Additive Transport, Adsorption, and Inclusion in Electroless Deposition of Copper

Abstract

Not requiring an external current source, electroless deposition provides unique capability in laying down thin metallic layers on top of non-conductive substrates. This is particularly advantageous for the semiconductor industry, enabling the metallization of electrically isolated features and narrower, future generation, interconnects. The latter are currently electroplated with copper, using special additives mixtures that provide the essential bottom-up fill. While electroplating additives have been extensively studied and their interactions in the electroplating process have been well-characterized and quantitatively modeled1,2, no similar characterization is available for electroless plating. The absence of appropriate electroless plating additives and the non-availability of a model for their function impose a barrier on the technological adaptation of electroless plating to hi-tech precision applications. The research described here addresses these limitations by providing a quantitative model for the deposit thickness distribution in a general, additives-containing electroless process. The model quantifies the effects of additives in an electroless plating system, accounting for additives transport, adsorption, and incorporation within the deposit. The model builds on and extends a previously presented model for prediction of electroless plating rate in an additives-free system, accounting for reactant concentration, pH, and transport3. The model presented here includes an important extension: a prediction of the average plating rate based on the fractional surface coverage of the additive. This model is compared to experimental data obtained in an electroless plating system consisting of copper and glyoxylic acid, with mercaptopropane sulfonic acid (MPS) as the additive. An additive mass balance model, similar to that developed by Roha and Landau4, is used to estimate the concentration of the additive on the electrode surface. [1] Adsorption: NA=kACS,Add(Γsat-Γ) [2] Desorption: N-A=k-AΓ [3] Inclusion: NI=kIiΓ Here, N is the additive flux, k is its adsorption rate constant, and CS,Add is the additive concentration in solution adjacent to the surface. Γ and Γsat are the surface concentration of the adsorbed additive and its saturation concentration, respectively. i is the current density, or the equivalent current density associated with the deposit build-up. The subscripts A, -A, and I refer to the adsorption, desorption, and inclusion processes, respectively. Langmuir-type adsorption is assumed. While equations 1-3 are general, additional assumptions were made for the specific system studied here: (i) plating is completely inhibited at sites occupied by the additive, and (ii) the additive desorption flux is negligible as compared to adsorption and inclusion. At steady-state, the rate of adsorption and inclusion are therefore equal, leading to the following equation for the fractional additive surface coverage as a function of known or measurable parameters: [4]: (kACb,Add[1-θ])/(1+[δkAΓsat(1-θ)]/D)=kIiθ Here, Cb,Add is the bulk additive concentration, θ is its fractional surface coverage, δ is the additive diffusion boundary layer thickness, and D is its diffusion coefficient. As the plating does not occur on sites with adsorbed additive, we can relate the surface coverage to current density (and to an equivalent plating rate): [5]: i=iNo Add(1-θ) As shown in Fig. 1, the model matches the experimental data well. Of particular importance, the model captures the trend of increasing plating rate with rotation speed (and higher transport rates) to a maximum of around 100 rpm before dropping to almost zero at high rotation speeds. Additional analysis and further experimental details are provided in the presentation. Acknowledgements Atotech GMBH is acknowledged for funding this study and for providing helpful input. References R. Akolkar and U. Landau, "A time-dependent transport-kinetics model for additive interactions in copper interconnect metallization", J. Electrochem. Soc., 151, C702 (2004).R. Akolkar and U. Landau, “Mechanistic analysis of the "bottom-up" fill in copper interconnect metallization”, J. Electrochem. Soc., 156, D351 (2009).R. A. Zeszut Jr., and U. Landau, “Transport and reactants concentration effects in electroless deposition of copper”, Metallization of Flexible Electronics Symposium, 231st Electrochemical Society Meeting, (2017).D. Roha, U. Landau, “Mass transport of leveling agents in plating: steady-state model for blocking additives,” J. Electrochem. Soc., 137, 824 (1990).R. A. Zeszut Jr., “Effect of transport and additives on electroless copper plating”, PhD Dissertation, Department of Chemical and Biomolecular Engineering, Case Western Reserve University, (2017). Figure 1