{Local electrochemical deposition of copper from sulfate solution:} http://dx.doi.org/10.5599/jese.1352 557 J. Electrochem. Sci. Eng. 12(3) (2022) 557-563; http://dx.doi.org/10.5599/jese.1352 Open Access : : ISSN 1847-9286 www.jESE-online.org Original scientific paper Local electrochemical deposition of copper from sulfate solution Georgii Vasyliev, Viktoria Vorobyova, Dmytro Uschapovskiy and Olga Linucheva National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, 37, Prospect Peremohy, Kyiv-56, 03056, Ukraine Corresponding author: g.vasyliev@kpi.ua ; Tel.: +38-096-924-9888; Fax: +38-044-204-9773 Received: April 20, 2022; Accepted: May 26, 2022; Published: June 7, 2022 Abstract Local electrochemical deposition is a type of electroplating used to plate metal locally or form metal objects using electrochemical principles at a short distance from the working electrode. In this work, deposition of the copper spot was modelled using COMSOL software and experimentally tested in copper sulfate electrolyte using soluble copper anode. The working capillary diameter was 4 mm and the interelectrode distance was 5 mm. The deposited copper of 100 µm thickness was investigated using the 3D-profilometry technique. The geometry of deposited metal was found to be in good accordance with the COMSOL model. The inclusions of anodic sludge were responsible for the surface inhomogeneity of the deposited copper. Keywords Local electrochemical deposition; additive manufacturing; copper plating; computer modelling; throwing power; anodic sludge Introduction Additive manufacturing (AM) is an emerging trend in industry and materials science that provides simple manufacturing of complex objects and opens new approaches in the materials design and manufacturing process. The variety of AM materials becomes wider every year simultaneously with the development of manufacturing techniques. Metal AM is a challenging task due to their high melting temperature. The most industrially developed technique is based on the local melting of metal powder with a laser or electron beam [1,2]. Despite the ability to produce complex objects, the technique requires metal powder preparation and a high energy power source. Electrochemical deposition can be an alternative to the local laser melting because of its low energy consumption (the process takes place at room temperature) and the object is not subjected to temperature influence. Several electrochemical AM techniques are known today. They can be divided into two main gro- ups, depending on the deposition conditions: mask-based deposition and mask-less deposition [3]. In the former technique, the cathode is covered with the insulating mask covering areas where deposition should not occur. So, the metal deposits on uncovered surfaces [4]. Another mask-based http://dx.doi.org/10.5599/jese.1352 http://dx.doi.org/10.5599/jese.1352 http://www.jese-online.org/ mailto:g.vasyliev@kpi.ua J. Electrochem. Sci. Eng. 12(3) (2022) 557-563 LOCAL ELECTROCHEMICAL DEPOSITION OF COPPER 558 technique is instant masking plating. In this technique, the mask moves with an anode. The depo- sition starts only when the mask is attached to the surface. Once the electrolyte is depleted, the depositions stop, the mask is lifted away, and the electrolyte is renewed [5]. Electrochemical fabrication is a manufacturing technique where the deposition of structural material is alternated with the deposition of sacrificial metal, which plays the role of a mask [6]. Once the deposition of the whole structure is completed, the sacrificial metal is etched to obtain the final structure. Mask-less deposition techniques are carried out with no mask on the surface and the environment around the electrode is open. The first example of this group is localized electrochemical deposition (LECD), proposed in 1995 [7]. The technique is based on the deposition of metal in the localized area under the working electrode. To get a very localized electric field, an ultrafine inert anode tip and small interelectrode gap are used. Further development of the proposed technique was focused on anode construction improvement. The anode was placed inside the dielectric capillary that provides an additional focus of the electric field [8-10]. Jet electrodeposition is another LECD technique, where the impinging anodic electrolyte jet as anode tool to deposit metal selectively deposits metal on the cathode [11]. In meniscus-confined electroplating, the deposition is located in a meniscus-shaped electrolyte bridge between a narrow anode tip and cathode substrate [12-16]. The deposition rate is determined by the diffusion rate of ions and is very low. Extremely small objects can be obtained using fluidic force microscope electrodeposition [17]. In this technique, the hollow AFM cantilever with an aperture at the tip is used. During deposition, the cantilever serves as a working electrode and is filled with electrolyte. The flow of electrolyte from the probe is controlled by pressure regulation. Once the probe is positioned, the deposition starts and ends when the deposited metal touches the tip. Then the probe moves to the next position. Comparing different electrochemical additive manufacturing techniques, it can be seen that the higher precision of manufacturing, the lower the deposition rate. Further investigations are required to develop a highly productive ECAM technique that is able to produce large-scale objects at commercially acceptable rates. In this work, the local electrochemical deposition of copper was performed using a computer model and compared to electrochemically deposited metal using 3D-profilometry. Computer modelling The computer model of the local electrochemical deposition was built using COMSOL Multiphysics 5.2 software. The scheme of the cell is given in Figure 1a. The computer model gives the ability to predict the localization of the electric field and thus the location of the deposited metal. Moreover, the model gives the expected profile of the deposited copper based on the electrical field distribution and deposition duration. The localization of deposition is highly dependent on the throwing power of the electrolyte. The letter one depends on the conductivity of the solution. The conductivity of the copper plating electrolyte depends on the CuSO4 content, but in technological processes, the sulfuric acid is added. So, the addition of H2SO4 increases the conductivity, which makes the distribution of deposited copper more uniform and decreases the voltage load on the plating bath. On the contrary, in this study copper deposition is localized, so the amount of acid should be reduced. To study the influence of H2SO4 content on copper local deposition, the calculations were performed for several concentrations. The COMSOL electrochemical deposit module requires the value of polarizability of the cathodic electrochemical process. The polarizability di/dE was determined experimentally (Figure 2). G. Vasyliev et al. J. Electrochem. Sci. Eng. 12(3) (2022) 557-563 http://dx.doi.org/10.5599/jese.1352 559 Galvanodynamic cathodic polarization curves were obtained using conventional three-electrode cell. The scan rate was 20 mA s-1. The plates of 11 cm2 manufactured from copper grade M0 served as working and auxiliary electrodes, and a saturated silver chloride electrode (SSCE) was used as a reference one (Eref = 0.2 V vs. NHE). The potential values on polarization curves are given vs. SSCE. Standard electrode potential of copper ECu2+/Cu = +0.34 V vs. NHE or ECu2+/Cu = +0.14 V vs. SSCE, which agrees well with experimental data (Figure 2). a b Figure 1. Scheme (a) and photo (b) of electrochemical cell: 1 – cathode; 2 – anode; 3 – polypropylene tube; 4 – electrolyte Figure 2. Cathodic polarization curves of copper deposition in the electrolyte containing 200 g/L CuSO4 and sulfuric acid in different concentrations The limiting current density appeared to depend on the sulfuric acid content even in the same concentration of copper ions [18]. This can be explained by the influence of migration part of mass transfer. According to equation (1), the limiting current density, iL, of copper reduction is dependent on the copper ion transport number: ( ) 2+ 2+ Cu L Cu 2 1 FDC i t  = − (1) where D – diffusion coefficient of copper ions, CCu2+ - copper ions concentration, tCu2+ - copper ions transport number,  - diffusion layer thickness. 0 40 80 120 160 -0.50-0.40-0.30-0.20-0.100.000.100.20 i / m А cm -2 Е / V 10 g/L acid 30 g/L acid 150 g/L acid di / dE = -796 di / dE = -516 di / dE = -400 http://dx.doi.org/10.5599/jese.1352 J. Electrochem. Sci. Eng. 12(3) (2022) 557-563 LOCAL ELECTROCHEMICAL DEPOSITION OF COPPER 560 The electrolyte solution contains the following ions: H+, Cu2+, SO42– and copper ions transport number can be determined by the equation (2): 2+ 2+ 2+ 2+ 2+ 2- 2- + + 4 4 Cu Cu Cu Cu Cu SO SO H H 2 2 2 2 C t C C C     + = + + + + + (2) where  Cu2+, SO42-, H+ - are molar conductivity of copper, sulfate, and proton ions. Due to the much higher molar conductivity of protons, the ion transport number of copper will decrease as acid concentration increases, so the limiting current density of copper deposition will decrease as well. The configuration of the cell is given in Figure 3a. The diameter of the working nozzle was 4 mm, and the copper wire was placed inside it and worked as a soluble anode. The distance between the copper anode and the surface was 5 mm. The gap between the nozzle and the surface was 0.1 mm. The deposition duration was 70 min. The rest of the parameters are given in Table 1. Table 1. Copper local deposition modelling parameters Parameter Study number 1 2 3 H2SO4 content, g L-1 10 30 144  / S сm-1 0.050 0.125 0.433 (di / dE) / mA cm-2 V-1 796 516 400 φext / V 0.82 0.55 0.49 The results of modelling are given in Figure 3b. The growth profiles show that the copper was deposited locally in the form of a cylinder, with a thickness of 100 µm. However, the curvature of the side surface depends on the acid concentration. The closest profile was obtained in the lower tested acid concentration of 10 g L-1. The increase of H2SO4 content to 150 g L-1 makes the deposited copper spot nearly 2 times wider. So, for experimental testing, the electrolyte with 200 g L-1 CuSO4 and 10 g L-1 H2SO4 was chosen. a b Figure 3. Current density distribution in the interelectrode area of the cell (a) and the growth profiles of copper (b), modelled for a 0.1 mm gap between the nozzle and the surface in 200 g/L sulfate electrolyte and different sulfuric acid concentrations Experimental The deposition of copper was performed in the same electrochemical cell, consisting of two electrodes (Figure 1b). The same cell configuration was used for computer modelling. The copper 0 20 40 60 80 100 120 0.146 0.148 0.150 0.152 0.154 δ / μ m x / m 10 g/L acid 30 g/L acid 150 g/L acid Surface: Electrolyte current density, A m-2 x / mm y / m m G. Vasyliev et al. J. Electrochem. Sci. Eng. 12(3) (2022) 557-563 http://dx.doi.org/10.5599/jese.1352 561 plate was placed at the bottom of the vessel, filled with electrolyte, and attached to the negative pole of the current source. The counter electrode was made from copper wire and placed inside the polypropylene tube of 50.5 mm. The tube was oriented in the center of the copper plate and the distance between the plate and the pipe was 5 mm. This arrangement was maintained during the deposition process. The electrolyte used in the study contained 200 g L-1 CuSO4 and 10 g L-1 H2SO4. Working current density was determined from polarization curves and composed 65 mA cm-2. The height of the deposited copper was set to 100 m, so the calculated deposition time was 70 min. The electrolysis was performed at ambient temperature. The characterization of the deposited copper was performed with the 3D-profilometry technique. The sensitive piezoelectric sensor with a vertical resolution of 250 nm/dot and horizontal resolution of 625 nm dot-1, scanned the area of 77 mm2 around the deposited copper spot. The obtained data were collected using a PC and SURFARE software was used to get a 3D image of the surface. The height of the deposited copper, surface roughness and precision of deposition was determined and compared to the COMSOL model. Results and discussion Copper deposition is an electrochemical reduction process which undergoes according to equation (3): Cu2+ + 2e- = Cu (3) The rate of copper deposition is determined by the rate of copper supply to the surface, so the working current density was determined experimentally. Figure 2 shows the polarization dependence of copper that was used to determine the working current density. The equilibrium potential of copper in the electrolyte was 0.1 V vs. SSCE. Limiting current density was found to be 120 mA cm-2. However, limiting current densities are never applied for metal deposition due to the formation of low-quality deposits with powder inclusions. The working current density was selected to be half of the limiting – 65 mA cm-2, which gives the deposition rate of 85 µm h-1. Deposited copper is shown in Figure 4. It can be seen that the surface of the deposited metal is not smooth and dark inclusions can be observed. Figure 4. Locally deposited copper on the surface of copper plate during 3D-profilometry The measured profile of the deposited copper and the comparison of the modeled and experimentally deposited copper are given in Figure 5. The profile of the deposited copper shows that the average height agrees well with the COMSOL computer model. http://dx.doi.org/10.5599/jese.1352 J. Electrochem. Sci. Eng. 12(3) (2022) 557-563 LOCAL ELECTROCHEMICAL DEPOSITION OF COPPER 562 a b a b Figure 5. 3D-vizualization of the deposited copper (a) and comparison of deposited, modelled and target profiles of copper (b) The side surface of the profile obtained in the experiment is even more straight than in model. However, some areas on the top of the deposited copper demonstrate the presence of peaks, which makes the surface rough. These peaks result from anodic sludge inclusion in the deposits and deposition of copper dendrites. The lower the sulfate acid content in the electrolyte, the lower is copper deposition overvoltage that leads to the formation of dendrites. Also, the dendrite formation may be caused by monovalent copper ions reproportion described by equation (4). Cu+ ions are formed by staged copper anodic dissolution. This process is known to stimulate the increase of roughness and dendrites formation. 2Cu+ = Cu + Cu2+ (4) In the experiment, the soluble anode was used, and the sludge was able to detach from the anode and fall on the surface, where it was grown in the copper. To avoid the inclusions of anodic sludge in the deposited metal and reproportion of copper in further studies, the insoluble anode will be used, and copper overvoltage will be increased using organic additives. Conclusions COMSOL computer model is adequate and can be used to select preferable electrolyte conductivity, determine interelectrode gap and predict current distribution during local electrochemical deposition of metals. In the present study the conductivity was 50 mS cm-1, IEG - 5 mm. The deposited copper profile was found in good agreement with the model. The thickness of the metal layer was 0.1 mm, and the diameter 4 mm, which agrees with the model. The inclusions of the anodic sludge and dendrites formation were responsible for the surface roughness of the deposited copper. The use of insoluble anode and prevention of dendrites formation in further studies would be preferable to prevent sludge formation. Acknowledgements: The work was supported by the Ministry of Science and Education of Ukraine, grant number 2514, 2022. References [1] C. Korner, International Materials Reviews 61 (2016) 361-377. https://doi.org/10.1080/09506608.2016.1176289 0.00 0.05 0.10 0.15 0.20 0.25 0 1 2 3 4 5 6  / m m x / mm Experimental Model Target https://doi.org/10.1080/09506608.2016.1176289 G. Vasyliev et al. J. Electrochem. Sci. Eng. 12(3) (2022) 557-563 http://dx.doi.org/10.5599/jese.1352 563 [2] P. Regenfuss, A. Streek, L. Hartwig, S. Klotzer, Th. Brabant, M. Horn, R. Ebert, H. Exner, Rapid Prototyping Journal 13(4) (2007) 204-212. https://doi.org/10.1108/13552540710776151 [3] L. Xinchao, M. Pingmei, A. Sansan, W. Wei, International Journal of Machine Tools and Manufacture 173 (2021) 103848. https://doi.org/10.1016/j.ijmachtools.2021.103848 [4] S. D. Leith, D. T. Schwartz, Journal of Microelectromechanical Systems 8(4) (1999) 384-392. https://doi.org/10.1109/84.809052 [5] A .L. Cohen, U. Frodis, F. G. Tseng, G. Zhang, M. Florian, P. M. Will, Micromachining and Microfabrication Process Technology V 3874 (1999) 236-247. https://doi.org/10.1117/12.361227 [6] A. L. Cohen, G. Zhang, F. G. Tseng, F. Mansfield, U. Frodis, P.M. Will, International Solid Freeform Fabrication Symposium, EFAB: batch production of functional, fully-dense metal parts with micron-scale features, Austin, USA, 1998, p. 161-168. https://doi.org/10.26153/tsw/590 [7] J. D. Madden, S. R. Lafontaine, I.W. Hunter, Sixth International Symposium on Micro Machine and Human Science, Fabrication by electrodeposition: building 3D structures and polymer actuators, Nagoya, Japan, 1995, pp. 77-81. https://doi.org/10.1109/MHS.1995.494221 [8] J. Xu, W. Ren, Z. Lian, P. Yu, H. Yu, The International Journal of Advanced Manufacturing Technology 110 (2020) 1731-1757. https://doi.org/10.1007/s00170-020-05799-5 [9] P. Hanekamp, W. Robl, F.‑M. Matysik, Journal of Applied Electrochemistry 47 (2017) 1305- 1312. https://doi.org/10.1007/s10800-017-1124-8 [10] S. Morsali, S. Daryadel, Z. Zhou, A. Behroozfar, D. Qian, M. Minary-Jolandan, Journal of Applied Physics 121 (2017) 024903-024908. https://doi.org/10.1063/1.4973622 [11] H. Fan, Y.P. Zhao, S.K. Wang, Key Engineering Materials 667 (2016) 259-264. https://doi.org/10.4028/www.scientific.net/KEM.667.259 [12] J. Hu, M.F. Yu, Science 329 (2010) 313-316. https://doi.org/10.1126/science.1190496 [13] S. K. Seol, D. Kim, S. Lee, J. H. Kim, W. S. Chang, J. T. Kim, Small 11(32) (2015) 3896-3902. https://doi.org/10.1002/smll.201500177 [14] A. Behroozfar, S. Daryadel, S. R. Morsali, S. Moreno, M. Baniasadi, R. A. Bernal, M. Minary- Jolanda, Advanced Materials 30(4) (2017) 1705107. https://doi.org/10.1002/adma.201705107 [15] P. Liu, Y. Guo, Y. Wu, J. Chen, Y. Yang, Crystals 10(4) (2020) 257. https://doi.org/10.3390/cryst10040257 [16] K. Nakazawa, M. Yoshioka, Y. Mizutani, T. Ushiki, F. Iwata, Microsystem Technologies 26 (2020) 1333-1342. https://doi.org/10.1007/s00542-019-04665-z [17] G. Ercolano, C. van Nisselroy, T. Merle, J. Vörös, D. Momotenko, W. W. Koelmans, T. Zambelli, Micromachines 11(1) (2020) 6. https://doi.org/10.3390/mi11010006 [18] O. V. Linyucheva, M. I. Donchenko, D. Yu. Ushchapovskiy, M. V. Byk, D. M. Skladanniy, Eastern-European Journal of Enterprise Technologies 6 (2014) 48-55. https://doi.10.15587/1729-4061.2014.30660 ©2022 by the authors; licensee IAPC, Zagreb, Croatia. This article is an open-access article distributed under the terms and conditions of the Creative Commons Attribution license (https://creativecommons.org/licenses/by/4.0/) http://dx.doi.org/10.5599/jese.1352 https://doi.org/10.1108/13552540710776151 https://doi.org/10.1016/j.ijmachtools.2021.103848 https://doi.org/10.1109/84.809052 https://doi.org/10.1117/12.361227 https://doi.org/10.26153/tsw/590 https://doi.org/10.1109/MHS.1995.494221 https://doi.org/10.1007/s00170-020-05799-5 https://doi.org/10.1007/s10800-017-1124-8 https://doi.org/10.1063/1.4973622 https://doi.org/10.4028/www.scientific.net/KEM.667.259 https://doi.org/10.1126/science.1190496 https://doi.org/10.1002/smll.201500177 https://doi.org/10.1002/adma.201705107 https://doi.org/10.3390/cryst10040257 https://doi.org/10.1007/s00542-019-04665-z https://doi.org/10.3390/mi11010006 https://doi.10.15587/1729-4061.2014.30660 https://creativecommons.org/licenses/by/4.0/)