Solubility behavior of aluminum anodes for aluminum deposition

Aluminum and aluminum alloys are interesting materials for a wide range of applications, e.g. in the field of corrosion protection [4,5] and in energy storage [6-8]. Their electrodeposition is a key technology, but can only be realized using aprotic electrolytes (e.g. ionic liquids). The use of soluble anodes is an important basis for continuous galvanic processes. They allow the constant replenishment of metal ions consumed by reduction on the workpiece and prevent undesirable anodic electrolyte decomposition. A large number of metals that passivate in aqueous electrolytes (e.g. aluminum) can be dissolved anodically in ionic liquids [9]. Nevertheless, passivation phenomena can occur under certain conditions, the understanding of which is fundamental for the realization of stable coating processes.

Ionic liquids (ILs) have proven to be suitable media for the deposition of aluminum and aluminum alloys [4,10]. One of the most intensively investigated ILs in this context is the mixture of 1-ethyl-3-methylimidazolium chloride, [EMIm]Cl, with _AlCl3. These electrolytes exhibit good solubility for a large number of metal salts and thus the ability to deposit alloys [11-13]. The Lewis acidity of these electrolytes can be adjusted via the molar ratio of the two components, which determines the ions present (Fig. 1).

Fig. 1: Theoretical anion proportions xAn (Cl-, AlCl4-, Al2Cl7- and Al3Cl10- [14]) and phase diagram (black line with squares) [14-16] of AlCl3/[EMIm]Cl mixtures. Different electrolyte compositions, expressed as molar ratio of AlCl3:[EMIm]Cl, are indicated by vertical dashed lines and the mean temperature of the glove box is indicated by a horizontal dashed line. The conversion of the lower x-axis was based on the data from [1,2]. [1]

For the electrodeposition of aluminum (equation 1), the electrolyte must be Lewis acidic, which is achieved by a molar excess of _AlCl3 [10,17-19].

<1> 4 _Al2Cl7-+ 3 e- → Al + 7 _AlCl4-

By using soluble aluminum anodes, a continuous deposition process (reversal of the reaction in equation 1) can be achieved while avoiding anodic electrolyte decomposition. However, in very Lewis-acid electrolytes and at high current densities, a sharp increase in cell voltage can be observed, which leads to the process coming to a standstill. A large ratio of anode to cathode surface area or intensive convection can counteract this process [20], but an inhomogeneous field distribution in the bath can still lead to the critical current density being exceeded locally and the process coming to a standstill.

Anodic passivation

Despite the high reversibility of the deposition and dissolution of aluminum in the electrolyte [1,2], an abrupt drop in the anodic current density j can be observed in an electrolyte with a molar ratio of 2.0:1 between _AlCl3 and [EMIm]Cl, starting from a value of about 8.5 mA ^cm-2 at potentials above 150 mV against an ^Al/AlIII reference electrode [1,3] (Fig. 2 left). This behavior is typical for passivation processes. Due to the work under inert gas and the exclusion of moisture, this observation cannot be attributed to the formation of an oxide layer on the aluminum anode. In the case of electrolytes with a lower Lewis acidity (e.g. 1.5:1 electrolyte), the dissolution is continuous, although a limitation of the anodic dissolution has also been reported for these electrolytes [20]. Studies on anodic passivation were carried out for both high-temperature molten salts (HTMS) [21-23] and ILs [20,24,25], but were unable to clarify their cause in detail.

The investigation of the process using electrochemical quartz crystal microbalance (EQCM) [1,26-29] shows the onset of the reduction of aluminum at the gold electrode of the quartz crystal used [3] and the associated mass increase at approx. -80 mV vs. ^Al/AlIII(Fig. 2 right), corresponding to the Sauerbrey equation (Equation 2) [27,30,31]. At the same time, the attenuation of the quartz increases, which according to the Kanazawa equation (equation 3) [27,28,32,33] can be explained by the increase in the viscosity-density product of the electrolyte due to the decreasing Lewis acidity as a result of the aluminum reduction at the electrode surface [34,35].

<2>

<3>

In equations 2 and 3, ∆f is the resonant frequency change, _f0 is the resonant frequency of the unloaded quartz, _ρL and ηd_,L are the density and dynamic viscosity of the electrolyte, respectively, _μQ and _ρQ are the shear modulus and density of the unloaded quartz, respectively, ∆ω and ∆m are the change in damping and mass change, respectively, and A is the active electrode area. The change in mass with the amount of charge transferred is on average (84.2±0.1) µg ^C-1 and therefore around 90% of the theoretical value for aluminum, based on equation 4.

<4>

In equation 4, M and z are the molar mass and the number of electrons transferred, respectively, and F is the Faraday constant. Due to the change in electrolyte viscosity and density, the determined mass change is influenced. In addition, the incorporation of light elements (C, N) into the aluminum layer as a result of the cathodic decomposition of [EMIm^]+ must be taken into account, as element analyses of the layer show. A quantitative evaluation is not possible, as the charge-to-mass ratio for this is not known [36,37]. Last but not least, the increase in the roughness of the electrode surface can lead to an increase in the energy dissipation of the quartz and thus to an increase in its attenuation [31]. After reaching the current density or attenuation maximum in the region of the reversal potential at -700 to -800 mV vs. ^Al/AlIII, both decrease again during the anodic scan. The attenuation does not reach the initial values, which is due to diffusion-related equalization processes near the electrode. The mass continues to increase due to the continuing cathodic current densities until a potential of 0 mV vs. ^Al/AlIII is reached and anodic current densities or the anodic dissolution of aluminum and a corresponding decrease in mass can be observed. At a potential of approx. 430 mV vs. ^Al/AlIII, the maximum current density of 8.5 mA ^cm-2 is reached, which has already been observed previously (Fig. 2 left). Analogous to the linear polarization, a passivation of the electrode can be observed. The EQCM data also show a high increase in frequency and mass coupled with an increase in attenuation. The Sauerbrey equation (equation 2) is no longer fully valid at this point [1], which no longer allows an exact calculation of the mass increase. The absence of cathodic currents no longer allows an explanation of the mass increase due to the reduction of aluminum. The oxidation of chloride ions, which limits the electrochemical stability window of the electrolyte used on the anodic side [10], can also not justify these data. Based on the Kanazawa equation (equation 3), however, an increase in the viscosity-density product of the electrolyte can be assumed, which is typical for solidifying liquids. Since a layer that remains permanently on the electrode cannot be observed, it must be assumed that any precipitates will dissolve quickly.

Cause of anodic passivation

The phase diagram for mixtures of [EMIm]Cl and _AlCl3(Fig. 1) shows that at values above 67 mol% _AlCl3 (2.0:1 electrolyte), the melting temperature of the electrolyte increases. This rises from approx. -90 to over 100 °C. In connection with the above observations, it can be concluded that rapid aluminum dissolution leads to an increase in the aluminum ion concentration and thus to an increase in the melting temperature. This is consistent with the described change in attenuation and the assumption of a solidifying liquid. Electrode passivation due to an insulating precipitate on the electrode surface would be the result and would explain the observed decreasing current densities.

On this basis, the cause of the anodic passivation would have to be due to slow diffusion processes. Therefore, current-controlled jump experiments are a suitable method for verifying the hypothesis put forward. The time until solidification and thus passivation occurs can thus be determined and compared with diffusion processes, which is discussed below.

The critical aluminum ion concentration_ccrit(^AlIII), at which electrolyte solidification occurs for a given temperature T, i.e. the solidus line is cut above 66 mol% _AlCl3, can be determined from the phase diagram (Fig. 1). This value corresponds to approx. 7.22 mol ^l-1 (at a temperature of approx. 27 °C prevailing in the glove box). For details on the calculation of this value, please refer to [3]. Based on the work of Sand [38], the concentration profile upstream of the aluminum anode can be calculated as a function of distance s and time t for a 2.0:1 electrolyte with a bulk concentration c* of (6.82 ± 0.01) mol ^l-1 according to equation 5 (Fig. 3).

Fig. 3: Location- and time-dependent concentration profiles of the aluminum ion concentration c(s,t) in front of an Al working electrode in a 2.0:1 electrolyte at 5 mA cm-2 (short-dashed), 10 mA cm-2 (long-dashed) and 20 mA cm-2 (solid) after 30 s (black), 60 s (red) and 120 s (blue), based on equation 5 and data from [1,2]. The horizontal lines represent the bulk concentration c* (green) and the critical concentration ccrit (purple). [1]

<5>

Where D is the diffusion coefficient and erfc is the complementary Gaussian error function.

Despite an expected change in the diffusion coefficient of the aluminum ions with the electrolyte viscosity according to the Stokes-Einstein [2,39] or Sutherland equation [2,40], a constant diffusion coefficient D(_Al2Cl7-) of (7.2 ± 0.3) ^{10-11m2 s-1} is assumed below for simplicity. When the current density is increased from 5 to 20 mA ^cm-2, the time until_ccrit is exceeded decreases from approx. 2 min to less than 30 s, which is qualitatively consistent with observations in galvanostatic experiments. Based on equation 5, the critical transition time _τcrit, i.e. the time until_ccrit is reached starting from the bulk concentration c*, can be defined according to equations 6 and 7.

<6>

<7>

Fig. 4 shows the plot of _τcrit vs. j and the corresponding linearized plot ^τcrit-1/2 vs. j. The experimentally determined values lie between the theoretical values for a 1.9:1 and a 2.0:1 electrolyte with a clear proximity to the latter. The regression of the experimental data shows a linear behavior (Fig. 4 right) with an intersection at the origin of the coordinates (^R2 > 0.999), which confirms diffusion control according to the theory described by Sand [38,41]. The observed deviations compared to theoretical values of a 2.0:1 electrolyte can be explained by delayed nucleation of the solidifying electrolyte (undercooling) [40]. This can also be deduced from the EQCM data, which show an increase in attenuation even before an increase in mass. A further deviation from experiment to theory can be explained by deviations from semi-infinite diffusion and infinite electrode expansion, as required by the underlying theory. Furthermore, the method of calculating the aluminum ion concentration [3], the preparation of the electrolyte (e.g. due to errors in the balance used) and the calculation of the critical concentration from the phase diagram have an effect on the accuracy of the results [1]. Nevertheless, the agreement between the experimental and theoretical data points to reliable conclusions.

For low current densities, an increasing deviation between experiment and theory can be observed (Fig. 4), which can be explained by natural convection. It is striking that the values approach a current density of 8.5 mA ^cm-2, which corresponds to the value known from linear polarization and EQCM experiments and, based on the previous discussion, corresponds to the maximum anodic current density that can be impressed without trans-port limitation or passivation occurring.

The results presented prove a clear correlation between the anode passivation and the melting behavior of the electrolyte. Based on the data discussed, the passivation process at high anodic current densities is an interplay of a local increase of aluminum ions in front of the anode due to anode dissolution, slow diffusion of these ions into the electrolyte interior and the formation of a precipitate on the anode due to local excess of the melting temperature of the electrolyte as a result of the increased aluminum ion concentration. Previously, Wang et al [20] investigated the anode passivation in the electrolyte used here and explained it with the formation of solid _AlCl3 on the electrode surface based on observations in HTMS electrolytes based on _AlCl3-NaCl[22,42]. The phase diagram of the HTMS electrolyte [16,43] shows that the liquidus line is exceeded as the AlCl3 _{concentration} increases, leading to the formation of a partially liquid phase with solid _AlCl3. Although this supports the argumentation in [22] regarding passivation by _AlCl3(s), a transfer to the IL used here is not justified due to the differences in the phase diagrams of the two electrolytes. In contrast to the HTMS electrolyte, the formation of a partially liquid phase does not occur and instead the electrolyte solidifies completely [15,16]. It can therefore be assumed that a precipitate of the solidified electrolyte ([EMIm]_Al2Cl7) is formed, which quickly dissolves by continuous diffusion after the power supply is switched off and could therefore not be analyzed in more detail.

Summary and conclusions

The passivation of soluble aluminium anodes in very Lewis-acid electrolytes and at high current densities was investigated using a combination of linear polarization, electrochemical quartz microbalance and galvanostatic jump experiments. The data show a correlation between diffusion-controlled processes (Fig. 4) and the melting behavior of the electrolyte (Fig. 1). The aluminum ions that form during anodic dissolution of the metal lead to a local increase in concentration, as diffusion into the electrolyte interior is slower than the subsequent supply by aluminum oxidation at the anode (Fig. 3). The local isothermal increase in concentration leads to the critical aluminum ion concentration being exceeded, at which the electrolyte solidifies due to a locally increased melting temperature (Fig. 1). The resulting insulating precipitate then leads to passivation of the electrode surface (Fig. 2) and brings the galvanic process to a standstill.

The increase in melting temperature takes place in a narrow concentration range. Due to melting temperatures far below the freezing point of up to approx. 130 °C (Fig. 1), increasing the process temperature does not appear to make sense. This is particularly true because at temperatures above 60 °C, increasing thermal and accelerated electrochemical decomposition of [EMIm^]+ is to be expected [35], which has a negative effect on the service life of the electrolyte and thus the economic efficiency of the process. In contrast, efficient convection of the electrolyte and a reduction in Lewis acidity are promising measures that can significantly minimize the tendency of the electrolyte to passivate (Fig. 2 left). However, since the solubility of many metal salts in the electrolytes under consideration decreases with decreasing Lewis acidity, a suitable compromise between Lewis acidity and passivation tendency must be found. This plays a particularly important role for the deposition of aluminum alloys, but also has an effect on the cell geometry and the achievable deposition rates. The influence of the passivation tendency of the electrolyte and the type and scope of countermeasures should therefore be weighed up in detail for each application.

This technical article is a supplement to the article "Aluminum and aluminum alloys as cadmium substitutes in the aerospace industry" from the August issue. With his work on the electrochemical deposition of aluminum and aluminum alloys [1], Dr. Böttcher won the 2024 Nasser-Kanani Prize alongside Frank Simchen. This article builds on [1] and has already been published in English in two scientific journals [2,3].

References
[1]R. Böttcher, Electrochemical deposition of aluminum and aluminum alloys from ionic liquids. Dissertation, Ilmenau, 2021.
[2]R. Böttcher, A. Ispas, A. Bund, Determination of transport parameters in [EMIm]Cl-based Ionic Liquids - Diffusion and electrical conductivity, Electrochim. Acta 366 (2021) 137370. https://doi.org/10.1016/j.electacta.2020.137370.
[3]R. Böttcher, A. Ispas, A. Bund, Anodic dissolution of aluminum and anodic passivation in [EMIm]Cl-based ionic liquids, Electrochem. Commun. 115 (2020) 106720. https://doi.org/10.1016/j.elecom.2020.106720.
[4]R. Böttcher, Aluminum and aluminum alloys as cadmium substitutes in aerospace applications, Galvanotechnik (2024) 981-988.
[5]A. Endrikat, R. Böttcher, T. Engemann, A. Ispas, R. Peipmann, A. Bund, Deposition of base metals from ionic liquids for corrosion protection, Galvanotechnik 108 (2017) 2179-2188.
[6]R. Piterek, On the way to the battery of the future, Galvanotechnik (2024) 64-69.
[7]P. Scherzl, M. Kaupp, T. Sörgel, Electroforming of aluminum foils, Galvanotechnik (2024) 156-162.
[8]P. Scherzl, M. Kaupp, T. Sörgel, Electroforming of aluminum foils, Galvanotechnik (2024) 311-317.
[9]A.P. Abbott, G. Frisch, K.S. Ryder, Electroplating Using Ionic Liquids, Annu. Rev. Mater. Res. 43 (2013) 335-358. https://doi.org/10.1146/annurev-matsci-071312-121640.
[10]A.P. Abbott, D.R. MacFarlane, F. Endres (Eds.), Electrode-position from ionic liquids, Wiley-VCH, Weinheim, 2017.
[11]Q.X. Liu, S.Z. El Abedin, F. Endres, Electroplating of mild steel by aluminum in a first generation ionic liquid: A green alternative to commercial Al-plating in organic solvents, Surf. Coat. Tech. 201 (2006) 1352-1356. https://doi.org/10.1016/j.surfcoat.2006.01.065.
[12]L. Simanavičius, A. Staknas, A. Šarkis, Codeposition of aluminum with some metals from AlBr3-dimethylethylphenylammonium bromide solutions containing acetylacetonate of selected metal, Electrochim. Acta 46 (2000) 499-507. https://doi.org/10.1016/S0013-4686" target="_blank">https://doi.org/10.1016/S0013-4686(00)00537-5.
[13]A. Ispas, E. Wolff, A. Bund, An Electrochemical Quartz Crystal Microbalance Study on Electrodeposition of Aluminum and Aluminum-Manganese Alloys, J. Electrochem. Soc. 164 (2017) H5263-H5270. https://doi.org/10.1149/2.0381708jes.
[14]D. Orbakh (Ed.), Nonaqueous electrochemistry, 1st ed., Marcel Dekker, New York, 1999.
[15]A.A. Fannin Jr, D.A. Floreani, L.A. King, J.S. Landers, B.J. Piersma, D.J. Stech, R.L. Vaughn, J.S. Wilkes, J.L. Williams, Properties of 1, 3-dialkylimidazolium chloride-aluminum chloride ionic liquids, J. Phys. Chem. 88 (1984) 2614-2621. https://doi.org/10.1021/j150656a038.
[16]G.R. Stafford, C.L. Hussey, Electrodeposition of transition metal-aluminum alloys from chloroaluminate molten salts, in: R.C. Alkire, D.M. Kolb (Eds.), Advances in Electrochemical Science and Engineering, Wiley-VCH, Weinheim, 2001, pp. 275-348.
[17]T. Jiang, M.C. Brym, G. Dubé, A. Lasia, G.M. Brisard, Electrodeposition of aluminum from ionic liquids: Part I-electrodeposition and surface morphology of aluminum from aluminum chloride (AlCl3)-1-ethyl-3-methylimidazolium chloride ([EMIm] Cl) ionic liquids, Surf. Coat. Tech. 201 (2006) 1-9. https://doi.org/10.1016/j.surfcoat.2005.10.046.
[18]A. Bakkar, V. Neubert, Electrodeposition and corrosion characterization of micro-and nano-crystalline aluminium from AlCl 3/1-ethyl-3-methylimidazolium chloride ionic liquid, Electrochim. Acta 103 (2013) 211-218. https://doi.org/10.1016/j.electacta.2013.03.198.
[19]H.M.A. Abood, N.L. Dawood, Morphology of electrodeposited aluminum metal from aluminum chloride-urea room temperature ionic liquid (RTIL) at variable parameters, Int. J. Res. Sci. 4 (2015) 753-760.
[20]C. Wang, A. Creuziger, G. Stafford, C.L. Hussey, Anodic Dissolution of Aluminum in the Aluminum Chloride-1-Ethyl-3-methylimidazolium Chloride Ionic Liquid, J. Electrochem. Soc. 163 (2016) H1186-H1194. https://doi.org/10.1149/2.1061614jes.
[21]G.L. Holleck, J. Giner, The Aluminum Electrode in AlCl3-Alkali-Halide Melts, J. Electrochem. Soc. 119 (1972) 1161-1166. https://doi.org/10.1149/1.2404433.
[22]C. Wang, C.L. Hussey, Aluminum Anodization in the LiAlBr 4 -NaAlCl 4 -KAlCl 4 Molten Salt, J. Electrochem. Soc. 162 (2015) H151-H156. https://doi.org/10.1149/2.0591503jes.
[23]B.S. Del Duca, Electrochemical Behavior of the Aluminum Electrode in Molten Salt Electrolytes, J. Electrochem. Soc. 118 (1971) 405-411. https://doi.org/10.1149/1.2408069.
[24]R.T. Carlin, Aluminum Anodization in a Basic Ambient Temperature Molten Salt, J. Electrochem. Soc. 136 (1989) 1409. https://doi.org/10.1149/1.2096931.
[25]A.P. Abbott, F. Qiu, H.M.A. Abood, M.R. Ali, K.S. Ryder, Double layer, diluent and anode effects upon the electrodeposition of aluminum from chloroaluminate based ionic liquids, Phys. Chem. Chem. Phys. 12 (2010) 1862-1872. https://doi.org/10.1039/B917351J.
[26]A.J. Bard, L.R. Faulkner, Electrochemical methods: Fundamentals and applications, 2nd ed., John Wiley & Sons, New York, 2001.
[27]V.M. Mecea, From quartz crystal microbalance to fundamental principles of mass measurements, Anal. Lett. 38 (2005) 753-767. https://doi.org/10.1081/AL-200056171.
[28]A. Bund, G. Schwitzgebel, Investigations on metal depositions and dissolutions with an improved EQCMB based on quartz crystal impedance measurements, Electrochimica Acta 45 (2000) 3703-3710. https://doi.org/10.1016/S0013-4686" target="_blank">https://doi.org/10.1016/S0013-4686(00)00463-1.
[29]A. Ispas, A. Bund, F. Endres, Application of the Electrochemical Quartz Crystal Microbalance for the Investigation of Metal Depositions from Ionic Liquids, ECS Transactions 16 (2009) 411-420. https://doi.org/10.1149/1.3159345.
[30]G. Sauerbrey, Verwendung von Schwingquarzen zur Wägung dünner Schichten und zur Mikrowägung, Z. Physik 155 (1959) 206-222. https://doi.org/10.1007/BF01337937.
[31]E.M. Moustafa, S. Zein El Abedin, A. Shkurankov, E. Zschippang, A.Y. Saad, A. Bund, F. Endres, Electrodeposition of Al in 1-Butyl-1-methylpyrrolidinium Bis (trifluoromethylsulfonyl) amide and 1-Ethyl-3-methylimidazolium Bis (trifluoromethylsulfonyl) amide Ionic Liquids, J. Phys. Chem. B 111 (2007) 4693-4704. https://doi.org/10.1021/jp0670687.
[32]K.K. Kanazawa, J.G. Gordon, Frequency of a quartz microbalance in contact with liquid, Anal. Chem. 57 (1985) 1770-1771. https://doi.org/10.1021/ac00285a062.
[33]K.K. Kanazawa, J.G. Gordon, The oscillation frequency of a quartz resonator in contact with liquid, Anal. Chim. Acta 175 (1985) 99-105. https://doi.org/10.1016/S0003-2670(00)82721-X.
[34]Y. Zheng, Y. Zheng, Q. Wang, Z. Wang, D. Tian, Density and viscosity of binary mixtures of 1-ethyl-3-methylimidazolium heptachlorodialuminate and tetrachloroaluminate ionic liquids, J. Chem. Eng. Data 62 (2017) 4006-4014. https://doi.org/10.1021/acs.jced.7b00702.
[35]S. Zhang, N. Sun, X. He, X. Lu, X. Zhang, Physical Properties of Ionic Liquids: Database and Evaluation, J. Phys. Chem. Ref. Data 35 (2006) 1475-1517. https://doi.org/10.1063/1.2204959.
[36]J. Tang, K. Azumi, Optimization of pulsed electrodeposition of aluminum from AlCl 3-1-ethyl-3-methylimidazolium chloride ionic liquid, Electrochim. Acta 56 (2011) 1130-1137. https://doi.org/10.1016/j.electacta.2010.10.056.
[37]Q.X. Liu, S.Z. El Abedin, F. Endres, Electrodeposition of nanocrystalline aluminum, J. Electrochem. Soc. 155 (2008) D357-D362. https://doi.org/10.1149/1.2884369.
[38]H.J.S. Sand, III. On the concentration at the electrodes in a solution, with special reference to the liberation of hydrogen by electrolysis of a mixture of copper sulphate and sulphuric acid, Philos. Mag. 1 (1901) 45-79. https://doi.org/10.1080/14786440109462590.
[39]C. Schreiner, S. Zugmann, R. Hartl, H.J. Gores, Fractional Walden rule for ionic liquids, J. Chem. Eng. Data 55 (2010) 1784-1788. https://doi.org/10.1021/je900878j.
[40]W. Sutherland, LXXV. A dynamical theory of diffusion for non-electrolytes and the molecular mass of albumin, Philos. Mag. 9 (1905) 781-785. https://doi.org/10.1080/14786440509463331.
[41]H.J.S. Sand, On the Concentration at the Electrodes in a Solution, with special reference to the Liberation of Hydrogen by Electrolysis of a Mixture of Copper Sulphate and Sulphuric Acid, Proc. Phys. Soc. 17 (1899) 496-534. https://doi.org/10.1088/1478-7814/1