density. FEM studies demonstrated that current crowding was mainly controlled by UBM thickness, metal trace design, and passivation opening diameter. A large temperature gradient of the order of 10³ °C/cm was generated across the sample to induce noticeable TM and to compare its effect against that of EM. TM-induced voiding was observed in Ni UBM solders while UBM dissolution with IMC formation occurred in Cu UBM solders. However, the relative effect of TM was found to be several times smaller than that of EM even at this large temperature gradient. / text