Y. Nagendra Prasad

Bio: Y. Nagendra Prasad is an academic researcher from Hanyang University. The author has contributed to research in topics: Chemical-mechanical planarization & Polishing. The author has an hindex of 8, co-authored 11 publications receiving 217 citations. Previous affiliations of Y. Nagendra Prasad include Indian Institute of Technology Madras.

Analysis of Scratches Formed on Oxide Surface during Chemical Mechanical Planarization

Abstract: Scratch formation on patterned oxide wafers during the chemical mechanical planarization process was investigated. Silica andceria slurries were used for polishing the experiments to observe the effect of abrasives on the scratch formation. Interleveldielectric patterned wafers were used to study the scratch dimensions, and shallow trench isolation patterned wafers were used tostudy the effect of polishing parameters, such as pressure and rotational speed head/platen . Similar shapes of scratches chattertype were observed with both types of slurries. The length of the scratch formed might be related to the period of contact betweenthe wafer and the pad. Large particles would play a significant role in increasing the number of scratches. The probability ofscratch generation is more at higher pressures due to higher friction force and removal rate. The optimization of the head to platenvelocity could decrease the number of scratches.© 2009 The Electrochemical Society. DOI: 10.1149/1.3265474 All rights reserved.Manuscript submitted August 10, 2009; revised manuscript received October 26, 2009. Published December 15, 2009.

Role of Amino-Acid Adsorption on Silica and Silicon Nitride Surfaces during STI CMP

Abstract: Selectivity (oxide/nitride polish rate) is a critical factor during chemical mechanical polishing (CMP) of shallow trench isolation (STI) structure, and it can be modified by adding amino acids to the slurry. The role of adsorption of the amino acids L-proline and L-arginine, on silicon dioxide and silicon nitride surfaces was characterized as a function of pH and concentration using thermogravimetric analysis. The results suggest that the adsorption behavior does not correlate with the polishing behavior of STI CMP and hence it may not play a key role in changing the selectivity.

Chemical mechanical planarization: slurry chemistry, materials, and mechanisms.

Abstract: 3. FEOL Applications: Device Level 180 3.1. Shallow Trench Isolation (STI) CMP 180 3.2. Replacement Metal Gate CMP 184 3.3. Poly-Si CMP for FinFET Devices 186 4. MOL Applications: Contact Level 187 4.1. Tungsten CMP 187 5. BEOL Applications: Multilevel Interconnects 188 5.1. Copper Interconnect Technology 188 5.2. CMP Challenges in Cu Interconnects 189 5.2.1. Low-k and Ultralow-k Material Challenges 189 5.2.2. Integration Challenges 190 5.2.3. CMP Process Challenges 191 5.3. Copper Planarizarion Process 191 5.4. Ta/TaN Liner CMP Process 193 6. CMP Process-Induced Defects 194 6.1. Defects in FEOL CMP 194 6.2. Defects in MOL Tungsten CMP 195 6.3. Defects in Cu BEOL CMP 195 6.3.1. Corrosion of Copper 196 6.3.2. Scratches 196 6.3.3. Dishing, Erosion, and Trenching 196 6.3.4. Mechanical Damage 197 6.3.5. Other Defects 197 7. Models of CMP Processes 198 7.1. Models Based on Contact Mechanics 198 7.2. CMP Process Models 199 8. Alternative CMP Processes 200 9. Concluding Remarks 201 10. Acknowledgments 201 11. References 201

Sub-surface mechanical damage distributions during grinding of fused silica

Abstract: The distribution and characteristics of surface cracking (i.e. sub-surface damage or SSD) formed during standard grinding processes has been investigated on fused silica glass. The SSD distributions of the ground surfaces were determined by: (1) creating a shallow (18-108 {micro}m) wedge/taper on the surface by magneto-rheological finishing; (2) exposing the SSD by HF acid etching; and (3) performing image analysis of the observed cracks from optical micrographs taken along the surface taper. The observed surface cracks are characterized as near-surface lateral and deeper trailing indent type fractures (i.e., chatter marks). The SSD depth distributions are typically described by a single exponential distribution followed by an asymptotic cutoff in depth (c{sub max}). The length of the trailing indent is strongly correlated with a given process. Using established fracture indentation relationships, it is shown that only a small fraction of the abrasive particles are being mechanically loaded and causing fracture, and it is likely the larger particles in the abrasive particle size distribution that bear the higher loads. The SSD depth was observed to increase with load and with a small amount of larger contaminant particles. Using a simple brittle fracture model for grinding, the SSD depth distribution has been related tomore » the SSD length distribution to gain insight into ''effective'' size distribution of particles participating in the fracture. Both the average crack length and the surface roughness were found to scale linearly with the maximum SSD depth (c{sub max}). These relationships can serve as useful rules-of-thumb for nondestructively estimating SSD depth and to identify the process that caused the SSD. In certain applications such as high intensity lasers, SSD on the glass optics can serve as a reservoir for minute amounts of impurities that absorb the high intensity laser light and lead to subsequent laser-induced surface damage. Hence a more scientific understanding of SSD formation can provide a means to establish recipes to fabricate SSD-free, laser damage resistant optical surfaces.« less