Modeling Macro-Sized, High Aspect Ratio Through-Hole Filling by Multi-Component Additive-Assisted Copper Electrodeposition

Abstract

A multi-element, time-dependent model is used to examine additive-assisted copper electroplating in macro-channels. This model is an adaptation of the work of Akolkar and Landau [J. Electrochem. Soc., 156, D351 (2009)], used to describe plating in micro-vias for integrated circuits. Using their method for describing species movement in the channel, the model has been expanded to include transport and adsorption limitations of the inhibitor and accelerator, as well as the copper ions in solution. The model is used to investigate copper plating as an infiltration method across many size scales and aspect ratios. Biomorphic graphite scaffolds produced from wood are used as a representative system and the results of a two-additive bath are used to characterize the behavior of the additives and determine the effectiveness of the plating. The results indicate that at macro-scales, channel dimensions play an increasingly important role in dictating the behavior of additive-assisted plating. Because additive systems are designed to establish differential surface coverage within the channel, the success of which is determined by the additive's rates of diffusion and adsorption, certain size scale/aspect ratio combinations preclude such coverage. A guide for sample geometries that may be successfully infiltrated with a two-additive bath is provided. Electrodeposition is a common method of producing multicomponent material systems.1,2 This technique is useful in overcoming common melt-based composite processing constraints, such as high melting points, fragile scaffolds, non-wetting components, and significant coefficient of thermal expansion (CTE) mismatch. Unlike melt infiltration or reactive wetting, electrodeposition is a low-temperature, non-reactive process. While electroformed copper is employed in the fabrication of items such as musical instruments, reflectors, and heat exchangers, 1 the electronics sector is responsible for the majority of advancements in the field, especially in terms of use as an infiltration method. Electrodeposition is used for interconnect metallization in integrated circuits, where small channels in a silicon wafer, known as vias, are filled with high purity copper. IBM pioneered this technique in 1998 to replace aluminum vapor deposition.3,4 While the microvias used in current integrated circuit technology can reach several hundred microns in dimension, channel size is trending toward nanosized, closely packed channels. [5][6][7] In other functional composites, one might contend with larger length scales and aspect ratios compared to metal interconnects. With this as motivation, we explore the feasibility of using existing electroplating techniques for composite processing of non-interconnect systems through a multi-component, time-dependent model. Specifically, we investigate infiltration of a metal into highly porous ceramic scaffolds with macro-sized channels.As a macro-porous system for study, we examine biomorphic SiCand graphite-Cu composites produced by electrodeposition, which may ...

RETRACTED: Copper Electrodeposition in Through-Silicon Via under Galvanostatic Condition

“…37,44 Considered that the diffusive coefficient of accelerator (SPS) is two orders of magnitude bigger than that of suppressor (PEG), 3,25,45 the difference in diffusion rate of SPS and PEG was first taken into model to explain the inhibited via-top and accelerated via-bottom in via. Diffusion of additives in electrolyte is described by Fick's law…”

Interfacial frictional stresses and fracture toughness of biomorphic graphite/copper interfaces

“…The surface was abraded with 120-grit sandpaper to provide roughness similar to that of the graphitized surface in the scaffold. Copper was deposited on the 12 x 7 mm 2 face of one of the rods using an electrodeposition process similar to those described in [24] and [25].…”

A novel model for through-silicon via (TSV) filling process simulation considering three additives and current density effect

“…Akolkar and Landau [20] developed a model that accounts for the time and position dependences of both the PEG and the SPS transport, adsorption, and interaction effects; however, the depletion of the cupric ion and additives within the via has been ignored [21,22]. Based on this model, Childers et al proposed a model considering the effect of diffusion of both the cupric ions and the additives (accelerator and suppressor) [23]. Furthermore, a physicochemical model, which included the diffusion, adsorption, desorption, and incorporation of two additives (accelerator and suppressor) and ions (cupric ion and chloride ion) is presented by Zhang et al [24,25].…”