versión On-line ISSN 0717-9707
J. Chil. Chem. Soc. v.54 n.4 Concepción dic. 2009
J. Chil. Chem. Soc., 54, Nº 4 (2009), págs. 439-444.
EFFECT OF THIOUREA CONCENTRATION ON THE ELECTROCHEMICAL BEHAVIOR OF GOLD AND COPPER ELECTRODES IN PRESENCE AND ABSENCE OF Cu(II) IONS.
H. GÓMEZ, H. LIZAMA*, C. SUÁREZ, A. VALENZUELA
Instituto de Química, Facultad de Ciencias, Pontifcia Universidad Católica de Valparaíso, Avenida Brasil 2950, Valparaíso, Chile. e-mail address: email@example.com
The effect of thiourea (TU) concentration on copper electrodeposition was studied employing cyclic voltammetry and electrochemical crystal quartz microbalance techniques. It was found that the polarization effects are related to the type of species generated in the electrode interface. Due to formation of formamidine disulfde (FDS) at concentrations greater than 10-4 mol dm-3, the electrode surface is blocked, and copper electrodeposition shifts towards more negative potential values (polarizing effect). Lower concentrations favors the formation of Cu(I) -TU contributing to the de-polarization of copper electrodeposition reaction.
Key words: thiourea; electrodeposition; copper; crystal quartz microbalance; formamidine disulfde.
Combination of glue, thiourea, and chloride are used as inhibitors in nearly all copper electrorefning tankhouses 1-6. Although numerous investigations having been conducted, the mechanism of action of these inhibitors system has not been fully clarifed. It is generally accepted that the additives produce an additional local overpotential at the more active cathode sites, favoring the formation of a low roughness homogenous surface7. In the particular case of thiourea (TU), its role in copper electrodeposition is mainly related to the formation and distribution of grain size8 but the associated chemical and electrochemical mechanisms of these processes are still not completely understood. Most of the papers related to TU are devoted to the study of anodic processes, such as the protection against corrosion of copper-iron alloys structures9, lixiviation10, gold cementation11, etc.
Some studies focused in the interaction of TU with different electrodes surfaces (Au, Pt, Ag, Cu), provide evidences that support the participation of the TU/FDS redox couple, according to the following irreversible reaction12,14
[Cux(TU)x+y]x+ complexes on copper electrodes polarized at positive potentials has been also reported12. The formation of this type of complexes is restricted to metals less noble than platinum, including the gold electrode9-15. Besides, in the potential domain of cathodic processes the electroreduction of metallic complexes as well as the FDS adsorption are involved16. On platinum electrodes there is no evidence of complexes formation and only the partial or total oxidation of FDS has been reported1.
Although there are an important amount of papers dealing with the electrochemical behavior of TU on different electrodes17,20, information regarding the role that this additive plays in the copper electrodeposition process is scarce. Aspects such as the interaction of TU with the electrode and electrolyte, chemical and electrochemical formation of FDS in strongly acid solutions, Cu-TU complexes formation, and electrode surface inhibition are important points that need to be systematically addressed. In current paper, we have studied the electrochemical behavior of TU on gold and copper electrodes in an acid electrolyte in presence and absence of Cu(II) ions. Cyclic voltammograms and nanogravimmetric measurements recorded on both electrodes at variables TU concentrations give useful information to get a deeper knowledge of the Cu/TU/acid solution interface, helping to interpret the complexity of reactions involved in these electrochemical systems.
Cyclic voltammetry and nanogravimmetric measurements were carried out in a conventional three-electrode cell. Table 1 summarizes the characteristics of the working electrodes used.
A saturated mercurious sulfate electrode was used as reference (-0.690 vs NHE), all potentials in the text are referred to this electrode. For voltammetric experiments the counter-electrode was a platinum wire with a geometric area of 3 cm2, whereas for measurements with EQCM a 1.37 cm2 titanium platinized wire was employed. Before each run of CV and EQCM experiments, the electrolytic solutions were de-aerated for 45 minutes with high purity argon, a fux of this gas was maintained during the measurements. A computer controlled potentiostat (Autolab, Model PGSTAT30 ) and a EQCM (Mastex, Model QCM) were used for cyclic voltammetry and nanogravimmetric experiments, respectively. The following solutions were prepared with analytical grade reagents (Merck P.A.) in de-ionized water: 0.30 mol l-1 H2SO4 (soln. A ), 0.8 mol l-1 H2SO4 (soln. B), 1.8 mol l-1 H2SO4 (soln. C), 0.3 mol l-1 H2SO4 , 50 mmol l-1 CuSO4 (soln.D).
RESULTS AND DISCUSSION
1. Cyclic Voltammetry and nanogravimetric studies at gold substrates.
The voltammetric response of TU was frst investigated employing a gold working electrode in sulfuric acid solution, in absence of Cu(II) ions. In absence of TU, the voltammogram of the Au/H2S04 interface recorded after 15 minutes of potential scans presents the typical anodic and cathodic current contributions related to the electroformation and electroreduction of gold oxide species, (Fig.1).
The addition of variable amounts of TU produces new anodic current contributions in the voltammogram (Fig. 2): a current plateau between -0.65 and 0.30 V (A1), not signifcatively affected by TU concentration; A2 peak with a maximum at 0.54 V, strongly dependent on TU concentration and, A3 peak in the potential region of gold oxides formation. As TU concentration increases, the latter contribution undergoes a small current increase, probably due to the presence of adsorbed TU species which can block electrode surface sites. The latter assumption is confrmed during the cathodic scan because the charge of current peak C3, related to the electroreduction of gold oxides formed in the anodic scan, progressively diminishes as a consequence of a TU blocking effect.
The increasing in TU concentration (in the range 4.7⋅10 - 4 to 1.4⋅10 - 3 mol L - 1 ) reveals an increase in the current associated to process A1 and the apparitionof a new cathodic peak (C1) at -0.41 V (Fig. 3). Considering the irreversible character of the electrochemical processes involved in these conjugated peaks, they are probably related to the TU/FDS redox couple. On the other hand, the developing and shifting of peak A2 towards positive potential (overlapped with peak A3), would be related to the formation of Au(I)-TU complexes [21 - 23].
In order to establish the conjugation of the different redox couples involved, the anodic switching potential, Eλ, was systematically moved from 1.0 V towards negative values (Fig.4). Under these conditions, the conjugation between the electroformation and electroreduction of gold oxides (peaks A3 and C3, respectively) appears clearly resolved. Moving E between 0.70 and 0.30 V produces peak C2 (conjugated with the A2 process), which is assigned to the reduction of Au(I)-TU complexes . Similarly, between 0.30 and -0.14 V the conjugation of the A1/C1 peaks, associated to the irreversible TU/FDS couple is also confrmed [18, 21 - 23].
In order to get a deeper insight on the electrochemical system under study, a set of voltammetric/nanogravimetric coupled experiments were conducted,. Figure 5 presents the j/E and Δm/E curves corresponding to the AuEQCM / solution A interface, in presence of 1.4⋅10- 3 mol⋅L- 1 TU. It is observed that in the -0.09 y 0.25 V potential interval, the electrooxidation of TU to FDS is produced together with a mass loss attributed to the formation and liberation of hydrogen ions according to reaction:
This process probable occurs simultaneously with the loss of either water molecules or sulfate ions adsorbed at the interface, which are removed due to the FDS formation. In the 0.25 to 0.5 V potential interval there is a new mass loss, attributed in this case to the gold electrodissolution and further formation of a Au-Tu complex produced through reaction,
This reaction is followed by the formation of gold oxide with the corresponding increase of mass. As in the 1.05 and 0.75 V potential interval the formation of gold oxides is favored, the mass still increases during the cathodic sweep Between 0.75 and -0.25 V there is an important mass decreasing, associated to the oxides gold reduction. This is confrmed by the reduction peak appearing in the voltammogram of Figure 5 in the same potential interval. Finally, from -0.25 V backwards, there is again a mass increase, now related to the reduction of FDS and Au-TU complex to TU.
2. Cu/H2SO4/TU interface.
Figure 6a presents the voltammetric behavior of Cu/solution A interface recorded between -1.24 and -0.34 V, starting in the anodic direction from the open circuit potential(ca -0.39 V). The typical behavior of a copper electrode is observed in absence of TU. An important diminution of the charge associated to copper redox process is observed when TU is progressively added to the electrolytic solution (see inset of figure 6a). This effect can be again explained assuming the blocking of copper sites either by TU, FDS or by both species.
Scanning the potential from the open circuit potential and E = -0.36 V in the anodic direction, the increase of TU concentration from 5.710- 4 to 210- 3 mol-L 1 (see inset of figure 6b) produces peaks A1 and A2. The former has been attributed to the electroadsorption of TU to FDS , whereas the latter is associated to the electroformation of Cu(I)-TU complexes, previously reported in the literature . The corresponding conjugated peaks C1 and C2 are observed during the cathodic scan. The correlation of these processes at constant TU concentration was followed moving progressively E towards cathodic values (Figure 7) and recording the j/E and Δm/E ) curves obtained with the gold electrode of the EQCM, covered with a copper layer.
Figure 7a shows the relationship between A2/C2 and A1/C1 conjugated processes: the former are related to the participation of the Cu-TU(I) complex, represented by the redox couple, and, the latter is associated to the TU/FDS couple according to reaction:
So far, it is important to highlight the differences observed for copper and gold electrodes regarding the potential interval where the complexes are reduced at both interfaces. On the copper electrode, the process takes place at potentials more positives than the electroreduction of FDS to TU. On the contrary, on the gold electrode, the TU complexes are reduced at potential negatives to the FDS/TU couple, indicating a greater electrochemical reversibility of the copper complexes. This behavior suggests the possibility that TU may act as a de-polarizing agent for copper electroreduction reaction.
As it was found for the gold electrode, in the potential region where the formation of CU(I)-TU complexes prevails, the general tendency of the Δm/E curves(figure 7b) associated to the TU-FDS couple is towards a mass decreasing. When the potential excursion is extended towards the Eλ,A anodic switching potential limit, a mass increase due to formation of copper oxides is produced. Due to their thermodynamic stability, this mass increasing is observed until a potential of ca -0.60 V is reached. Afterwards, a loss of mass due to the reduction of both, copper oxides and Cu(I)-TU complexes adsorbed at the electrode surface, is observed at more cathodic potentials. As in the case for gold electrodes, after these processes the ∆m/E profles show a mass increase due to the electrochemical reduction of FDS to TU.
3. Au/CuSO4/H2SO4/TU interface
The presence of Cu(II) ions introduces important modifcations in the electrochemical behavior of the electrodes under study. Actually, in aqueous solutions they can react with TU giving a Cu(I)-TU complex and FDS, the same species that are electrochemically formed at the working electrodes as just described in preceding sections.
Besides the characteristic behavior of the gold substrate in absence of TU, the voltammogram recorded in theanodic direction, starting from the open circuit potential, presents current peak A1, with a maximum at -0.10 V, which is related to the stripping of the copper deposited after successive potential sweeps (Figure 8). The electroreduction of Cu(II) to Cu(0) starts at - 0.42 V, with a maximum at -0.70 V. Adding increasingly amounts of TU produces an additional cathodic peak at -0.50 V (C1),attributed to the reduction of CuTU+ formed from reaction:
According to previous voltammetric description, the following tendencies are observed: i) for TU concentrations ranging from 1⋅10-5 mol⋅L-1 to around 4⋅10- 5mol⋅L-1 the current of peak C1 increases and remains constant at higher concentrations. In turn, in agreement with the formation of a complex, the current of peak C1 diminishes as the TU concentration increases; ii) the depolarization of the copper electrodeposition process takes place in the same concentration range, the inset in figure 8 accounting for this behavior. Depolarization is produced by the increase in the stripping (peak A1) of the electrodeposited copper that was generated with the participation of the Cu-TU complex (peak C1). In fact, as the Cu-TU complex requires one equivalent of charge in order to achieve copper electrodeposition, additionally contributes to increase the overall electrodeposition rate.
Regarding the electrode polarization, the increase of TU concentration over 7.9⋅10-5 mol⋅L- 1 produces the inverse behavior observed in the voltammogramas of Figure 8. This is appreciated in the cathodic hemicycle (Figure 9) where a charge decrease in the Cu(I)/Cu (close to 0.5 V), and in the Cu(II)/Cu couples is observed. This means that the blocking of the electrode surface takes place in an extension that not only modifes the charge of the processes involved, but also shifts their respective potentials towards more negative values. Under these circumstances, the electrode surface is completely modifed by the massive adsorption of FDS according to:
The progressive decrease of the peak associated to the oxidation of Cu electrogenerated in the successive scans confrms this voltammetric behavior.
4. EQCM Au/Cuelectrodeposited/CuSO4/H2SO4/TU interface
In order to establish the charge/mass relationships of the electrochemical system under study, the j/E and Δm/E response were evaluated (Figure 10).
Figure 10a shows the typical response of the gold electrode in a copper acid solution ( i.e. mass increase in the cathodic scan and mass loss during copper stripping). In the presence of TU (figure 10b), the blocking of the electrode surface, is evidenced. Although in the 0.5 and -1.0 V interval appear two mass changes attributed to Cu(I)-TU → Cu(0) and Cu(II) → Cu(0) reductions, this behavior reveals that copper electrodeposition is still the most important process involved. To confrm these assumptions, relationships for the variation of charge and mass over time were determined through the Q/t and Δm/t transients for further determining Q/ Δm, and the number of electrons involved in the copper electrodeposition process.
Figure 11a shows the evolution of the charge and Δmass curves in the potential interval where only the copper electrodeposition takes place. From Figure 11b is deduced the exchange of either 1 and 2 electrons in the electrochemical reactions related to the following redox couples :
According to these charge/mass relationships, the Cu(I)-TU redox couple should participate in the beginning of the copper electrodeposition process. Furthermore, as it appears nearby the open circuit potential, it behaves like a reversible system.
The ability of TU to form Cu(I) complexes that depolarizes the copper electrodeposition reaction could be the reason behind its wide use as a leveling agent in copper refneries. Because relatively high TU concentrations produces an increase in FDS, resulting in the blocking of the active sites at the electrode surface, the appropiate monitoring of the concentration of this additive in the tanks of industrial copper refneries deserves to be considered as an important factor for obtaining copper cathodes with an adequate morphology.
Cyclic voltammetry and EQCM experiments showed that in acid solutions onto gold and copper electrodes TU is irreversibly electroadsorbed as FDSads, which remains at the surface blocking active sites. Current diminution and a shift in the potential of the electrochemical processes involved in the absence of TU in these electrodes were observed. A mass loss, attributed to the exit of hydronium ions and water molecules from the adsorbed species was observed during the electroadsorption process. On the other hand, the ∆m vs E profle related to the electrochemical formation of the Cu(I)-TUads presented a loss of mass attributed to the exit of TU from the electrode surface.
The formation of Cu(I)-TU and FDS was confrmed in the bulk of copper sulfate acid solutions. Actually, themass changes associated to TU/ FDS were not signifcant enough to prevent the copper electrodeposition process. However, the addition of TU produces a change in the mass response because Cu(I)-TU is electroreduced to copper in the initial stages of the nucleation. Afterwards, as was observed in the charge-mass transient, the growth of the phase would be ruled by the electroreduction of Cu(II) to Cu(0).
This work has been supported by FONDEF (Chile), project D03I1148.
1. A.E. Bolzán, I.B. Wakenge, R.C. Salvarezza, A.J. Arvia, J. Electroanal. Chem, 475, 181. (1999). [ Links ]
2. A. Lukomska, S. Smolinski, J. Sobkowski, Electrochim. Acta, 46, 3111. (2001). [ Links ]
3. M. Alodan, W. Smyrl, Electrochim. Acta, 44, 299. (1998). [ Links ]
4. V. San Martín, S. Sanllorente, S. Palmero, Electrochim. Acta, 44, 579. (1998). [ Links ]
5. G.M. Brown, G.A. Hope, D.P.Schweinsberg, P.M. Fredericks, J. Electroanal. Chem, 380, 161. (1995). [ Links ]
6. G.M. Brown, G.A. Hope, J. Electroanal. Chem, 413, 153. (1996). [ Links ]
7. A. Vicenzo, P.L. Cavallotti, J. Appl. Electrochem. 32, 743. (2002). [ Links ]
8. P. Schilardi, S. Méndez, R. C. Salvarezza, and A. J. Arvia, Langmuir, 14, 2515. (1998). [ Links ]
9. C.M. Juarez and A.J. B. Dutra. Minerals Engineering, 13, 1083. (2000). [ Links ]
10. Z.D. Stankovic and M. Vukovic Electrochim. Acta, 41,) 2529. (1996). [ Links ]
11. H.G. Zhang, J.A.Doyle, C.C. Kenna, S.R. La Brooy, G.T. Hefter and I.M. Ritchie. Electrochim. Acta, 41. 389. (1996). [ Links ]
12. A.E. Bolzán, P.L. Schilardi, R.C.V. Piatti, T. Iwasita, A. Cuesta, C. Gutiérrez, A.J. Arvía. J. Electroanal Chem, 571, 59. (2004). [ Links ]
13. B. Reents, W. Plieth, V.A. MAcagno, G. I. Lacconi. J. Electroanal. Chem. 453, 121. (1998). [ Links ]
14. M. Szklarczyk, N. Nhu Hoa, P. Zelenay. J. Electroanal. Chem. 405, 111. (1996). [ Links ]
15. A.E. Bolzán, T. Iwasita, A.J. Arvia, J. Electroanal.Chem. 554-555, 49. (2003). [ Links ]
16. G. García, V. A. Macagno, G. I. Lacconi. Electrochim. Acta, 48, 1273. (2003). [ Links ]
17. A.E. Bolzán, P.L. Schilardi, R.C.V. Piatti, T. Iwasita, A. Cuesta, C. Gutierrez, A.J. Arvia, J. Electroanal.Chem. 571, 59. (2004). [ Links ]
18. A.E. Bolzán, R.C.V. Piatti, A.J. Arvia, J. Electroanal.Chem. 552, 19. (2003). [ Links ]
19. M. Yan, K. Liu, Z. Jiang, J. Electroanal.Chem. 408, 225. (1996). [ Links ]
20. I. Lukovits, I. Bakó, A. Shaban, E. Kálmán, Electrochim. Acta, 43, 131. (1998). [ Links ]
21. G. García, J. L. Rodríguez, G. I. Lacconi, E. Pastor, Langmuir, 20, 8773. (2004) [ Links ]
22. M. Tian, B.E. Conway, J. Appl. Electrochem. 34,) 533. (2004). [ Links ]
23. A. Fanigliulo, C. Mele, B. Bozzini, Transactions of the Institute of Metal Finishing, 81, 75. (2003). [ Links ]
(Received: September 22, 2008 - Accepted: August 25, 2009).