SERS investigations of additive interactions at Cu surfaces for Cu electrodeposition and Cu electroless deposition

Lindsay, Gavin Scott

Permalink

https://hdl.handle.net/2142/129832 Copy

Description

Title

SERS investigations of additive interactions at Cu surfaces for Cu electrodeposition and Cu electroless deposition

Author(s)

Lindsay, Gavin Scott

Issue Date

2025-07-09

Director of Research (if dissertation) or Advisor (if thesis)

Gewirth, Andrew A

Doctoral Committee Chair(s)

Gewirth, Andrew A

Committee Member(s)

• Rodríguez-López, Joaquín
• Shen, Mei
• Kenis, Paul

Department of Study

Chemistry

Discipline

Chemistry

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• SERS
• Electrodeposition
• Electrochemistry
• Cu
• Plating
• Spectroelectrochemistry
• Surface

Abstract

Semiconductor metallization is a key step in the production of microelectronic devices, including processing units and memory. Cu electrodeposition and Cu electroless deposition can be utilized to achieve this aim. Fundamental knowledge of these Cu deposition processes and their mechanisms is essential to keep up with increasing demands for processing power and data storage capabilities. To this aim, direct spectroscopic evidence of additives at the Cu solution interface is paramount to elucidate their interactions and mechanisms. Chapter 1 reviews the background pertinent to the metallization of interconnects and the methods utilized in studying them. Chapter 2 reports the interaction of stabilizers in tartrate-based Cu electroless baths. Chapter 3 explores the interaction of halides with the accelerator in Cu electrodeposition baths. Chapter 4 investigates the effect of the complexing agent on Cu electroless deposition baths. Chapter 5 shows the effects of anodic potentials on relative additive surface amounts during reverse-pulse plating for Cu electrodeposition. Together, this research provides fundamental knowledge of aqueous Cu deposition of common bath additives and their interactions. Chapter 2 reports the Cu electroless deposition rates in tartrate-based baths containing 2,2’ Bipyridyl (BP) or 3-mercaptobenzothiazole (MBT) in the presence and absence of Ni2+. The adsorption of BP and MBT decreases available surface sites for Cu electroless deposition and, therefore, decreases the Cu deposition rate. The addition of Ni2+ to the bath increases the Cu electroless deposition rate of baths containing BP as the stabilizing agent, but has no effect on baths with MBT as the stabilizing agent. A split-cell design, separating the formaldehyde and Cu half-reactions in combination with electrochemical measurements, yielded the mixed potential for each electroless deposition. Potential-dependent spectroelectrochemical measurements report a shift in BP adsorption to more negative potentials when Ni co-deposits with Cu, while MBT is adsorbed at all potentials studied, independent of the presence or absence of Ni. Time-dependent spectroelectrochemical measurements show a decrease in the BP adsorption rate in the presence of Ni, while the MBT adsorption rate remains the same regardless of the presence or absence of Ni. The decrease in BP adsorption in the presence of Ni increases the number of surface sites for Cu electroless deposition and therefore the Cu deposition rate when Ni is co-deposited. The lack of effect of Ni co-deposition on the adsorption of MBT causes the rate of MBT-containing baths to be independent of Ni. Chapter 3 investigates the effect of halides on the accelerator in Cu electrodeposition. Butler-Volmer fits of electrochemical data for solutions containing accelerator and halide indicate an increasing exchange current density in the order I- < Br- < Cl-. Spectroelectrochemical measurements show a halide-induced change in the orientation of the accelerator. This orientation change privileges the gauche conformation of the adsorbed accelerator in the same order. Oxygenfree contact angle measurements provide an explanation for the halide-dependent orientation, reporting increased hydrophobicity in the same order, I- < Br- < Cl-. Chapter 4 reports the Cu electroless deposition rates in baths with either HEDTA or tartrate as the complexing agents and the effect of the stabilizing agents, MBT or BP, on each of the baths. In the absence of stabilizing agents, HEDTA-based baths exhibit a slower electroless deposition rate than tartrate-based baths. In HEDTA baths, MBT, at certain concentrations, increases the Cu electroless deposition rate, while BP concentration has no effect on the electroless Cu deposition rate. Conversely, both stabilizers, MBT and BP, decrease the deposition rate in tartrate-based baths. Spectroelectrochemical measurements show adsorption of HEDTA but not of tartrate at relevant potentials. Time-dependent electrochemical in-situ spectroelectrochemical measurements report a slower adsorption of both MBT and BP in HEDTA-containing baths compared to tartrate-containing baths. These measurements also indicate competition adsorption between MBT and HEDTA for surface sites, but not between BP and HEDTA. The interactions of the stabilizers with HEDTA, as seen through SERS, suggest the accelerating mechanism of MBT is a result of increasing surface sites for Cu electroless deposition because of their competitive adsorption while BP has no effect on the Cu electroless deposition rate due to its slow desorption and lack of competition with HEDTA for surface sites. Chapter 5 explores the effect of reverse pulses on the additives at the Cu surface during Cu electrodeposition during Cu reverse-pulse plating. In-situ SERS throughout reverse pulse plating reports an increased Cl-to-accelerator ratio. Langmuir Kinetic model fits to time-dependent SERS during long pulse experiments yield the desorption rates for the additives. The SERS measurements during long pulse experiments indicate a dependence of the Cldesorption rate on the additives present, as well as a dependence on Cu+ at the surface. The desorption rates calculated from kinetic Langmuir fits to SERS data indicate that the slower desorption of Clrelative to the accelerator is the origin of the increased Cl-to-accelerator ratio in the fast pulse experiments.

Graduation Semester

2025-08

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/129832

Copyright and License Information

Copyright 2025 Gavin Lindsay

Owning Collections