Formation of Porous Gold Electrodeposits by Pulse Technique in AlCl₃–NaCl–KCl Molten Salt Containing AuCl

Abstract

This study investigated the electrodeposition behavior of Al–Au alloys in AlCl₃–NaCl–KCl molten salt containing AuCl at 443 K, and porous gold was formed by potential pulse electrolysis in the molten salt. The deposition/dissolution potential of Au in the molten salt was approximately 2.0 V vs. Al/Al(III). For the Al–Au alloy electrodeposits obtained by constant potential electrolysis, the Au concentration was less than 1 at% at potentials lower than 0.0 V. Then the Au concentration increased sharply to 84 at% at 0.0 V and increased gradually to 92 at% at 0.3 V and 98 at% at 0.9 V. Porous gold was formed by potential pulse electrolysis with repeated Al–Au alloys electrodeposition and Al dissolution. The specific surface area of the electrodeposits obtained by potential pulse electrolysis increased with decreasing frequency. The specific surface area of porous gold obtained in this experiment was up to 9.2 m² g⁻¹.

1. Introduction

A quartz crystal microbalance (QCM) is utilized as the highly sensitive mass sensor. Generally, quartz crystals are plate-shaped with thin gold films on both sides (Fig. 1a).

Figure 1.

Schematic illustration of QCM’s cross section (a) conventional QCM and (b) QCM device with porous gold thin film targeted by this research.

By forming porous gold on a smooth gold thin film and attaching media that adsorb specific gases to the surface (Fig. 1b), it can be utilized as a sensor to detect trace amounts of specific gases with high sensitivity. We attempted to form porous gold on a QCM element and develop a more sensitive sensor.

Physical vapor deposition is a possible method for forming porous gold. Novotny et al. reported that cauliflower-like structures of 2–3 µm can be obtained by thermal evaporation.¹ While this must be a useful method, it is costly in terms of creating a high-temperature vacuum environment. Chemical etching has been reported as an alternative method. Forty et al. reported the formation of porous gold by dissolving Ag–Au alloys in nitric acid.² This method involved several processes, including the formation of Ag–Au alloys film by vapor deposition and the dissolution of Ag in nitric acid. Electrochemical dealloying of Au alloy ingots obtained by heat treatment has been reported as another method³ for chemical etching, which also involves multiple processes and may not be suitable for the QCM element. Therefore, considering the formation of porous gold films for QCM, this study proposes a novel approach to form porous gold films by electrochemically repeating the formation of an alloy of a certain metal “M” and Au and the dissolution of the metal “M” in molten salt. Metal “M” should be a base metal with a large potential difference from gold; aluminum was selected for this study. This process is simple because it can be performed with a pulse technique. In addition, the molten salt is relatively inexpensive; thus, the process is considered economically advantageous.

Many aluminum alloys, such as Al–Li,⁴ Al–Cr,⁵ Al–Ni,⁶ Al–Ti,⁷ Al–Mn,⁸ Al–Pt,⁹ Al–Sn,¹⁰ Al–Ta,¹¹ and Al–Zr¹² have been reported to be formed by electrodeposition from low-temperature molten salt. Here, AlCl₃–NaCl–KCl molten salt containing AuCl at 443 K was utilized for the electrodeposition of Al–Au alloys. Firstly, the electrodeposition behavior of Al–Au alloys in molten salt was investigated. Finally, porous gold was formed by potential pulse electrolysis with repeated electrodeposition of Al–Au alloys and Al dissolution.

2. Experimental

Anhydrous gold(I) chloride (AuCl, Strem Chemicals, 97 %) was utilized as received. To prepare the electrolyte, sodium chloride (NaCl, Kanto chemical, 98.0 %) and potassium chloride (KCl, Wako Pure Chemical, 99.5 %) were dried in a vacuum oven at 393 K for at least 24 h in advance. Aluminum(III) chloride (AlCl₃, Kanto Chemical, 98.0 %) was mixed with the above-mentioned NaCl and KCl in a ratio of 61 mol% AlCl₃ −26 mol% NaCl −13 mol% KCl, and the mixture was melted at 443 K. Pre-electrolysis was performed to purify molten salt. In these experiments, 50 mL of molten salt containing approximately 0.1 g of AuCl was used. The molten salt was stirred at 300 rpm after adding AuCl to accelerate its dissolution. A glassy carbon plate (GC, Nilaco, GC-20SS) was utilized as the working electrode for the electrochemical cell. The GC surface area of 1 cm² was utilized for voltammogram measurements, and 2 cm² was utilized for constant potential electrolysis and potential pulse electrolysis. For the counter electrode, GC was utilized for voltammogram measurements and pure gold plates (Nilaco, 99.95 %, 4 cm²) for constant potential electrolysis and potential pulse electrolysis, respectively. Pure aluminum wire (Nilaco, 99.99 %, φ 0.80 mm) dipped into AlCl₃–NaCl–KCl molten salt filled in a Pyrex glass tube was employed as the reference electrode (R.E.). Electrical contact between the electrolyte and the R.E. was maintained by ceramics fiber inserted into a hole at the bottom of the tube. Voltammogram measurements, constant potential electrolysis, and potential pulse electrolysis were performed with a Hokuto–Denko HZ-PRO potentiostat (Hokuto Denko Co., Ltd.).

A portion of the molten salt was collected and weighed to determine the solubility of AuCl. The collected molten salt was dissolved in 0.1 mol dm⁻³ HCl, and the solution was analyzed by ICP-OES (ICAP PRO XP, Thermo Fisher Scientific Inc.). This experiment showed that the molten salt was saturated by AuCl, and the molten salt was used in the following electrochemical experiments. For voltammogram measurement, the potential range was set from −0.1 to 2.4 V vs. Al/Al(III) with a scan rate of 25 mV s⁻¹. For constant potential electrolysis, the potential range was from −0.2 to 1.8 V with an electric charge density of 40 C cm⁻². Potential pulse electrolysis was also conducted. Scanning electron microscopy (SEM, JSM-6010PLUS/LA, JEOL) and energy dispersive X-ray spectroscopy (EDS, JSX-3100R2, JEOL) were utilized to observe the surface morphology of the electrodeposits and determine the Au concentration in them. Furthermore, the Brunauer–Emmett–Teller (BET) method was applied to evaluate the surface porosity of the electrodeposits utilizing an automated surface area analyzer (automated surface area and pore size distribution analyzer, Nippon Bell Co., Ltd.). For optimal measurements, the surfaces of the electrodeposits were treated at 323 K for 1 h before the measurement, as in the previous paper.^13,14

3. Results

3.1 Solubility of AuCl in AlCl₃–NaCl–KCl molten salt

Initially, the dissolution of AuCl in AlCl₃–NaCl–KCl molten salt was investigated. Concentration change of Au ions with time after addition of AuCl into the molten salt is illustrated in Fig. 2. As illustrated in Fig. 2, the concentration reached saturation in 1 h. The solubility of AuCl in the molten salt was approximately 2.5 mmol dm⁻³ at 443 K. The molten salt saturated by AuCl was utilized in the following experiments.

Figure 2.

Concentration change of Au ions with time after addition of AuCl into AlCl₃–NaCl–KCl molten salt at 443 K. After adding AuCl, the molten salt was stirred at 300 rpm.

3.2 Voltammogram measurement

The voltammograms on GC electrode in AlCl₃–NaCl–KCl molten salt (blue line) and the molten salt containing AuCl (orange line) are presented in Fig. 3. These potential range was set from −0.1 to 2.4 V with a scan rate of 25 mV s⁻¹. An enlarged voltammogram was inserted into Fig. 3. In the molten salt without AuCl, when the potential shifted in the negative direction, the cathodic current due to Al electrodeposition increased from 0.0 V. When the potential was reversed, a peak anodic current due to the dissolution of the electrodeposited Al was observed from 0.0 V. When the potential was further shifted in the positive direction, an increase in the anodic current due to Cl₂ evolution was observed from approximately 2.3 V.

Figure 3.

Voltammograms on GC electrode in AlCl₃–NaCl–KCl molten salt (blue line) and molten salt containing AuCl (orange line). The enlarged figure was inserted at the potential between −0.1 and 1.1 V.

In the molten salt containing AuCl, as well as without AuCl, Al electrodeposition, dissolution, and Cl₂ evolution reaction occurred. In addition, a pair of anodic and cathodic currents was observed at approximately 2.0 V. This current corresponds to the electrodeposition and dissolution of Au. From the enlarged voltammogram in Fig. 3, a small cathodic current was observed at approximately 0.8 V, which gradually increased to 0.0 V as the potential shifted in the negative direction. When the potential was reversed in the positive direction, small anodic current peak was observed at 0.2 V. These currents were not observed in the voltammograms of the AlCl₃–NaCl–KCl molten salt without AuCl. Therefore, the currents may correspond to the electrodeposition and dissolution of the Al–Au alloy. To investigate the potential at which the Al–Au alloy electrodeposits were formed, the voltammograms were reversed at various potentials, and the results are presented in Figs. 4a–4f. In Fig. 4a, no reduction reaction and corresponding anodic reaction was observed, whereas in Fig. 4b, a small cathodic current was observed from 0.30 V, and an anodic peak appeared at approximately 0.40 V in the positive scan. In Figs. 4c and 4d, an anodic peak at approximately 0.35 V was observed, which was not observed in Fig. 4b. In Fig. 4e, a small cathodic current peak was observed below 0.10 V, and an anodic peak was observed at approximately 0.20 V when the potential was reversed. In Fig. 4f, a large cathodic current was observed from 0.0 V, and a large anodic current appeared when the potential was reversed. These can be understood as currents of Al electrodeposition and dissolution. Therefore, Al–Au alloys are expected to form in the range above 0 V. Also, peaks A₁/C₁, A₂/C₂, and A₃/C₃ in Fig. 4f are found to be the currents due to the couple reaction, respectively. In the Al–Au binary phase diagram, AlAu₄, Al₂Au₅, AlAu₂, AlAu, and Al₂Au are present as intermetallic compounds. The presence of AlAu₄ was confirmed by XRD analysis of the electrodeposits obtained by constant potential electrolysis. Two other alloy phases were expected to have formed based on voltammograms, but these have not been identified presently.

Figure 4.

Voltammograms on GC electrode in AlCl₃–NaCl–KCl molten salt containing AuCl with the sweep direction reversed at various potentials, (a) 0.30 V, (b) 0.25 V, (c) 0.20 V, (d) 0.15 V, (e) 0.00 V, and (f) −0.10 V.

3.3 Electrodeposition of Al–Au alloy by constant potential electrolysis

Based on voltammogram measurements, electrodeposition of Al–Au alloy was conducted on GC electrode in AlCl₃–NaCl–KCl–AuCl molten salt at the potential range from −0.2 to 1.8 V with an electric charge density of 40 C cm⁻². The relationship between the electrolytic potential and Au concentration in the electrodeposits, quantified by EDS, is illustrated in Fig. 5. In the figure, Au concentration in the electrodeposits was less than 1 at% at the potential lower than 0.0 V and increased drastically to 84 at% at 0.0 V. As the potential was more noble, the Au concentration increased gradually, reaching 89 at% at 0.3 V and 98 at% at 0.9 V. However, no significant concentration changes were observed at the more noble potentials. The color of the electrodeposit was light grey at lower than 0.0 V, dark brown from 0.0 to 0.3 V, and yellow above 0.6 V.

Figure 5.

Relationship between the electrolysis potential and concentration of Au in the electrodeposits obtained by constant potential electrolysis.

Surface SEM images of electrodeposits obtained by constant potential electrolysis at the potential range from −0.2 to 0.3 V are presented in Fig. 6. At −0.2 V and −0.1 V, the morphology of the electrodeposits was granular, while above 0.0 V, the morphology was spongy. Further observation revealed that the particle size of the electrodeposits was larger at −0.2 V than −0.1 V. Also, the more positive the electrolysis potential from 0.0 V, the smoother the surface morphology was formed.

Figure 6.

Surface SEM images of electrodeposits by constant potential electrolysis: (a) −0.2 V, (b) −0.1 V, (c) 0.0 V, (d) 0.1 V, (e) 0.2 V, and (f) 0.3 V.

3.4 Electrochemical formation of porous gold by potential pulse electrolysis

Potential pulse electrolysis was performed to form a porous gold layer, in which the Al–Au alloy electrodeposition and Al-only dissolution reactions were repeated. In this experiment, electrodeposits were formed by varying four parameters: the Al–Au alloy deposition potential (E_c), Al dissolution potential (E_a), E_c applying time (t_c), and E_a application time (t_a). Kigawa et al. reported the deposition of Al–Cr–Ni alloys in molten salt and the effect of pulse frequency on the surface morphology.¹⁵ Therefore, pulse frequency was controlled for change of surface morphology of porous gold in this study.

The surface SEM images of the electrodeposits obtained by potential pulse electrolysis are presented in Fig. 7. The electrolysis conditions with the Al dissolution potential (E_a) of 0.9 V and duty ratio (t_c/t_a) of 0.67 were unified to form all the electrodeposits; the deposition potential (E_c) and frequency (f = 1/T, T = t_c + t_a) of the Al–Au alloy was changed. Particularly noticeable between (a) and (d), when comparing the surface SEM images of the electrodeposits for the same E_c values, the one with the lower frequency had more pores. In comparison, the one with the higher frequency had a denser morphology because part of pores was occupied by Au small particles. The relationship between the pulse frequency and the specific surface area determined by BET method (m² g⁻¹) for each electrodeposition potential of the Al–Au alloy is illustrated in Fig. 8. In the figure, the specific surface area of the electrodeposits obtained by potential pulse electrolysis increased with decreasing frequency in every deposition potential.

Figure 7.

Surface SEM images of electrodeposits by potential pulse electrolysis with dissolution potential E_a = 0.9 V and duty ratio t_c/t_a = 0.67; (a) deposition potential E_c = 0.0 V, Frequency f = 0.1, (b) E_c = 0.1 V, f = 0.1, (c) E_c = 0.2 V, f = 0.1, (d) E_c = 0.0 V, f = 0.2, (e) E_c = 0.1 V, f = 0.2, and (f) E_c = 0.2 V, f = 0.2.

Figure 8.

Relationship between frequency of pulse wave and specific surface area of electrodeposits. Electrolysis condition is all common: a dissolution potential E_a = 0.9 V and duty ratio (t_c/t_a) = 0.67, and deposition potential of 0.0 V (grey circle), 0.1 V (blue circle), and 0.2 V (red circle).

4. Discussion

As illustrated in Fig. 5, Al is electrodeposited more noble potential than 0.0 V vs. Al/Al(III). This may be due to the presence of Au ions in the molten salt, which induces Al ions. The same phenomenon occurred in the electrodeposition of Al–Pt alloys previously reported by our group,⁹ and it is expected that alloys and intermetallic compounds are formed at potentials above 0.0 V.

As illustrated in Fig. 6, different surface morphologies were obtained depending on the applied potential. At −0.2 and −0.1 V, the electrodeposits had more than 99 at% of Al and a small amount of Au. This is considered to be the reason for the relatively dense granular morphology. At above 0.0 V, Al–Au alloys were electrodeposited, and the structure may be a solid solution of Al in the Au matrix or intermetallic compounds. Al was induced by the Au ions in the molten salt and electrodeposited. Since large potential difference exists between co-depositing Al and Au, continuous Al dissolution is expected to occur. Further electrodeposition of Al–Au alloys and partial dissolution of Al resulted in the Au spongy morphology. Electrodeposit at 0.3 V has higher Au concentration than others. Therefore, the dissolution of Al co-deposited with Au is also less than others, which is considered to have resulted in the formation of the flat surface.

Figures 7 and 8 demonstrate that the specific surface area of the electrodeposits obtained by potential pulse electrolysis decreased with increasing frequency. At higher frequencies, the applied potential was repeated shorter. This may have resulted in the formation of fine microstructures. In contrast, the applied potential was repeated longer when the frequency was low. This may be one of the reasons for the formation of electrodeposits with more porosity morphology.

Consequently, the surface area of the electrodeposited film may have decreased. Here, porous electrodeposited gold films with a specific surface area of 9.2 m² g⁻¹ were obtained under pulsed electrolytic conditions with the Al–Au alloy deposition potential of 0.0 V, Al dissolution potential of 0.9 V, frequency of 0.1, and duty ratio of 0.67. If the frequency is less than 0.1, porous gold with an even larger specific surface area can be obtained. However, if the Al–Au alloy to be electrodeposited in one pulse is thick, the Al on the deposit’s surface can be dissolved; however, Al in deeper areas may not be dissolved.

The porous gold obtained in this experiment with the specific surface area of maximum of 9.2 m² g⁻¹ is not inferior to that of other reports.^13,14 The method of forming porous gold used in this study was excellent because porous gold with such a specific surface area was achieved by substituting relatively inexpensive aluminum for silver, which is an expensive metal.

5. Conclusions

The electrodeposition behavior of Al–Au alloys in AlCl₃–NaCl–KCl molten salt containing AuCl at 443 K was investigated, and porous gold was formed by potential pulse electrolysis in the molten salt. The results are summarized as follows.

1. 1.    The deposition/dissolution potential of Au in this molten salt was approximately 2.0 V.
2. 2.    For the Al–Au alloy electrodeposits obtained by constant potential electrolysis, the Au concentration was less than 1 at% at potentials lower than 0.0 V. Then Au concentration increased sharply to 84 at% at 0.0 V. It increased gradually, reaching 92 at% at 0.3 V and 98 at% at 0.9 V.
3. 3.    The specific surface area of the electrodeposits obtained by potential pulse electrolysis decreased with increasing frequency. For example, the specific surface area was 9.2 m² g⁻¹ at a frequency of 0.1 and 1.7 m² g⁻¹ at a frequency of 0.2 for the Al–Au alloys electrodeposition potential of 0.0 V, the Al dissolution potential of 0.9 V, and duty ratio of 0.67.

Acknowledgments

This study was supported by JST SICORP Grant Number JPMJSC2108. We express our gratitude to all concerned parties. We also thank Dr. Tatsuhiro Shigyo of the Hokkaido Research Organization (HRO) for his assistance with specific surface area measurements.

CRediT Authorship Contribution Statement

Masaya Sugizaki: Investigation (Lead), Writing – original draft (Lead)

Hisayoshi Matsushima: Validation (Equal)

Mikito Ueda: Conceptualization (Lead), Supervision (Lead), Validation (Lead), Writing – review & editing (Lead)

Midori Kawamura: Conceptualization (Lead), Project administration (Lead)

Conflict of Interest

The authors declare no conflict of interest in the manuscript.

Funding

Japan Science and Technology Corporation: JPMJSC2108

Footnotes

The contents include the authors’ presentation #P31 at 2023 Joint Symposium on Molten Salts (MS12).

M. Sugizaki: ECSJ Student Member

H. Matsushima, M. Ueda, and M. Kawamura: ECSJ Active Members

References

• 1)  M. Novotny, P. Fitl, A. K. Sytchkova, J. Bulir, J. Lancok, P. Pokorny, D. Najdek, and J. Bocan, Open Phys., 7, 327 (2009).
• 2)  A. J. Forty and P. Durkin, Philos. Mag. A, 42, 295 (1980).
• 3)  Q. Zhang, X. Wang, Z. Qi, Y. Wang, and Z. Zhang, Electrochim. Acta, 54, 6190 (2009).
• 4)  R. A. Carpio and L. A. King, J. Electrochem. Soc., 128, 1510 (1981).
• 5)  T. P. Moffat, J. Electrochem. Soc., 141, L115 (1994).
• 6)  T. P. Moffat, J. Electrochem. Soc., 141, 3059 (1994).
• 7)  G. R. Stafford, J. Electrochem. Soc., 141, 945 (1994).
• 8)  J. Uchida, A. Shibuya, T. Tsuda, Y. Yamamoto, and H. Seto, J. Surf. Finishing Soc. Jpn., 42, 200 (1991).
• 9)  M. Ueda, H. Hayashi, and T. Ohtsuka, Surf. Coat. Technol., 205, 4401 (2011).
• 10)  M. Ueda, R. Inaba, and T. Ohtsuka, Electrochim. Acta, 100, 281 (2013).
• 11)  K. Sato, H. Matsushima, and M. Ueda, Appl. Surf. Sci., 388, 794 (2016).
• 12)  M. Ueda, T. Teshima, H. Matsushima, and T. Ohtsuka, J. Solid State Electrochem., 19, 3485 (2015).
• 13)  Y. H. Tan, J. A. Davis, K. Fujiwara, N. V. Ganesh, A. V. Demchenko, and K. J. Stine, J. Mater. Chem., 22, 6733 (2012).
• 14)  G. Pia, M. Mascia, and F. Delogu, Scr. Mater., 76, 57 (2014).
• 15)  M. Ueda, H. Kigawa, and T. Ohtsuka, Electrochim. Acta, 52, 2515 (2007).