Displacement of the suppressor polyethylene glycol (PEG) by the accelerant 3-mercapto-1-propanesulfonate (MPS) during copper electrodeposition was studied by a chronoamperometric addition technique. Displacement of PEG occurs through a process of MPS domain nucleation, mass-transfer limited domain expansion and an asymptotic approach to equilibrium between MPS in solution and that on the surface. An isotherm is proposed to account for the equilibrium coverage over a wide range of MPS concentrations, and the nearest neighbor separation between adsorbed MPS molecules during deposition is estimated. © The Author A typical additive ensemble for superfilling in copper electrodeposition comprises the suppressor polyethylene glycol (PEG) and the accelerant 3-mercapto-1-propanesulfonate (MPS). The complete accelerant is a complex of MPS with cuprous chloride, and it operates by exclusion or displacement of PEG from the electrode surface. [1][2][3][4][5][6][7][8][9][10][11][12] This type of additive interaction has been classified by Hai et al. as antagonistic because the suppressor and accelerant compete for space on the surface and do not otherwise interact.13 They have developed a detailed mechanism of accelerant adsorption that involves penetration of the accelerant though an adsorbed chloride layer through potential induced defects and the organization of a close-packed accelerant adlayer on the surface. 13Akolkar and Landau reported addition experiments in which they injected PEG and SPS into the electrolysis cell during plating. 14,15 They found that PEG adsorption was rapid and limited by masstransfer whereas SPS adsorption was kinetically limited. The latter result is also supported by Chiu et al. 16 In this study, we are motivated by the hypotheses that displacement of PEG by MPS occurs through nucleation of MPS domains followed by mass-transfer limited growth of the domains and finally their merger into a complete ad-layer. Based on this mechanism, we expect the current to follow the formWhere θ is the fractional accelerant coverage and i a and i s are the current densities on areas covered by accelerant and suppressor respectively. Equation 1 can be rearranged to give an explicit expression for θ.At long times, the current density reaches a steady final value i f that corresponds to equilibrium between the MPS in solution and that on the surface, so thatIn the present experiments, the working electrode is first plated in a solution containing only suppressor which covers the surface. MPS is then added in various concentrations. This procedure avoids the kinetically limited step of SPS reduction to MPS and permits us to evaluate the amount of accelerant that reaches the surface. In addition, by varying the MPS concentration over a very wide range, we obtain information on the equilibrium between MPS in solution with that on * Electrochemical Society Active Member. z