1

Electrochemical Impedance Spectroscopy (EIS) Analysis of BTA

Removal by TMAH during Post Cu CMP Cleaning Process

R. Prasanna Venkatesha, Byoung Jun Cho

a, S. Ramanathan

b, and Jin-Goo Park

a*

a Department of Materials Engineering and Bio-nano Technology, Hanyang University,

Ansan 426 691, Korea

b Department of Chemical Engineering, Indian Institute of Technology Madras, Chennai

600 036, India

*corresponding author:

Tel.: 82-31-400-5226, Fax: 82-31-400-417-3701, Email address: jgpark@hanyang.ac.kr

2

Abstract

In the present study, the effectiveness of tetramethyl ammonium hydroxide (TMAH) in

removing benzotriazole (BTA) layer from Cu surface after chemical mechanical

polishing (CMP) is evaluated through electrochemical impedance spectroscopy studies.

Since the insitu electrochemical impedance spectroscopy (EIS) measurements could not

be applied to characterize the cleaning process, ex situ EIS measurement is used in the

present work. The impedance data is modeled by electrical equivalent circuit (EEC)

analysis and the polarization resistance values are calculated. The BTA removal for

various concentrations of TMAH is quantified from polarization resistance values.

Contact angle, scanning electron microscopy and x-ray photoelectron spectroscopy

measurements were also conducted to complement the studies. The results show TMAH

is effective in removing the BTA layer and that when part of the BTA is removed, the

residual BTA appear as nodules. Some of the native oxides and hydroxides on the Cu

surface are also removed by TMAH. When the slurry contains 0.1% (wt) BTA, a

minimum concentration of 0.5% (wt) TMAH and a minimum cleaning time of 60 s are

required for complete removal of BTA from the Cu surface.

Keywords: BTA, TMAH, Cu, Post CMP cleaning, EIS

3

1. Introduction

The adsorption and removal of mono layer of organic molecules on metal films occur

routinely during semiconductor wet processing. For example, adsorption of benzotriazole

(BTA) on Cu surface occurs during chemical mechanical polishing (CMP) of Cu surface

using a slurry containing BTA [1, 2, 3]. This makes the Cu surface more hydrophobic and

leads to post CMP defects such as watermarks during drying and particles onto the

surface [4, 5]. Thus, the Cu surface needs to be treated with suitable chemistry to remove

the adsorbed BTA layer. Various cleaning formulations have been studied for post Cu

CMP cleaning process [6-8]. Based on the Cu-BTA Pourbaix diagram, Murarkami et al.

[6] studied the performance of a cleaning solution containing oxalic acid, anionic

surfactants and water for removing BTA from Cu surface and showed that a cleaning

solution with pH 2.4 showed better performance than the conventional acidic solution

with pH 4.2. Otake et al. [7] showed that BTA could be effectively removed by various

cleaning solutions, but their compositions were not disclosed. Ein-Eli and Starosvetsky

[8] evaluated two commercial post CMP cleaning solutions using potentiodynamic

polarization experiments and found that etching of Cu in these solutions were associated

with deposition of corrosion products on the Cu surface. Here also, the exact

compositions of the commercial products are not disclosed. Tetramethyl ammonium

hydroxide (TMAH) is one of the candidates investigated for the removal of organics and

particles from the Cu surface [9-12]. In our previous studies, we proposed a cleaning

solution containing TMAH as cleaning agent, arginine as complexing agent, uric acid as

corrosion inhibitor and water as diluent [13].

4

Techniques such as X-ray photoelectron spectroscopy (XPS) [14] and Time –off –flight

secondary ion mass spectrometer [15] have been employed to evaluate the organic layer

removal by cleaning agents. Electrochemical impedance spectroscopy (EIS)

measurements could be another choice to evaluate the organic removal from metallic

surfaces. Lin et al [1] evaluated the kinetics of cleaning performance of citric acid

solution in removing organic residues such as BTA, 5-Aminitetroazole and 1-

phenyltetrazolethiol by measuring in situ open circuit potential value in a microfluidic

device. Electrochemical impedance spectroscopy is a powerful technique to investigate

the processes occurring at the solid-liquid interface [16]. However, in EIS technique,

the conventional experimental procedure do not allow us to evaluate the cleaning

performance of the system in situ. The reason is, the typical post Cu CMP cleaning

processing time is just one minute whereas the EIS measurement in suitable frequency

range takes relatively longer time and during that time, the interaction between the

surface and the solution cannot be prevented. This means, the solid/liquid interface could

not be same at the beginning and end of the measurement. EIS data is valid only if the

system returns to the original state at the end of measurement and thus, the in situ EIS

measurement methodology is not suitable for characterizing the effectiveness of post

CMP cleaning solution. Recently post etch residue (PER) removal from Cu surface by

choline chloride – malonic acid mixture was characterized and it was reported that the in

situ EIS measurements were not very sensitive since the film is very thin and the removal

occurs within minutes [17]. Hence, in this present work, ex situ EIS is applied to evaluate

the cleaning performance of the system. The electrical equivalent circuit (EEC) model is

employed to estimate the polarization resistance and these polarization resistance values

in turn are used to quantify the removal of BTA by TMAH. To our knowledge, this is the

first report on using EIS to quantify the effectiveness of post-CMP cleaning of Cu. We

5

found that the suggested methodology is sensitive in evaluating removal efficiency of the

cleaning agent, especially when the cleaning rate is very high i.e. cleaning time is less

than typical EIS measurement time. Besides post Cu cleaning process, the same

methodology could be adapted to quantify organic removal from the metallic surface in

other immersion cleaning process also such as PER removal from the metallic surface..

2. Experimental

All the electrochemical experiments were carried out using a potentiosatat (VersaSTAT,

Princeton applied research, USA) in a conventional three electrode system. The working

electrode is Cu and is treated with various solutions. The reference and counter electrode

are Ag/AgCl (Satd. KCl) and platinum mesh, respectively. All the electrolyte solutions

were prepared from analytical grade chemicals and deionized water. The ex situ EIS

measurements applied in this study were conducted as follows: the Cu surface is dipped

in 0.1 wt% BTA solutions for the typical Cu CMP processing time of 1 minute, and

rinsed in the DI water twice in separate beakers (the first rinse for 20 s and the second

rinse for 30 s). This BTA treated Cu is again dipped in the cleaning solution for 1 minute,

then again rinsed with water as mentioned above and finally dried in N2 stream. Then EIS

measurements were carried out at open circuit potential, with this treated Cu surface as

the working electrode, in a solution containing only the supporting electrolyte of 0.1 M

NaClO4 unlike in conventional EIS measurement where the cleaning agent is typically

the electrolyte solution. The spectrum was acquired in the frequency range of 10 kHz –

0.1 Hz at an AC potential of 10 mV rms. Potentiodynamic polarization plots were

acquired by scanning the working electrode in the potential range -0.5 to 0.5 V vs. OCP

at a scan rate of 2 mV s-1

. It must be noted that potentiodynamic polarization studies can

also been used to estimate the corrosion current and potential [18], from which the

6

efficiency of the cleaning process can be calculated. However, if the surface passivated in

the electrolyte under anodic conditions, then the potentiodynamic polarization studies

would not yield accurate estimates of the corrosion current. To complement the

electrochemical studies, the surface of the copper is characterized using contact angle

measurements, Field emission scanning electron microscopy (FE-SEM) and X-ray

photoelectron spectroscopy (XPS) measurements. The above measurements are carried

out when the surface is in dry state. The contact angle of Cu surface was measured using

a static contact angle analyzer (Phoenix 300, SEO, Korea). The surface of bare Cu and

Cu treated with cleaning solutions at various conditions were characterized by FE-SEM

(Mira3, TESCAN, USA) and XPS (Sigma probe, Thermo VG, UK).

3. Results and Discussion

3.1 EIS Measurements of Cu-BTA-TMAH System

Figure 1 shows the impedance spectra of Cu treated with BTA and TMAH. All

the spectra show the same pattern, i.e. a depressed semi circle. A comparison of the

spectrum of bare Cu (Fig. 1f) and Cu treated with BTA (Fig. 1a) shows that the

magnitude of the impedance at low frequencies is significantly higher for the BTA

treated Cu surface. This indicates that the BTA adsorbed on the Cu surface is effectively

increasing the polarization resistance. The impedance spectra of Cu surface treated with

BTA and subsequently with TMAH of different concentrations for 60 s are presented

from Fig. 1b to Fig. 1e. The low frequency impedance of the Cu-BTA surfaces treated

with TMAH decrease with increasing concentrations of TMAH. However, treatment with

0.1% (wt) and 0.25% (wt) TMAH does not remove the BTA completely from the Cu

surface as the low frequency impedance magnitude for these surfaces are still higher than

that of bare Cu. However, for the Cu surface treated with 0.5% (wt) and 1% (wt) TMAH

7

show that the low frequency impedance magnitude is actually less than that of bare Cu

surface. Thus at least 0.5% (wt) of TMAH is necessary to completely remove BTA from

the Cu surface.

The impedance spectrum of an electrochemical system which is causal, linear

and stable during EIS measurement will obey Kramers Kronig transform (KKT) (19). All

the impedance spectra presented here were validated with KKT. Electrical equivalent

circuit in Figure 2 accounts for the presence of film [20, 21] and is used to model the

impedance data. In the above circuit model, Rs represents the solution resistance. Q1 and

R1 are the capacitance and resistance associated with the passive film. Q2 and R2

represent the capacitance and resistance associated with electrical double layer and

charge transfer resistance at metal/electrolyte solution interface [21]. The polarization

resistance is calculated by adding R1 and R2 since the presence of BTA film affects both

the parameters [21, 22]. The simulated value of impedance data is shown in the Figure 3

along with the experimental values and the best fit parameters are shown in the Table I.

From the polarization resistance values, the surface coverage of the corrosion inhibitor

can be calculated [22-24]. Hence, the cleaning performance of the system could be

estimated using the following relation

P, Max P, x

P, Max P, 0

- BTA Removal (%) = 100

-

R R

R R

×

where

P, xR : Polarization resistance of BTA adsorbed Cu treated with cleaning solution

P,0R : Polarization resistance of untreated Cu

8

P, MaxR : Polarization resistance of Cu treated with BTA solution

The estimated BTA removal is 2%, 66%, 101% and 101% for TMAH

concentration of 0.1% (wt), 0.25% (wt), 0.5% (wt) and 1% (wt), respectively. This

clearly says that a minimum concentration of 0.5% (wt) TMAH is required for the

complete removal of BTA. The results are consistent with the contact angle and

potentiodynamic polarization measurements which are reported previously [13]. For

results corresponding to 0.5% (wt) and 1% (wt) TMAH, the BTA removal is slightly

higher than 100%. While this may be considered as falling within experimental error, a

more likely explanation is that some of oxides or hydroxides present on the bare Cu

surface may be etched by TMAH solution leading to lesser value of polarization

resistance. This point is elaborated in the XPS results discussion.

3.2 EIS Data for 0.5% TMAH as a Function of Time

To find the rate of removal of BTA at 0.5% (wt) TMAH, the EIS studies were

conducted as a function of dipping time (15, 30, 45 and 60 s) in TMAH solution and the

corresponding spectra are shown in Figure 4. It is clearly seen that the total impedance

decreases with dipping time. The same circuit shown in Figure 2 is also used to model the

experimental data and the EEC parameters obtained are given in the Table II. The BTA

removal (%) is estimated from the polarization resistance values as a function of dipping

time. The values obtained are 17%, 66% 78% and 101% for 15, 30, 45 and 60 s,

respectively. The removal rate is roughly linear with treatment time.

3.3 Complementary techniques

Potentiodynamic polarization plots were also obtained for 0.5% (wt) TMAH as a

function of dipping time and the results obtained are shown in Figure 5. The corrosion

9

potential (Ecorr) and the corrosion current density (Icorr) values, estimated from these plots

by Tafel extrapolation method, are given in the Table III. Since, in certain cases, the

kinks were observed in the anodic and cathodic region of the polarization curves, the

estimated Icorr values reported in Table III would not be very accurate. However, the

overall trend of Icorr vs time is along the expected lines. Both the decrease in Ecorr value

and the increase in Icorr value with the treatment time clearly correspond to the removal of

BTA from the Cu surface. Similar to impedance measurements, Rp values can be

estimated from the cathodic and anodic slopes of potentiodynamic polarization curves

using Stern - Geary equation[18] and the values obtained are reported in Table III. While

the trends of Rp vs cleaning time match qualitatively for EIS and potentiodynamic

polarization estimates, they do not match quantitatively. It is mainly due to the presence

of kinks in both anodic and cathodic region of the polarization curves at overpotentials

whick make the estimation of Rp from these data more difficult. Contact angle values

were measured for the Cu samples treated with 0.5% (wt) TMAH as a function of

treatment time and the results are shown in Figure 6. The contact angle of fresh copper

surface is 20º and of the BTA treated Cu surface is 62º. When the BTA treated Cu surface

is dipped in the cleaning solution for 15 s, there is no significant change in the contact

angle. The contact angle is decreased to 47º when the treatment time increased to 30 s

which shows that BTA is partially removed. The contact angle decreases to < 10º for 1

min treatment, suggesting that the surface is hydrophilic. EIS data shows a clear change

even for 15 s treatment whereas the contact angle data shows a significant change only at

30 s or longer duration. Thus the contact angle measurements are not as sensitive as EIS

measurements at least when the removal is partial.

10

Figure 7a and 7b show the SEM images of bare Cu and BTA treated Cu. There is

no significant difference between these two images. Figure 7c to 7f show the images of

Cu samples which are treated with BTA first and then with various concentrations of

TMAH solutions. When the concentration of TMAH is less than 0.5% (wt) (Figure 7c

and 7d), nodule-like features are observed on Cu. It is known that BTA present on copper

can appear as needle like structure in SEM image [2]. Hence, they are likely to be islands

of BTA, and indicate that the BTA is not completely removed from the Cu surface. At

0.5% (wt) and 1% (wt) TMAH (Figure 7e and 7f ), the surface is very similar to that of

bare Cu surface which confirms that BTA film is completely removed from the Cu

surface. With further increase in TMAH concentration, the surface roughness also

increases. Figures 8(a-d) show the SEM images of BTA treated Cu which is treated in

0.5% (wt) TMAH for various durations. When the treatment time is less than 60 s,

nodule-like features are seen here also and at the end of 60 s, the surface is free of such

features. This also confirms that at least 60 s dipping time should be provided for the

complete removal of BTA. Thus the major conclusions of the EIS results viz. a minimum

of 0.5% (wt) TMAH and a minimum of 60 s treatment duration is necessary to remove

BTA from Cu surface, are supported by the SEM results.

XPS analysis of bare copper and copper treated with three different solutions (1. only

BTA, 2. only TMAH, 3. BTA followed by 0.5% TMAH) were carried out to determine

the status of the surface and also to identify the organic residues, if present. The spectrum

of Cu, O and N are shown in Figures 9 (a-c). The Cu 2p spectrum exhibits two main

peaks, Cu 2p3/2 and Cu 2p1/2 at the binding energies of 931.8 eV and 951.5 eV in addition

to the two satellite peaks at 943.6 and 962 eV. These characteristics reveal that CuO is

present on the copper surface. The peak at 531.1 eV in oxygen spectrum also confirms

11

the same. However for the bare and BTA treated Cu samples, when they are treated with

TMAH solution, the satellite peaks in the Cu 2p spectrum disappear. Besides, the

intensity of the CuO peak in the O 1s spectrum decreases significantly and instead, the

peak at 529.9 eV, corresponding to Cu2O, appears. In addition, the shoulder peak at 934.8

eV, which is attributed to copper hydroxides, disappears for the TMAH treated Cu

surface. These results suggests that when the Cu surface (whether it is bare Cu surface or

Cu treated with BTA) is cleaned in TMAH solution, a native oxide of Cu (+1) grows on

the surface during exposure to ambient. If some of the copper oxides and hydroxides on

the Cu surface are removed during TMAH treatment, the impedance of the treated

surface would be less than that of bare Cu and thus the XPS results corroborate EIS

results. The BTA treated Cu surface exhibits a nitrogen peak at 399.6 eV, but when the

surface is treated with TMAH, the nitrogen peak at 399.6 eV disappears. The small

intensity peak appears at 402.4 eV corresponding to nitrogen with alkyl group appears

when the Cu is treated with TMAH solutions. Thus, residues of TMAH may be present

on the surface, although they do not result in a hydrophobic surface and hence are not

expected to cause damage during further processing.

4. Conclusions

In this present study, ex situ EIS measurement is employed to study the removal of BTA

from Cu surface using TMAH. EIS data were modeled by EEC and the polarization

resistance of the Cu surfaces treated with various solutions is calculated from the model

fit. The efficiency of BTA removal is estimated from these polarization resistance values

and the studies show that a minimum concentration of 0.5% (wt) TMAH and a minimum

duration of 60 s are required for the removal of 0.1% (wt) BTA. The results indicate that

12

ex situ EIS technique is sensitive to the presence of BTA and also to the presence of

oxides and hydroxides on the Cu surface. The results obtained from contact angle

measurements are less sensitive when the BTA layer is only partially removed, but the

overall trend corroborates with other results. SEM images show that when BTA is

partially removed by TMAH, the residues remain as nodule-like structures on the Cu

surface. XPS measurements confirm that TMAH removes BTA as well as some of the

native oxide and hydroxide on the Cu surface. In summary, ex situ EIS experimental

procedure can be used as a sensitive method to characterize Cu post CMP cleaning

process.

Acknowledgements

We would like to thank Prof. D.D. Macdonald (macdonald@matse.psu.edu) for the

Kramers Kronig transform software. This research was supported under the framework of

International Cooperation Programme (# 2011-0027711) and Basic Science Research

Program (# R11-2008-044-02000-0) through the National Research Foundation of Korea

(NRF) funded by the Ministry of Education, Science and Technology.

References

1. Lin JY, West AC, Wan CC, Wang YY (2008) Electrochem Commun 10:677-680

2. Harada K, Ito A, Kawase Y, Suzuki T, Hara M, Sakae R, Kimura C, Aoki H

(2011) Jpn J Appl Phys 50:05EC06(1-4)

3. Peters DW (2007) Res Soc Symp Proc 991:215-225

4. Yeh CF, Hsiao CW, Lee WS (2003) Appl Surf Sci 216:46-53

13

5. Hong YK, Eom DH, Lee SH, Kim TG, Park JG, Busnaina AA (2004) J

Electrochem Soc 151(11):G756-G761

6. Murakami Y , Ishikawa N (2009) International Conference on Planarization/CMP

Technology (ICPT), Fukuoka, Japan, p. 495-499

7. Otake A, Kuroda A, Matsumoto T, Daviot J , Shang C (2009) International

Conference on Planarization/CMP Technology (ICPT), Fukuoka, Japan, p. 137-

142

8. Ein-Eli Y, Starosvetsky D (2007) Electrochim Acta 52:1825-1838

9. Huang WH, Raghavan S, Fang Y, Zhang L (1999) Mat Res Soc Syp Proc

566:161-166

10. Mishra A, Fischer ML, US patent application publication no. US 2005/0181961

11. Naghsineh S, Barnes J, Oldak EB, US patent no. 6,723,691 B2.

12. Chiou WC, Chen YH, Lee SN, Jeng SM, Jang SM, Liang MS (2004) IEEE

International Interconnect Technology Conf Proc, June 7, San Francisco,

California, USA, p.127-129

13. R. Prasanna Venaktesh, Tae-Young Kwon, Nagendra Prasad Y, Ramanathan S,

Park J-G (2011) Microelectronic Engineering (Accepted).

14. Kodera M, Nishioka Y, Shima S, Fukunaga A, Tsujimara M (2005) Jpn J Appl

Phys 44:8396

15. Nishi A, Sado M, Miki T, Fukui Y (2003) Appl Surf Sci 203-204:470

16. Orazem M, Tribollet B (2008) Electrochemical Impedance Spectroscopy. John

Wiley and Sons, New Jersey

17. Taubert J, Keswani M and Raghavan S , Microelectron. Engg

doi:10.1016/j.mee.2011.11.014

14

18. Bard AJ, Faulkner LR (2001) Electrochemical methods – Fundamentals and

Applications. 2nd edi., John Wiley & Sons, Canada

19. Macdonald JR (2005) In: Barsoukov E, Macdonald JR (eds.) Impedance

Spectroscopy, 2nd edn., Wiley, New Jersey

20. Tsai T-H, Wu Y-F, Yen S-C (2003) Appl Surf Sci 214:120-135

21. Fenelon AM, Breslin CB (2001) J Appl Electrochem 31:509-516

22. Kosec T, Merl DK, Milošev I (2008) Corros Sci 50:1987-1997

23. Noyel Victoria S, Ramanathan S (2011) J Solid State Electrochem 15:837-844

24. Zhang D, Cai Q, He X, Gao L, Zhou G (2008) Mat Chem Phys 112:353-358

15

TABLE CAPTIONS

Table I. EEC Parameters obtained for Cu treated with various solutions which is

presented in the Figure 2

Table II. EEC Parameters retrieved from the impedance data shown in the Figure 4

Table III. The values of Ecorr, Icorr, and Rp values estimated from polarization curves

16

FIGURE CAPTIONS

Figure 1. The impedance spectrum of Cu in a solution containing only supporting

electrolyte of 0.1 M NaClO4. Prior to the electrochemical experiments the Cu was treated

first with BTA and second with various concentrations of TMAH solution. The

concentration of TMAH in the second treatment step was A) 0% (wt), B) 0.1% (wt), C)

0.25% (wt), D) 0.5% (wt), E) 1% (wt) and F) untreated reference Cu. The inset figure

shows the expanded view at lower impedance for clarity.

Figure 2. Electrical equivalent circuit model used to fit the impedance data acquired for

Cu surfaces treated with various solutions

Figure 3. Simulated impedance data superimposed with experimental raw data. The

experimental conditions are presented in the Figure 1. The inset figure shows the

expanded view at lower impedance for clarity. The line represents the model fit and

points represent the experimental data.

Figure 4. (a) The impedance spectrum of Cu in a solution containing only supporting

electrolyte of 0.1 M NaClO4. Prior to the electrochemical experiments the Cu was first

treated with BTA and then with 0.5% (wt) TMAH solution for various times. The inset

figure shows the expanded view at lower impedance for clarity. The line represents the

model fit and points represent the experimental data.

Figure 5. (a) Potentiodynamic polarization curves of Cu in a solution containing only

supporting electrolyte of 0.1 M NaClO4. Prior to the electrochemical experiments the Cu

was first treated with BTA and then with 0.5% (wt) TMAH solution for various times; (a)

0 s, (b) 15 s, (c) 30 s, (d) 45 s and (e) 60 s

Figure 6. Contact angle measurements of Cu dipped in 0.1% (wt) BTA followed by

dipping in 0.5% (wt) TMAH as a function of time.

Figure 7. SEM images of (a) bare Cu, (b) BTA treated Cu, (c) – (f) Cu treated with BTA

first and then with 0.1% (wt), 0.25% (wt), 0.5% (wt) and 1% (wt) TMAH, respectively

17

Figure 8. SEM images of Cu treated with BTA first and then with 0.5% (wt) TMAH for

various dipping times; (a) 15 s, (b) 30 s, (c) 45 s and (d) 60 s

Figure 9. The XPS spectra of (a) Cu 2p (b) O 1s and (c) N 1s for the samples; (1) bare

Cu, (2) BTA treated Cu, (3) TMAH treated Cu and (4) Cu treated with BTA first and

then with 0.5% (wt) TMAH.

18

TABLE I.

Parameters Untreated

Cu

BTA

treated

Cu

Cu treated with BTA and subsequently with

various concentration (wt%) of TMAH

0.1 0.25 0.5 1

Rs, ohm cm2 105 108 108 105 107 106

Y01,S 1ns cm-2

3.4 10-6

2.4 10-6

3.0 10-6

4.1 10-6

1.6 10-6

7.8 10-6

n1 0.8 0.95 0.98 0.96 0.84 0.91

R1, ohm cm2 5000 7151 8022 9165 295 947

Y02,S 2ns cm-2

2.8 10-5

1.5 10-6

2.9 10-6

8.5 10-6

1.1 10-4

7.5 10-5

n2 0.8 0.57 0.59 0.5 0.62 0.54

R2, ohm cm2 2.7 10

4 3.5 10

5 3.5 10

5 1.4 10

5 2.7 10

4 2.5 10

4

19

TABLE II. EEC Parameters retrieved from the impedance data shown in the Figure 4

Parameters Untreated

Cu

BTA

treated

Cu

Cu treated with BTA and subsequently with

0.5 wt% TMAH for various times (s)

15 30 45 60

Rs, ohm cm2 105 108 107 105 110 107

Y01,S 1ns cm-2

3.4 10-6

2.4 10-6

3.6 10-6

4.0 10-6

5.6 10-6

1.6 10-5

n1 0.8 0.95 0.97 0.96 0.93 0.84

R1, ohm cm2 5000 7151 1.8 10

4 1.1 10

4 2802 294.5

Y02, S 2ns cm-2

2.81 10-5

1.5 10-6

3.8 10-6

7.9 10-6

3.3 10-5

1.1 10-4

n2 0.8 0.57 0.53 0.53 0.51 0.62

R2, ohm cm2 2.7 10

4 3.5 10

5 2.9 10

5 1.3 10

5 1.03 10

5 2.69 10

4

20

Table III. The values of Ecorr and Icorr

Dipping time of BTA deposited

Cu in 0.5% (wt) TMAH

Ecorr (mV) Icorr (µA) Rp(ohm-

cm2)

0 378 0.2 2.15 105

15 341 0.6 8.68 104

30 329 0.9 4.63 104

45 313 2 1.71 104

60 281 3.2 1.17 104

21

Figure 1

22

Rsol

R1

Q1

R2

Q2

Figure 2

23

Figure 3

24

Figure 4

25

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

1.00E-09 1.00E-07 1.00E-05 1.00E-03 1.00E-01

E/

V v

s. A

g/A

gC

l (s

atd

.)

log I (I, A cm-2)

A

BC

D

E

10-9 10-7 10-5 10-3 10-1

Figure 5

26

0

10

20

30

40

50

60

70

0 15 30 45 60

Co

nta

ct a

ng

le/

de

gre

e

Time / s

Figure 6

27

2 μm

2 μm

(a) (b)

2 μm

2 μm

(c) (d)

2 μm

2 μm

(e) (f)

Figure 7

28

2 μm

2 μm

(a) (b)

2 μm

(c) (d)

Figure 8

29

910930950970

Inte

nsi

ty (

a.u

.)

Binding energy/ eV

(1)

(2)

(3)

(4)

Figure 9 (a)

30

520525530535540545

Inte

nsi

ty (

a.u

.)

Binding energy/ eV

(1)

(2)

(3)

(4)

Figure 9 (b)

31

385395405415

Inte

nsi

ty (

a.u

.)

Binding energy/ eV

(1)

(2)

(3)

(4)

Figure 9 (c)