A Multi-scale Model for Copper Dishing in Chemical-Mechanical Polishing

The present success in the manufacture of multi-layer interconnects in ultra-large-scale integration is largely due to the acceptable planarization capabilities of the chemical-mechanical polishing (CMP) process. In the past decade, copper has emerged as the preferred interconnect material. The greatest challenge in Cu CMP at present is the control of wafer surface non-uniformity at various scales. As the size of a wafer has increased to 300 mm, the wafer-level non-uniformity has assumed critical importance. Moreover, the pattern geometry in each die has become quite complex due to a wide range of feature sizes and multi-level structures. Therefore, it is important to develop a non-uniformity model that integrates wafer-, die- and feature-level variations into a unified, multi-scale dielectric erosion and Cu dishing model. In this paper, a systematic way of characterizing and modeling dishing in the single-step Cu CMP process is presented. The possible causes of dishing at each scale are identified in terms of several geometric and process parameters. The feature-scale pressure calculation based on the step-height at each polishing stage is introduced. The dishing model is based on pad elastic deformation and the evolving pattern geometry, and is integrated with the wafer- and die-level variations. Experimental and analytical means of determining the model parameters are outlined and the model is validated by polishing experiments on patterned wafers. Finally, practical approaches for minimizing Cu dishing are suggested.

A Multi-scale Model for Copper Dishing in Chemical-Mechanical Polishing

The present success in the manufacture of multi-layer interconnects in ultra-large-scale integration is largely due to the acceptable planarization capabilities of the chemical-mechanical polishing (CMP) process. In the past decade, copper has emerged as the preferred interconnect material. The greatest challenge in Cu CMP at present is the control of wafer surface non-uniformity at various scales. As the size of a wafer has increased to 300 mm, the wafer-level non-uniformity has assumed critical importance. Moreover, the pattern geometry in each die has become quite complex due to a wide range of feature sizes and multi-level structures. Therefore, it is important to develop a non-uniformity model that integrates wafer-, die- and feature-level variations into a unified, multi-scale dielectric erosion and Cu dishing model. In this paper, a systematic way of characterizing and modeling dishing in the single-step Cu CMP process is presented. The possible causes of dishing at each scale are identified in terms of several geometric and process parameters. The feature-scale pressure calculation based on the step-height at each polishing stage is introduced. The dishing model is based on pad elastic deformation and the evolving pattern geometry, and is integrated with the wafer- and die-level variations. Experimental and analytical means of determining the model parameters are outlined and the model is validated by polishing experiments on patterned wafers. Finally, practical approaches for minimizing Cu dishing are suggested.

In vitro evaluation of marginal leakage in bonded restorations, with mechanical or chemical-mechanical (Carisolv) removal of carious tissue

This study carried out an in vitro evaluation and comparison of the occurrence of marginal leakage in bonded restorations using mechanical or chemical-mechanical (Carisolv) removal of carious tissue. For that purpose, 40 extracted decayed human molars were divided into 4 groups: GI (burs + Prime & Bond NT + TPH), GII (Carisolv + Prime & Bond NT + TPH), GIII (burs + SBMP + Z100) and GIV (Carisolv + SBMP + Z100). After accomplishment of the restorations and thermal cycling, the teeth were exposed to dye, sectioned and qualitatively evaluated. The results demonstrated that the system of removal of carious tissue did not influence the results of microleakage at any of the cavity margins. At dentinal margins, use of the Prime & Bond NT + TPH restorative system allowed the occurrence of less microleakage than the SBMP + Z100 system.

Uniaxially aligned ceramic nanofibers obtained by chemical mechanical processing

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Investigation of the Design of a Bistable Micro-Chemical-Mechanical Device Utilizing Lateral Buckling

The Gracias Laboratory at Johns Hopkins University has developed microgrippers which utilize chemically-actuated joints to be used in micro-surgery. These grippers, however, take up to thirty minutes to close fully when activated biochemicals in the human body. This is very problematic and could limit the use of the devices in surgery. It is the goal of this research to develop a gripper that uses theGracias Laboratory's existing joints in conjunction with mechanical components to decrease the closing time. The purpose of including the mechanical components is to induce a state of instability at which time a small perturbation would cause the joint to close fully.The main concept of the research was to use the lateral buckling of a triangular gripper geometry and use a toggle mechanism to decrease the closure time of the device. This would create a snap-action device mimicking the quick closure of a Venus flytrap. All developed geometries were tested using finite element analysis to determine ifloading conditions produced the desired buckled shape. This research examines lateral buckling on the micro-scale and the possibility ofusing this phenomenon in a micro-gripper. Although a final geometry with the required deformed shaped was not found, this document contains suggestions for future geometries that may produce the correct deformed shape. It was determined through this work that in order to obtain the desired deformed shape, polymeric sections need to be added to the geometry. This simplifies the analysis and allows the triangular structure to buckle in the appropriate way due to the added joints. Future work for this project will be completed by undergraduate students at Bucknell University. Fabrication and testing of devices will be done at Johns Hopkins University in the Gracias Laboratory.

Controlling the surface roughness of epitaxial SiC on silicon

The surface of cubic silicon carbide (3C-SiC) hetero-epitaxial films grown on the (111) surface of silicon is a promising template for the subsequent epitaxial growth of III-V semiconductor layers and graphene. We investigate growth and post-growth approaches for controlling the surface roughness of epitaxial SiC to produce an optimal template. We first explore 3C-SiC growth on various degrees of offcut Si(111) substrates, although we observe that the SiC roughness tends to worsen as the degree of offcut increases. Hence we focus on post-growth approaches available on full wafers, comparing chemical mechanical polishing (CMP) and a novel plasma smoothening process. The CMP leads to a dramatic improvement, bringing the SiC surface roughness down to sub-nanometer level, though removing about 200 nm of the SiC layer. On the other hand, our proposed HCl plasma process appears very effective in smoothening selectively the sharpest surface topography, leading up to 30% improvement in SiC roughness with only about 50 nm thickness loss. We propose a simple physical model explaining the action of the plasma smoothening.

CMP流场的数值模拟及离心力影响分析

化学机械抛光(chemical mechanical polishing,CMP)是一项融合化学分解和机械力学的工艺,其中包含了流体动力润滑的作用.在已有润滑方程的基础上,提出并分析了带有离心力项的润滑方程.利用Chebyshev加速超松弛技术对有离心力项的润滑方程进行求解,得到离心力对抛光液压力分布的影响.数值模拟结果表明,压力分布与不带离心力项的润滑方程得出的明显不同;无量纲载荷和转矩随中心膜厚、转角、倾角、抛光垫旋转角速度等参数的变化趋势相同,但数值相差较大,抛光垫旋转角速度越大差别越大.

磷酸盐激光玻璃的化学机械抛光

从磷酸盐激光玻璃的成分和结构出发,分析了其化学机械抛光(CMP)机制。通过实验验证了磷酸盐激光玻璃对抛光液pH值具有较强的选择性,在微酸性和中性条件下磷酸盐激光玻璃具有较高的抛光效率,在抛光液中添加pH值调节添加剂会保持抛光环境的酸碱性从而影响抛光效率和抛光质量。通过对pH值和抛光剂浓度的控制获得了均方根(RMS)优于0.6 nm的磷酸盐激光玻璃的超光滑表面。

Effects of Cu-Wire Surface Fluctuations on Early Failures

Different chemical mechanical polishing （CMP） slurries are used to obtain single-damascene Cu-wires with different surface fluctuations as well as pre-existing surface-defects in wires with rougher surfaces. The presence of such pre-existing defects strongly increases the rate of early failures to almost 100%, reduces electromigration lifetime rapidly to the level of early failures, and changes the multimodal failure distribution into monomodal. The activation energy （0. 74±0.02eV） for the failure mechanism associated with these pre-existing defects confirms a dominant surface diffusion. It shows how a weakest link approximation analysis can he applied to a single wire by dividing the wire into relevant segments and assigning different failure mechanisms to the various segments. The analysis confirms that, although surface-defects are not the fastest early failure mechanism, the ten times higher surface-defectdensity in the rougher wires is responsible for the observed high early-failure rate and unreliable performance.

Development of germanium/silicon integration for near infrared detection

Silicon (Si) is the base material for electronic technologies and is emerging as a very attractive platform for photonic integrated circuits (PICs). PICs allow optical systems to be made more compact with higher performance than discrete optical components. Applications for PICs are in the area of fibre-optic communication, biomedical devices, photovoltaics and imaging. Germanium (Ge), due to its suitable bandgap for telecommunications and its compatibility with Si technology is preferred over III-V compounds as an integrated on-chip detector at near infrared wavelengths. There are two main approaches for Ge/Si integration: through epitaxial growth and through direct wafer bonding. The lattice mismatch of ~4.2% between Ge and Si is the main problem of the former technique which leads to a high density of dislocations while the bond strength and conductivity of the interface are the main challenges of the latter. Both result in trap states which are expected to play a critical role. Understanding the physics of the interface is a key contribution of this thesis. This thesis investigates Ge/Si diodes using these two methods. The effects of interface traps on the static and dynamic performance of Ge/Si avalanche photodetectors have been modelled for the first time. The thesis outlines the original process development and characterization of mesa diodes which were fabricated by transferring a ~700 nm thick layer of p-type Ge onto n-type Si using direct wafer bonding and layer exfoliation. The effects of low temperature annealing on the device performance and on the conductivity of the interface have been investigated. It is shown that the diode ideality factor and the series resistance of the device are reduced after annealing. The carrier transport mechanism is shown to be dominated by generation–recombination before annealing and by direct tunnelling in forward bias and band-to-band tunnelling in reverse bias after annealing. The thesis presents a novel technique to realise photodetectors where one of the substrates is thinned by chemical mechanical polishing (CMP) after bonding the Si-Ge wafers. Based on this technique, Ge/Si detectors with remarkably high responsivities, in excess of 3.5 A/W at 1.55 μm at −2 V, under surface normal illumination have been measured. By performing electrical and optical measurements at various temperatures, the carrier transport through the hetero-interface is analysed by monitoring the Ge band bending from which a detailed band structure of the Ge/Si interface is proposed for the first time. The above unity responsivity of the detectors was explained by light induced potential barrier lowering at the interface. To our knowledge this is the first report of light-gated responsivity for vertically illuminated Ge/Si photodiodes. The wafer bonding approach followed by layer exfoliation or by CMP is a low temperature wafer scale process. In principle, the technique could be extended to other materials such as Ge on GaAs, or Ge on SOI. The unique results reported here are compatible with surface normal illumination and are capable of being integrated with CMOS electronics and readout units in the form of 2D arrays of detectors. One potential future application is a low-cost Si process-compatible near infrared camera.

Morphologically Diverse Cerium Oxide Nanostructures and Nanofields for Multi functional applications.

The prospective impact of nanomaterials in science and technology has followed an increasing trend due to their unique chemical and physical properties compared to bulk. Significant advances in current technologies in areas such as clean energy production, electronics, medicine, and environment have fuelled major research and development efforts in nanotechnology around the world. This leads to the opportunity to use such nanostructured materials in novel applications and devices. Ceria, zirconia, alumina and titania are some of the major oxides which find vast applications as a nanomaterial on a wider side.