Solid State Phenomena Submitted: 2020-05-01

ISSN: 1662-9779, Vol. 314, pp 270-274 Revised: 2020-08-14

© 2021 Trans Tech Publications Ltd, Switzerland Accepted: 2020-08-14
Online: 2021-02-09

Effect of Dissolved Oxygen on Removal of Benzotriazole from Co

During a Post-Co CMP Cleaning
Heon-Yul Ryu1,a, Palwasha Jalalzai1,b, Nagendra Prasad Yerriboina1,c,
Tae-Gon Kim2,d, Satomi Hamada3,e, and Jin-Goo Park1,f *
1
Department of Materials Science and Chemical Engineering, Hanyang University, Ansan 15588,
Korea
2
Department of Smart Convergence Engineering, Hanyang University, Ansan 15588, Korea
3
EBARA Corporation, Fujisawa, Kanagawa 251-8502, Japan
a
[email protected], [email protected], [email protected], [email protected],
e
[email protected], f*[email protected]

Keywords: cobalt, benzotriazole, adsorption, dissolved oxygen, post Co-CMP cleaning

Abstract. A systematic study of the adsorption of benzotriazole on Co surface and its removal in
aqueous solutions was carried out for post-CMP cleaning application. Static etch rate (SER)
measurements and electrochemical impedance spectroscopy (EIS) were employed. The experimental
results show that BTA adsorbed well on Co surface when it exposed to BTA solution at neutral and
alkaline pH. BTA did not adsorb due to active Co dissolution at acidic pH. The effect of dissolved
oxygen (DO) concentration in de-ionized (DI) water on the removal of Co-BTA complex layer was
investigated. At the DI water rinse process after BTA treatment, Co-BTA layer was maintained on
Co surface when DO concentration of DI water was low. Interestingly, BTA was removed by DI
water with high DO concentration.

Introduction
Cobalt (Co) has replaced the conventional materials used for barrier, electrodes, and interconnects
of integrated circuit (IC) device because of its excellent conductivity at sub 10 nm technology node
[1]. However, corrosion is one of the major issues for Co when it is applied to chemical mechanical
planarization (CMP) and post-CMP cleaning process [2]. Corrosion inhibitors are used not only for
Co CMP slurries but also for post-CMP cleaning solutions to prevent corrosion induced defects since
Co is more susceptible to corrosion than Cu. They adsorb on Co surface and form a Co-inhibitor
complex layer [3] similar to Cu-inhibitor complex layer [4, 5]. However, the presence of the inhibitor
after CMP may act as residual organic defects [4, 5]. It is desirable to achieve an inhibitor-free surface
after post-CMP cleaning by using a proper inhibitor which adsorbs well during post-CMP cleaning
and be removed during de-ionized (DI) water rinse step. It is reported that the corrosion of Cu occurs
during the DI water rinse step as functions of dissolved oxygen (DO) concentration and oxidation-
reduction potential (ORP) of DI water [6]. Thus, the condition of DI water for the rinse step also
needs to be considered for corrosion prevention and residual inhibitor removal. In this study,
corrosion inhibition of Co surface by benzotriazole (BTA) and its removal was investigated. The
effect of DO of DI water on the removal of Co-BTA complex was also studied.

Experimental Procedures
300-mm Co wafer (thickness 200 nm, physical vapor deposition) was cut into coupons (2 x 2 cm2)
for the experiments. The SER of Co was evaluated by dipping the Co coupons in solutions for 10 min
followed by DI water rinse and N2 dry. The solutions were prepared at three different pHs (3, 7, and
11) with and without 10 mM BTA (C6H5N3, Sigma Aldrich, USA). The pH of solutions was adjusted
with HNO3 and KOH. The static etch rate (SER) was calculated by measuring the sheet resistance of
Co film using a four-point probe (AIT, CMT-SR5000, Korea) before and after the treatment in
solutions.

All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans
Tech Publications Ltd, www.scientific.net. (#548071955-09/12/20,09:27:40)
 Solid State Phenomena Vol. 314 271

The prepared Co coupons were carried out for electrochemical impedance spectroscopy (EIS)
measurement to investigate the formation of Co-BTA complex on Co surface. For EIS measurements,
a 1 cm2 circular area of Co was exposed to the electrolyte solutions, and all the experiments were
performed using a potentiostat (VersaSTAT3, AMETEK, USA) in a conventional three-electrode
system. The Co coupon was used as a working electrode, whereas Ag/AgCl (saturated KCl) and
platinum mesh were used as the reference and counter electrodes, respectively. Since EIS is a non-
destructive technique, sequential measurements were performed with the Co coupon by treating the
different electrolyte solutions. DI water and 10 mM BTA solutions at pH 3, 7, and 11 were used as
electrolyte solutions. For all solutions, the pH was adjusted with HNO3 and KOH. Commercial
software (ZSimpWIN, EChem Software, USA) was used to obtain the electrical equivalent circuit
(EEC) parameters.
The concentration of DO in DI water was controlled by using Ar aeration and measured by a DO
probe (Thermo Fisher Scientific, USA). The oxidation-reduction potential (ORP) which is related to
DO concentration of DI water was also measured [7].

Results and Discussions

To understand the corrosion inhibition of Co by BTA, SER was measured at pH 3, 7, and 11 and
shown in Figure 1. The SER was higher at pH 3 and lower at pH 7 and 11 in the absence of BTA.
When a 10 mM BTA solution was used, SER was not much affected at pH 3 but it decreased
especially at pH 7. It shows that BTA is effective to inhibit the dissolution or corrosion at the different
pH values in the following order: pH 7 > pH 11 > pH 3.

Figure 1: Static etch rate of Co in solutions with 0 and 10 mM BTA at different pHs.

The EIS experiment was performed to understand the stability of Co-BTA complex at different
pHs. The EIS data were analyzed by EEC modeling as shown in Figure 2, which includes the porous
film on the metal surface. In this circuit model, Rsol is electrolyte solution resistance, and CPEc and
Rc are the capacitance and resistance related to the passive film such as oxide or Co-BTA complex.
CPEdl indicates the capacitance related to an electrical double layer at the interface of the metal
surface and electrolyte solution and Rct indicates the charge transfer resistance at the interface of the
metal surface and electrolyte solution. The polarization resistance is calculated by summing up Rc
and Rct. Through the polarization resistance, the presence of Co-BTA complex on Co surface could
be analyzed quantitatively.
The EIS results are shown in Figure 3 and Fig. 4 shows the polarization resistance values of Co
at different conditions calculated from Fig. 3. In the absence of BTA, the impedance curves and
polarization resistances of Co surface at pH 11 were higher than pH 7 due to different Co oxide
formation [8]. Co surface showed the lowest polarization resistance at pH 3 due to Co dissolution. In
the presence of BTA, Co surface showed the largest impedance curve and the highest polarization
resistance of 73,290 ohm-cm2 at pH 7 which means that a thicker Co-BTA complex was formed at
272 Ultra Clean Processing of Semiconductor Surfaces XV

pH 7 than at 11. At pH 3, the polarization resistance did not show any change in solutions with and
without BTA which indicates no adsorption of BTA on Co surface. Therefore, neutral and alkaline
pH is favorable for BTA containing post-CMP cleaning solution to prevent Co corrosion during the
cleaning process.

Figure 2: EEC model used to fit the impedance data acquired from Co surfaces treated with various
solutions. (a) a simple circuit model for the metal surface without any passive layer and (b) circuit
model for the metal surface with a porous passivation layer.

Figure 3: Measured and EEC modeled impedance spectra of Co surfaces from EIS measurements
with different electrolytes; (a) pH adjusted DI water and (b) pH adjusted 10 mM BTA.

Figure 4: Polarization resistance of Co surfaces in solutions with 0 and 10 mM BTA at different pHs.
To evaluate the Co-BTA complex layer on Co surface after post-CMP cleaning, sequential EIS
was used with two electrolytes: 10 mM BTA and DI water at pH 7. The DO concentration in DI water
was controlled by Ar aeration before EIS measurement to investigate the effect of DO on residual
BTA on Co surface. Figure 5 shows the impedance curves and polarization resistances of Co surface
 Solid State Phenomena Vol. 314 273

at different electrolytes. DI water originally showed 10 ppm of DO and 276 mV of ORP before Ar
aeration. After 3 hours of Ar aeration, DO and ORP of DI water decreased to 1 ppm and 183 mV,
respectively. The polarization resistance at step 2 was similar to step 1 when DI water with 1 ppm
DO was used, which indicates no removal of BTA from Co surface. However, when DI water with
10 ppm DO was used for EIS in step 2, the polarization resistance was decreased to lower than 10
kohm-cm2 which is similar to the metallic Co surface. It indicates that Co-BTA complex was removed
from Co surface by DI water rinsing.

Figure 5: (a) Impedance spectra and (b) polarization resistance of Co surfaces obtained from
sequential EIS. Different DO concentrations in DI water was applied for sequential EIS step 2.

As a result, the mechanism of BTA removal could be hypothesized as shown in Figure 6. Co

surface is protected from corrosion by forming Co-BTA complex layer when it is exposed to a post-
CMP cleaning solution containing BTA at pH 7 (Fig. 6 a). At DI water rinse step with high DO and
ORP, Co-BTA complex on Co surface is not stable and Co is oxidized by the high concentration of
oxygen in DI water (Fig. 6 b). This oxidation might be a reason for BTA removal. On the other hand,
when DO and ORP of DI water are low, Co-BTA complex layer is stable and remained on Co surface.

Figure 6: BTA adsorption and removal with different electrolytes (pH 7). (a) during post-CMP
cleaning step (b) DI water rinse step with high DO (c) DI water rinse step with low DO.

Summary
In this study, BTA adsorption on Co surface at different pH conditions and its removal with two
different concentrations of DO water was investigated by SER and EIS measurements. BTA was not
effective as a corrosion inhibitor in acidic solutions, while it was effective at neutral and alkaline pH.
EIS measurement confirms no BTA adsorption at pH 3 and the formation of the Co-BTA complex
layer on Co surface at pH 7 and 11. A low DO in DI water did not remove BTA, but high DO DI
274 Ultra Clean Processing of Semiconductor Surfaces XV

water effectively removed BTA from Co surface. As a result, BTA can be suggested as a suitable
corrosion inhibitor for post-Co CMP cleaning solution because it can protect the Co surface from
corrosion during the cleaning step and it can be removed easily by DI water with proper DO
concentration.

References
[1] J.-H. Han, Chemical Mechanical Polishing Innovation as a Key Technology for sub-10nm Logic
Device, China Semiconductor Technology International Conference (CSTIC) 2016, Shanghai, China
(2016).
[2] H. Nishizawa, H. Nojo and A. Isobe, Fundamental Study of Chemical-Mechanical Polishing
Slurry of Cobalt Barrier Metal for the Next-Generation Interconnect Process, Jpn. J. Appl. Phys. 49
(2010) 05FC03-1-05FC03-2.
[3] D. Gallant, M. Pezolet and S. Simard, Inhibition of cobalt active dissolution by benzotriazole in
slightly alkaline bicarbonate aqueous media, Electrochim. Acta 52 (2007) 4927-4941.
[4] B. J. Cho, S. Shima, S. Hamada and J. G. Park, Investigation of Cu-BTA complex formation
during Cu chemical mechanical planarization process, Appl. Surf. Sci., 384 (2016) 505-510.
[5] B. J. Cho, J. Y. Kim, S. Hamada, S. Shima and J. G. Park, Effect of pH and chemical mechanical
planarization process conditions on the copper-benzotriazole complex formation, Jpn. J. Appl. Phys.
55 (2016) 06JB01-1-06JB01-5.
[6] M. Imai, Y. Yamashita, T. Futatsuki, M. Shiohara, S. Kondo and S. Saito, Effect of Dissolved
Oxygen on Cu Corrosion in Single Wafer Cleaning Process, Jpn. J. Appl. Phys. 48 (2009) 04C023-
1-04C023-4.
[7] E. Lie and T. Welander, Influence of dissolved oxygen and oxidation–reduction potential on the
denitrification rate of activated sludge, Wat. Sci. Tech., 30 (1994) 91-100.
[8] W. A. Badawy, F. M. Al-Kharafi and J. R. Al-Ajmi, Electrochemical behaviour of cobalt in
aqueous solutions of different pH, J. Appl. Electrochem. 30 (2000) 693-704.