Competitive anion/anion interactions on copper surfaces relevant for Damascene electroplating

Hai, Nguyen Thi Minh; Huynh, Truc T.; Fluegel, Alexander; Arnold, Marco; Mayer, Dieter; Reckien, Werner; Bredow, Thomas; Broekmann, Peter

doi:10.1016/j.electacta.2012.03.054

Cited by 64 publications

(149 citation statements)

References 43 publications

Supporting

Mentioning

135

Contrasting

Order By: Relevance

“…Feng et al [45] proposed that Cl -acts as a strong anchor and secures the PEG-Cu + complex to the cathode surface. Hai et al [46] further suggested that the PEG branches could interlink creating a large network of interlinked suppressor complex that effectively covers a huge area of the substrate surface.…”

Section: Effect Of Single Additivementioning

confidence: 99%

Electrochemical effect of Copper Gleam additives during copper electrodeposition

Pena

Roy

2017

Transactions of the IMF

View full text Add to dashboard Cite

show abstract

Section: Effect Of Single Additivementioning

confidence: 99%

Electrochemical effect of Copper Gleam additives during copper electrodeposition

Pena

Roy

2017

Transactions of the IMF

View full text Add to dashboard Cite

show abstract

“…Though MPS is known as a strong accelerator, [20][21][22][23] MPS itself is unable to induce superfilling 22 since superfilling is a result of complex behaviors of the accelerators, including not only the acceleration of the deposition kinetics but also its adsorption on the topographic surface and interaction with the suppressor. To acquire a superfilling capability, either SPS or aged MPS (dimerized into SPS) as described by Eq.…”

Section: D35mentioning

confidence: 99%

“…Acceleration by SPS in Cu superfilling has been explained by the competitive adsorption theory [17][18][19][20][21] or by the catalytic action caused by the reduction of cupric ions. 22,23 In the competitive adsorption theory, acceleration is regarded as a recovery of the deposition rate previously suppressed by the polyethylene glycol (PEG)-Cl inhibition layer, and recovery occurs through the displacement of the PEG-Cl layer with SPS.…”

mentioning

confidence: 99%

See 1 more Smart Citation

High Accuracy Concentration Analysis of Accelerator Components in Acidic Cu Superfilling Bath

Choe

Kim

et al. 2015

J. Electrochem. Soc.

View full text Add to dashboard Cite

We have devised a modified cyclic voltammetry stripping (CVS) method to measure the concentrations of bis-(sulfopropyl) disulfide (SPS) and 3-mercapto-1-propane sulfonate (MPS) in Cu plating solutions. Though MPS, a breakdown product of SPS, enhances the Cu deposition rate on flat electrodes, it is not a superfilling-capable accelerator for the damascene structure, unlike SPS. Therefore, accurate measurement of SPS in damascene Cu plating baths is important. However, enhancement of the Cu deposition rate by MPS interferes with the electrochemical signal of SPS, leading to a significant error when using the modified linear approximation technique (MLAT)-CVS analysis method. To evaluate their concentrations individually, a two-step CVS analysis was performed in which the total accelerator concentration ([SPS] + 1/2[MPS]) and conversion ratio were separately determined. All MPS species in the bath were oxidized to SPS by controlling the plating solution pH. Subsequent MLAT-CVS analysis successfully revealed the total accelerator concentration in the Cu plating solution. Individual SPS and MPS concentrations were thereby calculated using the conversion ratio evaluated from the difference in their relative accelerating abilities. This modified method enabled determination of the SPS concentration with <10% error, suggesting a reliable and high accuracy tool to predict pattern filling capabilities of plating solutions. Owing to the merits of high throughput, low process cost, and good film quality, Cu electroplating has a wide application, from traditional Cu foil production to state-of-the-art semiconductor metallization processes. [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16] The Cu plating solution typically contains small amounts of organic additives, 1-16 which enhance the uniformity, 13 brightness, 14 and mechanical properties of the Cu film. 15 In addition, for particular application to damascene Cu plating, the additives enable bottom-up filling at trenches or vias by controlling the relative deposition rates of Cu at the top and bottom of the features.1-12 For this application, the organic additives are grouped as the accelerator, suppressor, and leveler based on their electrochemical behaviors and their roles in pattern filling.One of the most widely used accelerators is bis-(sulfopropyl) disulfide (SPS), a dimer of 3-mercapto-1-propane sulfonate (MPS). Acceleration by SPS in Cu superfilling has been explained by the competitive adsorption theory [17][18][19][20][21] or by the catalytic action caused by the reduction of cupric ions. 22,23 In the competitive adsorption theory, acceleration is regarded as a recovery of the deposition rate previously suppressed by the polyethylene glycol (PEG)-Cl inhibition layer, and recovery occurs through the displacement of the PEG-Cl layer with SPS.17-19 A recent study revealed that the acceleration is a result of the dissociative adsorption of SPS with displacement of the well-ordered Cl − layer. 20,21 MPS, a dissociated product of SPS, reconverts to SPS with re...

show abstract

“…[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. 13 Akolkar and Landau reported addition experiments in which they injected PEG and SPS into the electrolysis cell during plating.…”

mentioning

confidence: 99%

On the Displacement of Adsorbed Polyethylene Glycol by 3-Mercapto-1-Propanesulfonate during Copper Electrodeposition

Kreider

Barkey

Wong

2014

J. Electrochem. Soc.

View full text Add to dashboard Cite

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

show abstract

Competitive anion/anion interactions on copper surfaces relevant for Damascene electroplating

Cited by 64 publications

References 43 publications

Electrochemical effect of Copper Gleam additives during copper electrodeposition

Electrochemical effect of Copper Gleam additives during copper electrodeposition

High Accuracy Concentration Analysis of Accelerator Components in Acidic Cu Superfilling Bath

On the Displacement of Adsorbed Polyethylene Glycol by 3-Mercapto-1-Propanesulfonate during Copper Electrodeposition

Contact Info

Product

Resources

About