Home Complementary methods for characterization of the corrosion products on the surface of Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 brazing alloys
Article Publicly Available

Complementary methods for characterization of the corrosion products on the surface of Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 brazing alloys

  • Stevan P. Dimitrijević , Borislava D. Vurdelja , Silvana B. Dimitrijević , Filip M. Veljković , Željko J. Kamberović and Suzana R. Veličković EMAIL logo
Published/Copyright: January 11, 2020
Download Article (PDF)
Corrosion Reviews
From the journal Corrosion Reviews Volume 38 Issue 2

Abstract

Corrosion products formed on the surface of two silver brazing alloys after the potentiostatic polarization in 3.5% sodium chloride solution were characterized by the standard methods such as, the X-ray diffraction, micro-Raman spectroscopy, and scanning electron microscopy with energy-dispersive spectroscopy. This paper presents the results of a laser desorption/ionization mass spectrometry (LDI MS) analysis as a new approach to the characterization of corrosion products. The potential of the anodic polarization was 0.5 V versus saturated calomel electrode, and the process duration was 300 s. The corrosion layers on both investigated alloys were similar in composition with cuprous chloride and silver chloride as the main components and had strong indications of cuprous oxide formation. The major difference between these two layers was the existence of zinc hydroxychloride as the corrosion product of Ag-Cu-Zn alloy. Palladium compounds were not found in the case of Ag-Cu-Pd alloy. The results of different methods have shown a good consistency. Complementarity between the used methods was useful in the interpretation of the results for each used method. This study has demonstrated that LDI MS can be used as an efficient additional method together with the traditional ones.

Keywords: brazing silver alloys; characterization; corrosion film; LDI; mass spectrometry

1 Introduction

It is well known that silver alloys are widely used in the areas of energy, information, and communication technology, chemical processing industries, construction, heat exchangers, power transmission lines and electrical contacts, coinage, ornamental parts, etc. (Haynes, 2014; Zugic et al., 2017; Xiong et al., 2019). Silver brazing fillers are a group of silver-based alloys that are used for joining most ferrous and nonferrous metals and ceramics to a metal contact (Ye et al., 2010; Pereira et al., 2015; Wang & Xue, 2016; Grund et al., 2018). Contemporary use of Ag brazing fillers and silver-based pastes for joints in emerging and innovative technologies like the sensors, new energy sources (fuel cells, solar photovoltaic modules, oxygen transport membranes, and similar) and electronics industry is evident (Hardy et al., 2007; Zemen et al., 2012; Zaharinie et al., 2014; Chao et al., 2015; Raju et al., 2016).

The development of cadmium-free silver brazing alloys has given the importance of the Ag-Cu-Zn system. It is used as a pure ternary alloy or as a base for multicomponent alloys with Sn, Ni, In, and Ga as additional alloying metals (Greener & Szurgot, 1982; Lai & Xue, 2010; Xue et al., 2019). Another interesting ternary system without cadmium is Ag-Cu-Pd. It is used a lot for hydrogen separation process, but then with a high palladium content (Pati et al., 2017). Alloys with low Pd concentration are used as brazing fillers (Ren et al., 2018). The ternary Au-Cu-Zn and Ag-Cu-Pd alloys are used in dentistry, as amalgams improvers or joint fillers for different dental materials. Corrosion of the metal alloys for dental application is а severe issue, even when they are of noble metals (Upadhyay et al., 2006). An electrochemical study of experimentally obtained Ag-Cu-Zn brazing alloys, with an equivalent composition to commercial dental alloys, has shown fair corrosion resistance among them, but only one of them had a high corrosion resistivity in both tested solutions (3.5% NaCl and Ringer’s solutions) (Ntasi et al., 2014).

The addition of Pd in an amalgam alloy leads to better corrosion resistance and improvement of mechanical properties (Greener & Szurgot, 1982). For example, the nine Ag-Pd-Cu alloys were tested on corrosion in several solutions, with good results especially in Ringer’s solution (Nakayama & Ando, 1999). However, sulfide solution was shown to be a strong corrosion media even with the addition of gold to the system, even in a corrosion study of Au-Ag-Cu-Pd dental alloys (Chiba & Kusayanagi, 2005). In the same investigation of corrosion products of these alloys in 0.9% NaCl solution, it was found as silver chloride and copper chloride.

Generally, corrosion product identification in chloride media on the surface of the alloy could be challenging, especially when anodic polarization is applied. Most chloride complexes are soluble, but some of them are not and can be a part of anodic film. In a study of anodic film formed on Ag-Cu-Zn alloy in 3.5% NaCl solution has found that besides cuprous oxide and chloride, it could contain different zinc oxychlorides (Dimitrijević et al., 2016).

The investigation used the common characterization methods: X-ray diffraction (XRD), Raman, and scanning electron microscopy (SEM) with energy-dispersive spectroscopy (EDS).

In earlier work, within our research group, a laser desorption/ionization method on a commercial matrix-assisted laser desorption ionization (MALDI) time of flight (TOF) mass spectrometric instrument was successfully employed to study the anodic films on Ag43Cu37Zn20 alloy surface (Vurdelja et al., 2017). Generally, MALDI-TOF spectrometry is in very common use in many fields of chemistry, biology, and medicine. The initial applications were almost completely for the qualitative analysis of bio-molecules (proteins, peptides, lipids, metabolites) direct from a large number of different samples and in particular tissue sections (Karas & Hillenkamp, 1988; Tanaka et al., 1988; Hidaka et al., 2007; Sauer, 2007). Due to the high speed of analysis, restricted possibility for cross-contamination between samples/users, ease of data interpretation, ease of use, and relative low equipment cost, the MALDI method now has grown into a valuable tool in many fields of inorganic chemistry, for example, the analysis of various nano-layers or of solid materials (Rashidzadeh & Guo, 1999; Pangavhane et al., 2010; Ramirez-Galicia et al., 2010; Mawale et al., 2019).

In the case of inorganic sample, there is often no need for a matrix to obtain good mass spectra; in this way, the problem of low-mass detection has been overcome. This type of analysis is named as laser desorption/ionization (Duncan et al., 2008).

In this work the chemical composition of the film anodically formed on Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys was evaluated by XRD, micro-Raman spectroscopy (MRS), and SEM/EDS. Earlier results have shown that laser desorption/ionization mass spectrometry (LDI MS) analysis was in good agreement with the results obtained using the Raman spectra and XRD analysis (Vurdelja et al., 2017). Hence, this paper discusses the strategy for implementation of the LDI MS method to analyze the corrosion films of these alloys after potentiostatic polarization in the presence of chloride ions.

2 Materials and methods

2.1 Materials and electrode preparation

Constituent metals Ag, Cu, Zn, and Pd with purity of 99.99% were used for the preparation of the alloys (Sigma-Aldrich, Steinheim, Germany). Samples for the electrodes were prepared by the ingot metallurgy method. The obtained ingots were subjected to homogenization annealing at 600°C for 24 h in a nitrogen atmosphere and slowly cooled to 350°C for the next 8 h. They were annealed for 8 more hours at 350°C and then slowly cooled to room temperature for 4 hours in the same protective atmosphere.

Compositions of samples corresponded to the standard brazing alloys, AG 202 and PD 105, respectively, according to EN 1044:1999 standard, Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10, and allowed tolerances in standards (e.g. standard SRPS EN ISO 17672:2012 or EN 1044:1999). Tolerances for EN AG 202 alloy are ±1% for Ag and Cu and ±2% for Zn, while they are ±0.5% for all three metals in EN PD 105 alloy.

The sodium chloride was of analytical grade produced by Merck (Darmstadt, Germany). The solution of 3.5% NaCl was prepared with double distilled water of conductivity less than 1 μS/cm.

2.2 Electrochemical procedure

Electrochemical experiments were conducted in a conventional three-electrode cell. The cylindrical electrodes of the alloys embedded in polytetrafluoroethylene, with an active surface of 0.40 cm2, were working electrodes. Platinum sheet (area of 2.0 cm2) served as a counterelectrode. All electrochemical measurements were performed using a Interface 1000 potentiostat/galvanostat/zero resistance ammeter (Gamry instruments, Philadelphia, PA, USA) and software package Gamry Framework (Version 6.25). For analysis of the electrochemical data, the Gamry Echem Analyst software package was used.

Potentiodynamic polarization measurements were made after stabilization of the open circuit potential (OCP), after 1 h immersion in the working solution, in the potential range between −1.0 and +1.0 V versus OCP and at a scan rate of 1 mV/s.

Anodic films on Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys were obtained after the potentiostatic polarization in 3.5% NaCl solution at 25±0.5°C. The applied potential was +500 mV versus saturated calomel electrode (SCE). Duration of polarization was 300 s.

2.3 Surface analyses

2.3.1 SEM/EDS examination

The microstructure and compositions of the metallurgical phases and corrosion layers were studied using a scanning electron microscope operated at 20 keV (model: JSM IT 300LV, JEOL, Tokyo, Japan) with an EDS detector (model: X-max, Oxford Instruments, Abington, UK; element detection: Z>5; 0.1 wt.%; 126 eV resolution). Samples after anodic polarization were prepared in an auto fine coater (model: JFC-1300, JEOL, Tokyo, Japan) with automatic control of coating thickness by a thickness controller (model: FC-TM20, JEOL, Tokyo, Japan).

2.3.2 Micro Raman spectroscopy (MRS)

The micro-Raman spectra were taken in the backscattering configuration and analyzed using a T64000 spectrometer (Horiba Jobin Yvon, Lille, France), equipped with a nitrogen cooled charge-coupled device detector. The 532 nm line of a Ti:sapphire laser was used as an excitation source. The measurements were performed at 20 mW laser power.

2.3.3 X-ray diffraction (XRD)

The XRD patterns of investigated samples were recorded on an X-ray diffractometer (Model Explorer, GNR Analytical Instruments Group, Novara, Italy) using Cu Kα1/2 radiation and wavelength of 0.154 nm and operated at 40 kV and 30 mA. For the X-ray analysis, the software PDXL 2 Version 2.4.2.0 was used. The resulting diffractograms were compared with the data from the database ICDD (PDF-2 Release 2015 RDB).

2.3.4 Laser desorption/ionization mass spectrometry

The LDI MS measurements were carried out on a commercial MALDI Voyager-DE PRO mass spectrometer from Sciex (Foster City, CA, USA) using an ultraviolet nitrogen laser (wavelength of 337 nm, pulse width of 3 ns, and repetition rates of 20.00 Hz) and TOF analyzer. In these experiments, the acceleration voltage was 20–25 kV, with 94% grid voltage and an extraction delay time of 150 ns. The sample was mechanically removed from the electrode. For the LDI analysis, a small volume of the sample, mixed with deionized water, is applied directly to the sample plate of stainless steel. An analyte of 0.5 μl was dried at room temperature and pressure. Within the mass spectrometer, the laser is pulsed (200 laser pulses per spectrum) onto the sample surface, where simultaneously, desorbing and ionizing of the analyte molecules have occurred. A typical LDI mass spectrum is a plot of intensities, i.e. the relative abundance of ions on the ordinate axis and the corresponding mass/charge values on the x-axis. The isotopic pattern was determined from the ratios of signal-to-noise values of the individual signal in a group of the most intense signals of detected compounds.

3 Results and discussion

In the first part, the metallurgical phase composition of the Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys before anodic polarization was examined using thermodynamic calculations, SEM/EDS, and XRD. Earlier research has shown that the ternary phase diagrams of Ag-Cu-Zn and Ag-Cu-Pd systems show that these alloys are two-phase systems. The alloy with zinc consists of Ag-rich phase (Ag) and Cu-rich phase (Cu), both with fcc structure. These phases are virtually two components (binary alloys) where the (Ag) phase contains mainly Ag-Zn and the (Cu) phase contains mainly Cu-Zn. Identically, the alloy with palladium consists of two fcc phases, Ag and Cu-Pd (Chang et al., 1977). Thermodynamic calculations in Pandat™ software for Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10, as presented in Table 1, confirm that at equilibrium at low temperature (350°C), only traces of the third metal from the alloy is present in the “binary” phase, Cu and Ag, respectively. Nevertheless, it should be noted that even as stated, ordered fcc (L12) CuPd phase was not shown in their ternary diagram. Only the newer literature (Subramanian & Laughlin, 1991) has shown the stability zones for ordered CuPd structures.

Table 1:

Calculated metallurgical phase compositions for the investigated alloys in wt.% (equilibrium at 350°C).

Alloy, designation
 Ag60Cu26Zn14 (EN AG 202)
 Ag58.5Cu31.5Pd10 (EN PD 105)
Phase (Ag) (Cu) (Ag, Pd) (Cu, Pd)
Fraction of the phase 0.546 0.454 0.508 0.492
Silver, Ag 90.93 1.181 95.54 0.223
Copper, Cu 0.718 74.08 0.364 80.28
Zinc, Zn 8.350 24.74 – –
Palladium, Pd – – 4.099 19.49

SEM observations for these samples, made before the electrochemical experiments, are presented in Figure 1, while the EDS analysis results are summarized in Table 2.

Figure 1: SEM images of the silver brazing alloys before electrochemical experiments: Ag60Cu26Zn14 (A); Ag58.5Cu31.5Pd10 (B).
Figure 1:

SEM images of the silver brazing alloys before electrochemical experiments: Ag60Cu26Zn14 (A); Ag58.5Cu31.5Pd10 (B).

Table 2:

Metallurgical phase compositions for the investigated alloys in wt.% determined by the EDS.

Alloy, designation
 Ag60Cu26Zn14 (EN AG 202)
 Ag58.5Cu31.5Pd10 (EN PD 105)
Phase (Ag) (Cu) (Ag, Pd) (Cu, Pd)
Silver, Ag 83.86 6.57 92.60 5.44
Copper, Cu 5.93 70.35 2.57 73.03
Zinc, Zn 10.71 23.08 – –
Palladium, Pd – – 4.83 21.53

Figure 1A and B show the microstructures of Ag60Cu26Zn14 (sample 1) and Ag58.5Cu31.5Pd10 (sample 2), respectively. In sample 1, two-phase (Ag)+(Cu) microstructures were identified, and in sample 2, two-phase (Ag)+(Cu, Pd) microstructures were found.

It is known that in a real system, despite long annealing and slow cooling, the rate of diffusion in solid solutions is low and equilibrium is much closer to the temperature between temperatures of annealing and solidus. Consequently, the content of the third metal in the each of the four presented metallurgical phases is much higher (Dimitrijević et al., 2018). Here, the EDS analysis also confirms this. By comparing the results in Tables 1 and 2, the higher contents of Ag in the (Cu) phase can be noticed, and vice versa, as well as Ag in (Cu, Pd) and Cu in (Ag) than in the ideal cases, calculated in Table 1.

The obtained X-ray diffractograms for sample 1 and sample 2 are depicted in Figure 2.

Figure 2: XRD pattern of the silver brazing alloys before electrochemical experiments: Ag60Cu26Zn14 (A); Ag58.5Cu31.5Pd10 (B).
Figure 2:

XRD pattern of the silver brazing alloys before electrochemical experiments: Ag60Cu26Zn14 (A); Ag58.5Cu31.5Pd10 (B).

Figure 2A shows the two-phase structure of Ag-Cu-Zn alloy; therefore, it can be said that the results of the XRD analysis support the results of the thermodynamic calculations quite well. In the diffractogram for the Ag60Cu26Zn14 alloy (Figure 2A), one phase is a solid solution of silver (labeled as Ag) and the other is a Cu solid solution (labeled as Cu0.7Zn0.3). The silver phase has an almost perfect match with the pure silver. Small differences in peak positions and intensities in the diffraction patterns relative to the identified Cu0.7Zn0.3 phase indicate that the identified (Cu) phase beside zinc contains some amount of dissolved silver, as can be seen in Table 2.

The diffractogram of the Ag58.5Cu31.5Pd10 alloy in Figure 2B shows that it consists of two disordered fcc phases and one intermetallic. Similar to the previous consideration, negligible deviations from the ideal pattern are due to the small Cu content in the Ag phase (black circle). This suggests that the silver phase is a silver solid solution of palladium and copper in silver. The software has detected double peaks in diffractogram (Cu – white circle, Cu3Pd – black rhomboid) as two phases that are shown in the insertion. Peaks are very close to each other, indicating the similar composition of the two phases with similar crystallographic features, although they have different structures.

The strong tendency of palladium to form the ordered Cu3Pd superlattice in the copper-rich phase of the ternary Ag-Cu-Pd alloys can cause such behavior. More precisely, superlattice (that should be the ordered Cu3Pd I phase) was found to be characteristic for Ag-Cu-Pd with a Pd concentration from 10 to 25 wt.% with a similar (57–62 wt.%) silver content (Chen et al., 1996). Additionally, the Cu-Pd ratio in the studied alloy was where the intermetallic phase Cu3Pd with the AuCu3-type (L12) (~8–22 at.% Pd in Cu) is stable and not a one-dimensional long-period superstructure (~17–28 at.% Pd) (Kee-Baek et al., 2019). This explanation is in good correlation with detected ~73 wt.% of Cu and ~21 wt.%, in the CuPd reach phase, by the EDS analysis (Table 2) and XRD analysis.

The results of the XRD and SEM/EDS analysis are in good agreement with that of the Ag-Cu-Zn alloy. The Cu3Pd phase was not found by the SEM/EDS method, similar to the published study (Chen et al., 1996), because of the small dimensions of the ordered phase particles, where only the transmission electron microscopy with selected area diffraction was suitable for detection.

The results obtained in this work are in conformity with earlier literature data (Chang et al., 1977) for Ag60Cu26Zn14. Also, our results for Ag58.5Cu31.5Pd10 are consistent with the newer literature for ternary alloys (Subramanian & Laughlin, 1991) and the newest published redetermined binary Cu-Pd phase diagram (Kee-Baek et al., 2019).

Potentiodynamic polarization is a useful method to obtain initial data of the electrochemical system. Here, the main purpose of the method was to confirm thermodynamic considerations about the choice of the potential for anodic polarization which produces corrosion films at the surface of the alloys. Therefore, the polarization curves for sample 1 and sample 2 are shown in Figure 3.

Figure 3: Polarization curves for Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys in a 3.5% NaCl solution at 25°C.
Figure 3:

Polarization curves for Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys in a 3.5% NaCl solution at 25°C.

The Ag-Cu-Pd alloy had more positive OCP compared to Ag-Cu-Zn, −5.3 mV versus SCE and −195.5 mV consequently. This indicates a higher corrosion resistance of the alloy with palladium. The OCPs were measured during 3600 s. A little more negative corrosion potentials were obtained during the dynamic polarization, in both cases for about 8 mV.

One peak has been found for sample 1, and two clear peaks have been observed in the polarization curve for sample 2. The first one is at +10 mV (±5 mV), and a more positive value is for the alloy with palladium. The first peak can be attributed to the multiphase reaction: Cu+Cl−→CuCl+e−, and it is confirmation of the literature data that a CuCl surface layer formed near the observed potential leads to a decrease in the current density at polarization measurement (Lee & Nobe, 1986; Vvedenskii & Grushevskaya, 2003). The second peak corresponds to the reaction Ag+Cl−→AgCl+e− and the decrease in the current density in the polarization curve due to the formation of silver chloride layer at the surface of the silver/silver phase (Hassan et al., 2010). At the potential of the second peak, the maximum of current density is observed for sample 1 (not fully developed peak). However, in a complex system where ternary alloys are oxidized under polarization, simultaneous parallel reactions should be considered. In the peak areas of potential, consequently, the above reactions are just prevalent and not the only one. The porosity of the corrosion films has stressed this kind of behavior even more, and this is explained in the “Characterization of the corrosion layers on the surface of alloys” section.

At high potentials, the polarization curve for an alloy containing zinc shows a nearly constant value. The same is with alloy with palladium, but with a broad peak at +0.52 V. This peak should not be attributed to a concrete reaction, but to the coverage of alloy surface with a corrosion layer. Values of the current densities for the pseudo-steady state are about 1 mA/cm2 (approx. half of that for the Ag-Cu-Pd) and cannot be considered as passivation states. The porosity of the corrosion films is probably responsible for the shape of the polarization curves at potentials higher than the second peak and the fact that the board peak at about +0.5 V was observed only for sample 2.

Potentiostatic polarization is performed with the aim to obtain a corrosion film at the surface of the electrodes. The change in current versus time for the Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 electrodes, at 0.50 V, in 3.5% NaCl solution, is presented in Figure 4. A potential of +500 mV was chosen due to a higher silver content (about 60 wt.% against 43 wt.%) than in a previous study (Dimitrijević et al., 2016), additionally because it is close to but higher than the potential of palladium oxidation in chloride solution. The standard electrode potential for a half reaction for Pd to Pd(II) due to the reaction [PdCl4]2−+2e−→Pd+4Cl− is +0.621 V versus the standard hydrogen electrode (+0.380 V vs. SCE) (Hassan et al., 2010). However, the palladium spice as a soluble anodic oxidation potential was planned to affect all the metals in both alloys. Additionally, the potential of +0.5 V was chosen because it was the last (broad) peak in the polarization curve.

Figure 4: Chronoamperometric current-time curves obtained in aerated 3.5 wt.% NaCl solutions at E=0.50 V and at 25°C for Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10.
Figure 4:

Chronoamperometric current-time curves obtained in aerated 3.5 wt.% NaCl solutions at E=0.50 V and at 25°C for Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10.

Examination of the potentiostatic current transients, in Figure 4, reveals that the transient current density decreases monotonically with time reaching nearly (quasi) steady-state current values. The values of instantaneous and steady-state current densities are higher for Ag-Cu-Zn alloy. A total of five measurements were performed, and the results with a lower current maximum for Ag-Cu-Zn alloy and higher for Ag-Cu-Pd alloy are shown in Figure 4, for clarity reasons. These current densities for sample 1 and sample 2 have been varied for lower than 5% of the mean values. The instantaneous current for the alloy containing zinc is about two and a half times higher than the palladium-based alloy, and the steady-state current is nearly double.

For sample 1 and sample 2, it was found that the continuously falling current was linear versus t−1/2 (Figure 5). These deviations from the Cottrell equation are probably due to the coupled chemical reactions. The distinctively nonlinear parts in the diagrams are most probably a consequence of the simultaneously electrochemical reactions. Linear parts, on the other hand, are possibly intervals where one reaction is prevalent. There are three such parts for Ag-Cu-Zn alloy and two for the Ag-Cu-Pd. Only linear parts at the end of experiments (the shortest in diagrams, but with the longest duration) pass through the origin, and all others have and interception at the ordinate. Processes in these intervals (from about 120 s until the end of the experiment) are fully diffusion controlled. This is characteristic of the formation of CuCl and AgCl films on the surface of electrode (Vvedenskii & Grushevskaya, 2003; Hassan et al., 2010). The existence of current density intercepts at the ordinates points to the mixed-control process (kinetic control contribution).

Figure 5: Dependence of the current density on t−1/2 for descending portions of the current transients recorded in 3.5% NaCl solution at E=0.50 V for Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B).
Figure 5:

Dependence of the current density on t−1/2 for descending portions of the current

transients recorded in 3.5% NaCl solution at E=0.50 V for Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B).

3.1 Characterization of the corrosion layers on the surface of alloys

In the second part, to study in detail the composition of corrosion films after anodic polarization, we used XRD, MRS, SEM/EDS, and LDI MS methods. Figure 6 shows the diffractograms of the Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 samples after the potentiostatic treatment.

Figure 6: XRD patterns of the anodic films after anodic polarization on Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B).
Figure 6:

XRD patterns of the anodic films after anodic polarization on Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B).

It can be seen in Figure 6A that the diffractogram of the Ag60Cu26Zn14 sample contains the most prominent peaks, with a full pattern from Ag and CuCl. The peaks with small intensity, besides those that belong to the CuCl pattern, are from silver chloride and possibly cuprous oxide. Previous research has shown that the formation of a porous Cu2O oxide layer occurs with the following reaction (Sherif, 2012):

(1) 2CuCl ads +H 2 O = Cu 2 O+2H + +2Cl −

However, in neutral chloride media, the formation of Cu2O layer is not a fast process (Vvedenskii & Grushevskaya, 2003; Sherif, 2012). In our conditions, the time of the experiment (300 s) was not long enough for significant occurrence of this reaction (Sherif, 2012); hence, the Cu2O oxide film was barely formed. The main reaction on the surface of the (Cu) metallurgical phase is the formation of cuprous chloride by the multiphase reaction with reversible steps and can be simplified to (Lee & Nobe, 1986):

(2) Cu = Cu + +e − ,  followed with Cu + +Cl − = CuCl

It has also been noted that in the same phase (Cu-Zn) after the initial dealloying, where zinc is dissolved, zinc hydroxychlorides are also formed as part of the layer (Figure 6A).

It is obvious that CuCl is the dominant species at the (Cu) phase and that the (Ag) phase of the alloy is subjected to the strong dealloying process (mostly dezincification) with only a thin film of AgCl on the surface. This AgCl film is probably porous with poor coverage of the surface. Silver chloride film formation cannot be expected to a great extent in the short period, like in the test. It can be significantly formed only in a long time period, when silver will not be protected anymore by zinc or copper.

It is noteworthy that silver solutions (Ag-Zn) have a similar pattern with pure Ag; its peaks appear near the peaks of Ag, as also displayed in Figure 6A.

In the diffractogram of the Ag58.5Cu31.5Pd10 sample (Figure 6B) dominant peaks and full XRD patterns of AgCl and CuCl are clearly visible. The main difference with the diffractogram in Figure 6A shows the more dominant presence of AgCl. It can be concluded that the main reactions are the formation of a CuCl layer on the surface of the Cu-Pd phase and the formation of a AgCl film on the surface of the Ag phase. According to the theory, the copper oxidation is even easier when copper is together with palladium (in the phase) since it has much lower electrode potential. The same is for silver; the reaction Ag=Ag++e− is easier when palladium is in the phase, and not zinc since silver, acting as an anode in a metallurgical silver phase (Ag-Pd). The less pronounced cuprous chloride pattern may be due to the fact that in the Cu-Pd phase, copper can be more easily dissolved (oxidized to cupric ion) than in the Cu-Zn phase. In the Cu-Pd system, similar to the case of the Ag-Pd phase for silver, copper acts as an anode.

The presence of Pd, together with the weak XRD pattern of the Cu3Pd after anodic polarization (Figure 6B), indicates the dealloying of the Cu3Pd phase. This is not unexpected since the dealloying of Cu3Pd in sulfate solutions is already reported in the literature (Ankah et al., 2013).

It should be noted that the silver pattern is still present, but much less pronounced than in the case of the Ag60Cu26Zn14 sample. This indicates that the AgCl layer is thin and porous, with relatively low protection ability. This is similar to CuCl, which shows a lower effect since the Pd pattern is barely visible (peaks with low intensity). These palladium peaks are probably mostly from the dealloyed Cu3Pd phase since there is a low concentration of it in the silver phase (Ag-Pd-Cu).

The Raman spectra of the investigated alloys recorded after anodic potentiostatic polarization treatment is shown in Figure 7. The main Raman peaks are marked in these spectrums.

Figure 7: Raman spectra of the anodic films at room temperature after anodic polarization on Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B).
Figure 7:

Raman spectra of the anodic films at room temperature after anodic polarization on Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B).

Figure 7A shows the results for the Ag60Cu26Zn14 alloy. The observed Ag peaks (black solid circles) at 1188, 1365, and 1568 cm−1 are in general agreement with the previously published results (Liang et al., 2009). Peaks at 1500 and 1536 could also be attributed to the silver or colloidal silver on the surface.

Very strong silver peaks indicate the formation of Ag areas, segregate from the (Cu) phase. These areas suggest the process of dealloying of the (Ag) phase or dissolution of zinc (and in less extent copper) from it.

Cuprous chloride peaks (white circle with cross) are observed at 113, 148, 204, 614, and 1113 cm−1. This is in good agreement with literature data (Frsot, 2003; Zhang et al., 2014). The first two of the peaks also originated from silver chloride, and the peak at 614 cm−1 is mutual/overlaps with Cu2O.

Silver chloride peaks (open down triangle) are found at 113, 148, and 239 cm−1. The very strong peak at 148 cm−1 is mutual/overlaps with CuCl, as well as a weak one at 113 cm−1, and makes identification of AgCl harder. Nevertheless, the pronounced peak at 239 cm−1 confirms the presence of AgCl. These peaks were compared with those reported in the literature (Von der Osten, 1974) with general compliance. Results are in agreement with the XRD analysis.

In the Raman spectra of treated alloy, several Cu2O modes (white rhombuses) are seen. The modes that can be attributed to cuprous oxide are at 518, 614, 645, and 1594 cm−1 (Compaan & Cummins, 1973; Ma et al., 2015; Hosseinpour & Johnson, 2017). This strong indication of the presence of Cu2O is not fully supported by the XRD results. In the diffractogram, only a hint of the existence of copper(I) oxide can be seen.

The peaks of β-Zn(OH)Cl are marked in Figure 7A as white squares. Two main peaks are observed at 269 cm−1, assigned to a Zn-Cl bond, and 394 cm−1, attributed to a Zn-O vibration characteristic of this structure. Zinc hydroxychloride is also characterized by two main OH fundamental stretching bands at 925 and 1051 cm−1 (Bernard et al., 1993; Lutz et al., 1993). However, peaks are not pronounced, and the presence of the zinc hydroxychloride is not so strongly supported like with the XRD method.

In the case of Ag60Cu26Zn14 alloy, results display the presence of silver (Ag), cuprous chloride (CuCl), silver chloride (AgCl), zinc hydroxychloride (β-Zn(OH)Cl), and cuprous oxide (Cu2O).

The Raman spectrum for the Ag58.5Cu31.5Pd10 alloy is presented in Figure 7B. Strong silver peaks (black solid circles in the figure) indicate the formation of Ag in the Ag (Ag-Pd) phase areas. The observed Ag peaks are very strong at 1568 and 1602 cm−1, and weaker peaks at 886, 977, 1071, 1260, 1340, 1405, 1441, and 1740 cm−1 complement the pattern and are in agreement with the previous consideration.

Cuprous chloride (white circle with cross) has a prominent peak at 149 cm−1 and a weaker ones at 613, 1071, and 1229 cm−1. The lack of a few peaks and lower intensity indicate a porous surface film and the possibility of copper dissolution due to oxidation to Cu(II), which palladium could promote. This behavior is in full agreement with the results from the XRD analysis.

Modes at 518 and 613 cm−1 are from the cuprous oxide (white rhombuses), although the latter can originate from cuprous chloride also. It is the same as for the previous alloy, except for the lower intensity of the peaks, which is completely confirmed by the XRD analysis. However, a big difference with the XRD is the peak at 1594 cm−1, which strongly suggests the presence of Cu2O. This is more stressed than with AgCuZn alloy, which was not expected.

Silver chloride (open down triangle) has a strong peak in the spectrum at 239 cm−1, which undoubtedly confirms its presence. Weaker peaks at 886 and 977 cm−1 complement the AgCl characterization. Additionally, a peak at 149 cm−1 can also be related to AgCl. This is in full agreement with the XRD analysis. Unprotected (cathodically in the metallurgy phase) silver is more prone to oxidation. In the chloride solution, AgCl formation is easiest (occurring at the lowest electrode potential). An interesting possibility is that the peak at 1071 cm−1 could be from Ag2O since it can be assigned to Ag-O stretching/bending modes (Wang et al., 1999). Nevertheless, the presence of Ag2O was not confirmed by the XRD, and it is unlikely due to the applied electrical potential in the chronoamperometry test. Due to these reasons, silver(I) oxide is not marked in Figure 7B. It has been noted that the presence of silver (Ag), cuprous chloride (CuCl), silver chloride (AgCl), and cuprous oxide (Cu2O) is registered for the Ag58.5Cu31.5Pd10 sample.

The SEM/EDS micrographs of the corrosion films at Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 after anodic polarization are shown in Figure 8.

Figure 8: SEM images of the anodic films after anodic polarization on Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B).
Figure 8:

SEM images of the anodic films after anodic polarization on Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B).

In the case of the Ag60Cu26Zn14 sample (Figure 8A), it can be observed that the surface is covered by a porous layer. The EDS spectra indicate that it is formed mostly by CuCl since copper and chlorine are dominantly present in them. The ratio between these two elements is about 3:2 in average. This excess of chloride and almost constant 10% of zinc support that zinc is present in the form of zinc hydroxychloride that is incorporated in the main CuCl film. It fits in the results of the Raman and especially XRD analysis. The oxygen content between 22% and 28% supports that conclusion. The oxygen concentration is not highly precise when determined by the EDS, even in a high vacuum, so the excess of it is not of great importance, especially when the images do not support other compounds. Eventually, formed cuprous oxide cannot be confirmed by the EDS analysis, but it is possible that traces of it could be in the thin layer under CuCl (multilayer at the surface). Another explanation of high oxygen content may be that CuCl is in а hydrate form, at least some part of it.

On the contrary to the findings in the layer, in the holes of it, oxygen was not determined, the concentration of the zinc was lower, and silver was present. It is probable that those are areas where the Ag-rich phase is present on the surface of the alloy where the silver chloride is formed, but not completely since areas of almost pure silver (dealloyed (Ag) metallurgic phase) are also present. This is highly supported by the XRD and Raman analysis and is just an indication here.

In Figure 8B, the Ag58.5Cu31.5Pd10 sample surface is almost fully covered with AgCl. This is in full agreement with high-intensity peaks of silver chloride in diffractogram and Raman spectra. White crystals are pure AgCl crystals, but the dark area is a mix of AgCl and silver since the Ag/Cl ratio is much higher than stoichiometric. This ratio is about 5 and almost constant. Silver is always about 80 wt.% and chloride content is between 15 and 18 wt.%. Oxygen in those areas is zero or very low (max. 0.35%), palladium is not detected at all, and copper is between 2.37 and 3.35 wt.%. Copper is presumably in the form of CuCl, which stresses chlorine deficiency for the pure AgCl as the only component of the main surface layer. The defects in the surface corrosion film revealed Cu-Pd phase since both of the metals are detected. The oxygen content of 12.2 and 15.7 wt.% suggests the presence of cuprous oxide. Although SEM/EDS analysis is not fully in agreement with those of XRD and Raman, results are fairly consistent for both samples.

The results of the LDI MS analysis for the corrosion films at Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys after anodic polarization treatment are presented in Figure 9.

Figure 9: The full LDI mass spectra in the positive mode of the anodic corrosion film of Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B) alloys; in the inset are the isotopic envelope for Zn2Cl2O5(OH)3(H2O)+ (A) and Pd+(B) ions.
Figure 9:

The full LDI mass spectra in the positive mode of the anodic corrosion film of Ag60Cu26Zn14 (A) and Ag58.5Cu31.5Pd10 (B) alloys; in the inset are the isotopic envelope for Zn2Cl2O5(OH)3(H2O)+ (A) and Pd+(B) ions.

The positive mode LDI mass spectrum for the Ag60Cu26Zn14 (Figure 9A) sample contains five major signals. The other ions detectable in the positive ion LDI TOF mass spectra of this sample are of much lower intensity. The stoichiometry of all ions was determined by comparison of the theoretical and experimental isotopic patterns. Table 3 presents the experimental mean values of the relative abundances of the individual isotopic combinations in the group of signals arising from the major signals detectable in the positive ion LDI mass spectra of Ag60Cu26Zn14 alloy with the standard deviations, their corresponding theoretical values, and accuracy of experimental results compared with these theoretical values.

Table 3:

Experimental and theoretical values of the relative abundances the individual isotopic combinations in the group of signals arising from the major signals detectable in the positive ion LDI mass spectra of Ag60Cu26Zn14 alloy, as well as the accuracy of experimental results compared with these theoretical values.

Ions Relative abundance of isotopes by S/N in LDI spectra Theoretical values of the relative abundance of isotopes Accuracy of experimental results compared with these theoretical values (%)
Cu(H2O)+ 69.61±1.38 69.15 0.66
30.38±1.38 30.84 1.49
Ag+ 51.45±0.25 51.84 0.75
48.55±0.25 48.16 0.81
AgCl-Na+ 39.16±0.40 39.15 0.02
49.54±0.37 49.07 0.96
11.29±0.01 11.77 0.81
Ag2Cl+ 20.85±0.74 20.29 2.76
44.43±0.36 44.30 0.29
29.39±0.52 29.73 1.14
4.77±0.46 5.67 15.87
(CuCl)3Cu(OH)2+ 9.81±0.42 9.84 0.30
26.77±1.60 27.12 1.29
31.55±1.36 31.34 0.67
20.42±0.89 20.82 1.92
5.86±0.24 8.14 30.2
5.77±0.35 1.88 67.42

A detailed analysis of the LDI mass spectrum with the results in Table 3 found that the major signals originates from Cu(H2O)+ (80.93, 82.93 calcd 80.94, 82.94), Ag+ (106.89, 108.88 calcd 106.91, 108.90), AgCl-Na+ (164.80, 166.80, 168.78 calcd 164.86, 166.86, 168.86), Ag2Cl+ (248.70, 250.72, 252.70, 254.67 calcd 248.78, 250.78, 252.78, 254.78), and (CuCl)3Cu(OH)2+ (390.40, 392.41, 394.40, 396.38, 398.35, 401.12 calcd 390.63, 392.63, 394.63, 396.63, 398.63, 400.63).

The relative abundances of the individual isotopic combinations in the group of signals arising from the minor signals are not presented here. However, the isotopic patterns for these ions can be clearly seen, such as the isotopic patterns of Zn2Cl2O5(OH)3(H2O)+, which are presented in the inset of Figure 9A. It should be highlighted that the relative abundances of the individual isotopic combinations of 27.00±0.76 (calcd 27.08), 45.40±0.54 (calcd 46.00), 23.31±0.36 (calcd 23.27), and 4.29±0.94 (calcd 3.63) for the minor ion at m/z 204.76, 206.76, 208.76, and 210.80, respectively, unambiguously show the presence AgCuCl+ ions.

Hence, the minor peaks originates from Cu+ (m/z 62.83, 66.80 calcd 62.96, 64.96), CuO(H2O)+ (m/z 96.90, 98.88 calcd 96.94, 98.93), Cu2OH(H2O)2+ (m/z 178.77, 180.78, 182.78 calcd 178.88, 180.88, 182.88), AgCuCl+ (m/z 204.76, 206.76, 208.76, 210.80 calcd 204.80, 206.80, 208.80, 210.79), Ag2+ (m/z 213.63, 215.63, 217.02 calcd 213.81, 215.81, 217.81), AgCuCl(H2O)+ (m/z 222.71, 224.71, 226.71 calcd 222.81, 224.81, 226.81), Ag2Cl2-Na+ (m/z 306.61, 308.66, 310.66, 312.65 calcd 306.74, 308.74, 310.74, 312.65), Ag3+ (m/z 320.66, 322.60, 324.61, 326.57 calcd 320.71, 322.71, 324.71, 326.71), and Zn2Cl2O5(OH)3(H2O)+ (m/z 346.50, 348.46, 350.46, 352.46 calcd 345.78, 347.78, 349.78, 351.77).

The positive mode LDI mass spectrum of the Ag58.5Cu31.5Pd10 sample is presented in Figure 9B. The results for both samples show a high degree of similarity. It should be observed that in addition to the listed species, the mass spectrum of the anodic corrosion film of Ag58.5Cu31.5Pd10 alloys also contains Pd+ ions with low intensity (their isotopic patterns presented in the inset Figure 9B.

The main difference in Figure 9A and B is that the signal of Ag+ ions, which have an intensity of 100% in the LDI MS of Ag60Cu26Zn14 and about 50% for Ag58.5Cu31.5Pd10 alloys. It may be an indication of higher corrosion resistance of this alloy than the Ag-Cu-Zn alloy.

It is obvious from the LDI MS results that the experimental values of m/z are in very good agreement with their corresponding theoretical values. Also, in most cases, the experimental values of relative abundances in the individual isotopic patterns were below 1% of their theoretical values (it can be seen in Table 3). These facts suggest that with the LDI MS method, reliable determination of the composition of corrosion films is possible.

The results obtained via the LDI MS method in this work are comparable with previous investigations on the behavior of Ag-Cu alloys in the presence of chloride ions. For example, earlier, it has been shown that the addition of chloride ion resulted in competitive adsorbability between Cl− and OH− ions at the alloy surface. The consequence of this is the formation CuCl, CuCl2·H2O, CuCl2·3[Cu(OH)2], and AgCl, and the formation of Cu2O, Cu(OH)2, CuO, Ag2O2, and AgO on the alloy surface was retarded (Zaky, 2006; Hazzazi et al., 2008). This may be the reason that the ions Cu2OH(H2O)2+ and CuO(H2O)+ have lower intensities, followed by ions of (CuCl)3Cu(OH)2+, in the LDI MS of Ag60Cu26Zn14.

It has been observed that in the LDI mass spectrum, there are AgCl and Ag2Cl2 ions, formed with the assistance of traces of sodium ions. This is equivalent to an earlier study where in the mass spectra of vapors over the NaCl/AgCl mixture, AgCl-Na+ and Ag2Cl2-Na+ ions were detected (Kapala & Skudlarski, 1991). Previous studies have shown that in the laser ablation of a AgCl sample, positively charged species of Ag+, Ag2C1+, and Ag3CI+ are dominant in the LDI mass spectrum. In the same experiment, the formation of Ag2+, Ag2CI2+, and Ag3C13+ was detected as well (Qin et al., 1997). Due to this reason, it can be concluded that Ag2Cl+, Ag2+, Ag2Cl2-Na+, and Ag3+ ions, detected here, originate from AgCl in the anodic film on the Ag60Cu26Zn14 alloy.

Comparison results obtained using the LDI MS, the Raman spectroscopy, and XRD analysis for the corrosion films at Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys after the anodic polarization treatment are presented in Table 4.

Table 4:

Overview of detected species recorded after the anodic potentiostatic polarization treatment of Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys, using the Raman spectra and XRD, and LDI MS.

Sample Raman analysis XRD analysis LDI MS
Ag60Cu26Zn14 Ag Ag Ag
AgCl AgCl AgCl
/ / AgCuCl
CuCl CuCl (CuCl)3Cu(OH)2
Cu2O Cu2O Cu2OH(H2O)2 and CuO(H2O)
/ AgZn /
β-Zn(OH)Cl Zn2OCl2x2H2O Zn2Cl2O5(OH)3(H2O)
Ag58.5Cu31.5Pd10 Ag Ag Ag
AgCl AgCl AgCl
CuCl CuCl (CuCl)3Cu(OH)2
/ Cu3Pd /
/ Pd Pd
/ / AgCuCl
Cu2O / Cu2OH(H2O)2 and CuO(H2O)

It can be clearly observed from Table 4 that the results of chemical composition for the anodic film on Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys, obtained using the Raman, XRD, and LDI analysis, are very similar. For the LDI method, the formation of ions of the type M-H+ or M-H2O+ (M-metal atom) are expected; therefore, there are minor disagreements with the results obtained by the standard methods (the case of Cu2O). The presence of Cu3Pd (XRD analysis) was not confirmed by LDI MS. However, it should be highlighted that the LDI MS method identified species such as (CuCl)3Cu(OH)2+ and AgCuCl, which the Raman and XRD analysis did not detect.

The results presented here demonstrate that the LDI MS can detect new, unusual species. Therefore, inclusion of the LDI analysis as an additional technique for the characterization of corrosion films may be proposed.

4 Conclusion

The present work used the XRD, MRS, SEM/EDS, and LDI MS for investigating the chemical composition of the outer corrosion films of Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloys obtained after anodic potentiostatic polarization treatment (at +0.5 V for 5 min in 3.5% wt. NaCl solution).

  1. Polarization curves confirmed thermodynamic analysis for choosing the potential for the potentiostatic polarization. They indicate the formation of a CuCl film for both alloys and AgCl for AgCuPd alloy. Additionally, they suggest that the surface films are porous, which was confirmed by SEM images.

  2. The XRD analysis showed the existence of CuCl and AgCl in both anodic films. Besides them, Zn2OCl·2H2O was observed as the corrosion product for the zinc-bearing alloy. The weak peaks of Cu2O were detected only in the case of the same alloy but not for the palladium-bearing one. The XRD patterns of the Ag phase for Ag60Cu26Zn14 and Cu3Pd phase for Ag58.5Cu31.5Pd10 were also observed. This is consistent with the finding that the Cu-rich phase in the first alloy was corroded more heavily than the Ag-rich phase, and in the second alloy, the Ag-rich phase was corroded more strongly than Pd-rich phases.

  3. The MRS analysis showed that Ag, CuCl, AgCl, β-Zn(OH)Cl, and Cu2O form the anodic layer on the surface of the Ag60Cu26Zn14 alloy. Analysis of the Raman spectrum of Ag58.5Cu31.5Pd10 has a similar finding, where only β-Zn(OH)Cl was not part of it. The finding of Cu2O for the Ag-Cu-Zn alloy was the confirmation of the XRD results. The Raman method only detected the same compound in the surface layer of the Ag-Cu-Pd alloy.

  4. The SEM micrographs, obtained for anodic films, confirm the finding of the XRD analysis. The main components of the Ag-Cu-Zn alloy film were cuprous chloride and zinc oxychloride. For the Ag-Cu-Pd alloy, the main compound was silver chloride. Cuprous oxide cannot be detected, which supports that it is probably the inner film in the layer. The porosity of the film on the surface, indicated from the chronoamperometric curves, was confirmed by this method.

  5. The LDI mass spectra of both corrosion films contain the following ion species: Ag+, AgCl-Na+ (with their clusters, such as Ag2Cl+, Ag2Cl2-Na+, Ag2+, and Ag3+) AgCuCl+, AgCuCl(H2O)+, Cu+, Cu(H2O)+, (CuCl)3Cu(OH)2+, CuO(H2O)+, and Cu2OH(H2O)2+. Besides the mentioned species, the LDI MS of the corrosion film of Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 alloy contains yet Zn2Cl2O5(OH)3(H2O)+ and Pd+, respectively. The LDI MS analysis is shown that the ion species, obtained in this experimental condition, are similar for Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10, however, their intensities are varied according to the chemical composition of sample.

The LDI MS method offers high mass accuracy, very good agreement between the experimental and theoretical isotopic patterns of detected ions, high analysis speed (tens of spectra can be obtained in a few minutes), representative mass spectrum of the sample, and the possibility of detecting new unusual species.

Hence, the LD MS analysis, in conjunction with the other techniques (XRD, MRS, and SEM/EDS) can be useful to obtain the depth profiles of corrosion product on the surface of Au-Cu alloy.

Acknowledgments

The authors are grateful to Prof. D. Manasijević from TF Bor University of Belgrade for thermodynamic calculations, performed in Pandat 8.1 software.

  1. Funding: This work is the result of the Projects OI 172019 and TR 34033, funded by the Ministry of Education, Science and Technological Development of the Republic of Serbia.

References

Ankah GN, Meimandi S, Renner FU. Dealloying of Cu3Pd single crystal surfaces. J Electrochem Soc 2013; 160: C390–C395.10.1149/2.079308jesSearch in Google Scholar

Bernard MC, Hugot-Le Goff A, Massinon D, Phillips N. Underpaint corrosion of zinc-coated steel sheet studied by in situ Raman spectroscopy. Corros Sci 1993; 35: 1339–1349.10.1016/0010-938X(93)90356-LSearch in Google Scholar

Chang YA, Golberg D, Neumann JP. Phase diagrams and thermodynamic properties of ternary copper-silver systems. J Phys Chem Ref Data 1977; 6: 621–674.10.1063/1.555555Search in Google Scholar

Chao CL, Chu CL, Fuh YK, Hsu RQ, Lee S, Cheng YN. Interfacial characterization of nickel-yttria-stabilized zirconia cermet anode/interconnect joints with Ag-Pd-Ga active filler for use in solid-oxide fuel cells. Int J Hydrog Energy, 2015; 40: 1523–1533.10.1016/j.ijhydene.2014.11.028Search in Google Scholar

Chen KI, Chern Lin JH, Ju CP. Microstructure and segregation behavior of palladium in silver-copper-palladium alloys. J Dent Res 1996; 75: 1497–1502.10.1177/00220345960750071001Search in Google Scholar PubMed

Chiba A, Kusayanagi Y. Corrosion of dental Au-Ag-Cu-Pd alloys in 0.9% sodium chloride solution. Corr Sci Techn 2005; 4: 19–22.Search in Google Scholar

Compaan A, Cummins HZ. Resonant quadrupole-dipole Raman scattering at the 1S yellow exciton in Cu2O. Phys Rev Lett 1973; 31: 4150.10.1103/PhysRevLett.31.41Search in Google Scholar

Dimitrijević SP, Lazarević ZŽ, Rajčić-Vujasinović M, Dimitrijević SB, Petrović M, Gilić M, Jokić BM. Raman spectroscopy study of anodic film on Ag43Cu37Zn20 alloy. Optoelectron Adv Mater Rapid Commun 2016; 10: 777–780.Search in Google Scholar

Dimitrijević SP, Manasijević D, Kamberović Ž, Dimitrijević SB, Mitrić M, Gorgievski M, Mladenović S. Experimental investigation of microstructure and phase transitions in Ag-Cu-Zn brazing alloys. J Mater Eng Perform 2018; 27: 1570–1579.10.1007/s11665-018-3258-1Search in Google Scholar

Duncan MW, Roder H, Hunsucker SW. Quantitative matrix-assisted laser desorption/ionization mass spectrometry. Brief Funct Genomic Proteomic 2008; 7: 355–370.10.1093/bfgp/eln041Search in Google Scholar PubMed PubMed Central

Frsot RL. Raman spectroscopy of selected copper minerals of significance in corrosion. Spectrochim Acta Part A 2003; 59: 1195–1204.10.1016/S1386-1425(02)00315-3Search in Google Scholar

Greener EH, Szurgot K. Properties of Ag-Cu-Pd dispersed phase amalgam: compressive strength, creep, and corrosion. J Dent Res 1982; 61: 1192–1194.10.1177/00220345820610101801Search in Google Scholar PubMed

Grund T, Gester A, Wagner G, Habisch S, Mayr P. Arc brazing of aluminium, aluminium matrix composites and stainless steel in dissimilar joints. Metals 2018; 8: 16.10.3390/met8030166Search in Google Scholar

Hardy JS, Kim JY, Thomsen EC, Weil KS. Improved wetting of mixed ionic/electronic conductors used in electrochemical devices with ternary air braze filler metals. J Electrochem Soc 2007; 154: P32–P39.10.1149/1.2430691Search in Google Scholar

Hassan HH, Ibrahim MAM, Abd El Rehim SS, Amin MA. Comparative studies of the electrochemical behavior of silver electrode in chloride, bromide and iodide aqueous solutions. Int J Electrochem Sci 2010; 5: 278–294.10.1016/S1452-3981(23)15284-9Search in Google Scholar

Haynes WM, editor. CRC Handbook of chemistry and physics: a ready-reference book of chemical and physical data, 95th ed., Boca Raton, Florida: CRC Press, 2014.Search in Google Scholar

Hazzazi OA, Zaky AM, Amin MA, Abd El Rehim SS. Passivation and Cl− induced depassivation of Cu-Ag alloys in borate buffer solutions. Int J Electrochem Sci 2008; 3: 489–508.10.1016/S1452-3981(23)15468-XSearch in Google Scholar

Hidaka H, Hanyu N, Sugano M, Kawasaki K, Yamauchi K, Katsuyama T. Analysis of human serum lipoprotein lipid composition using MALDI-TOF mass spectrometry. Ann Clin Lab Sci 2007; 37: 213–221.Search in Google Scholar

Hosseinpour S, Johnson M. Vibrational spectroscopy in studies of atmospheric corrosion. Materials (Basel) 2017; 10: 413–421.10.3390/ma10040413Search in Google Scholar PubMed PubMed Central

Kapala J, Skudlarski K. The thermodynamics of {xNaCl + (1−x)AgCl} measured by mass spectrometry. J Chem Thermodyn 1991; 23: 667–670.10.1016/S0021-9614(05)80204-6Search in Google Scholar

Karas M, Hillenkamp F. Laser desorption ionization of proteins with molecular masses exceeding 10,000 Daltons. Anal Chem 1988; 60: 2299–2301.10.1021/ac00171a028Search in Google Scholar PubMed

Kee-Baek S, Su-Jung Y, Hyeong-Seop I, Tae-Yeon S. Ag-Pd-Cu alloy reflector to improve the opto-electrical performance and electromigration resistance of near ultraviolet GaN-based light-emitting diode. J Alloys Compd 2019; 800: 512–517.10.1016/j.jallcom.2019.06.119Search in Google Scholar

Lai Z, Xue S. Study on microstructure and property of brazed joint of AgCuZn-X (Ga, Sn, In, Ni) brazing alloy. Rare Met Mater Eng 2010; 39: 397–400.10.1016/S1875-5372(10)60087-2Search in Google Scholar

Lee HP, Nobe K. Kinetics and mechanisms of Cu electrodissolution in chloride media. J Electrochem Soc 1986; 133: 2035–2043.10.1149/1.2108335Search in Google Scholar

Liang H, Li Z, Wang W, Wu Y, Xu H. Highly surface-roughened ‘flower-like’ silver nanoparticles for extremely sensitive substrates of surface-enhanced Raman scattering. Adv Mater 2009; 21: 4614–4618.10.1002/adma.200901139Search in Google Scholar

Lutz HD, Schmidt M, Weckler B. Infrared and Raman studies on calcium, zinc and cadmium hydroxide halides Ca{O(H,D)}Cl, Cd{O(H,D)}Cl, Zn{O(H,D)}F and β-Zn{O(H,D)}Cl. J Raman Spectrosc 1993; 24: 797–804.10.1002/jrs.1250241113Search in Google Scholar

Ma AL, Jiang SL, Zheng YG, Ke W. Corrosion product film formed on the 90/10 copper–nickel tube in natural seawater: composition/structure and formation mechanism. Corros Sci 2015; 91: 245–261.10.1016/j.corsci.2014.11.028Search in Google Scholar

Mawale R, Halenkovič T, Bouška M, Gutwirth J, Nazabal V, Bora PL, Pečinka L, Prokeš L, Havel J, Němec P. Mass spectrometric investigation of amorphous Ga-Sb-Se thin films. Sci Rep 2019; 9: 10213.10.1038/s41598-019-46767-8Search in Google Scholar PubMed PubMed Central

Nakayama M, Ando N. Immersion test of the Ag-Pd-Cu alloy system. Shigaku 1999; 86: 768–778.10.1007/BF03039012Search in Google Scholar

Ntasi A, Al Jabbari YS, Silikas N, Al Taweel SM, Zinelis S. Metallurgical characterization of experimental Ag-based soldering alloys. Saudi Dent J 2014; 26: 139–144.10.1016/j.sdentj.2014.05.001Search in Google Scholar PubMed PubMed Central

Pangavhane SD, Nemec P, Wagner T, Janca J, Havel J. Laser desorption ionization time-of-flight mass spectrometric study of binary As-Se glasses. Rapid Commun Mass Spectrom 2010; 24: 2000–2008.10.1002/rcm.4607Search in Google Scholar PubMed

Pati S, Jat RA, Anand NS, Derose DJ, Karn KN, Mukerjee SK, Parida SC. Pd-Ag-Cu dense metallic membrane for hydrogen isotope purification and recovery at low pressures. J Membr Sci 2017; 522: 151–158.10.1016/j.memsci.2016.09.025Search in Google Scholar

Pereira JC, Nascimento RM, Martinelli AE, Aechar W, Rocha L. Wetting behavior of silver braze alloys onto metallized zirconia inserts. Welding J 2015; 94: 211–218.Search in Google Scholar

Qin Q, Zhou M, Mao D. Time-of-flight mass spectrometric study on UV laser ablation of silver chloride. Appl Surf Sci 1997; 119: 321–329.10.1016/S0169-4332(97)00204-3Search in Google Scholar

Raju K, Kim S, Song K, Yu JH, Yoon DH. Joining of metal-ceramic using reactive air brazing for oxygen transport membrane applications. Mater Des 2016; 109: 233–241.10.1016/j.matdes.2016.07.068Search in Google Scholar

Ramirez-Galicia G, Pena-Mendez EM, Pangavhane SD, Alberti M, Havel J. Mass spectrometry and ab initio calculation of (n=1–7) ion structures. Polyhedron 2010; 29: 1567–1574.10.1016/j.poly.2010.02.012Search in Google Scholar

Rashidzadeh H, Guo B. Generation of large gas-phase silver cluster ions by laser desorption/ionization of silver-containing salts. Chem Phys Lett 1999; 310: 466–470.10.1016/S0009-2614(99)00819-2Search in Google Scholar

Ren HS, Xiong HP, Long WM, Shen YX, Pang SJ, Chen B, Cheng YY. Interfacial diffusion reactions and mechanical properties of Ti3Al/Ni-based superalloy joints brazed with AgCuPd filler metal. Mater Charact 2018; 144: 316–324.10.1016/j.matchar.2018.07.024Search in Google Scholar

Sauer S. The essence of DNA sample preparation for MALDI mass spectrometry. J Biochem Bioph Methods 2007; 70: 311–318.10.1016/j.jbbm.2006.10.007Search in Google Scholar PubMed

Sherif EM. Inhibition of copper corrosion reactions in neutral and acidic chloride solutions by 5-ethyl-1,3,4-thiadiazol-2-amine as a corrosion inhibitor. Int J Electrochem Sci 2012; 7: 2832–2845.10.1016/S1452-3981(23)13918-6Search in Google Scholar

Subramanian PR, Laughlin DE. Cu-Pd (copper-palladium). J Phase Equilib 1991; 12: 231–243.10.1007/BF02645723Search in Google Scholar

Tanaka K, Waki H, Ido Y, Akita S, Yoshida Y, Yoshida T, Matsuo T. Protein and polymer analyses up to m/z 100,000 by laser ionization time-of-flight mass spectrometry. Rapid Commun Mass Spectrom 1988; 2: 151–153.10.1002/rcm.1290020802Search in Google Scholar

Upadhyay D, Panchal MA, Dubey RS, Srivastava VK. Corrosion of alloys used in dentistry: a review. Mater Sci Eng 2006; 432: 1–11.10.1016/j.msea.2006.05.003Search in Google Scholar

Von der Osten W. Polarized Raman spectra of silver halide crystals. Phys Rev B 1974; 9: 789–793.10.1103/PhysRevB.9.789Search in Google Scholar

Vurdelja BD, Dimitrijević SP, Dimitrijević SB, Kamberović ŽJ, Veličković SR. Characterization of the Ag43Cu37Zn20 alloy surface after potentiostatic polarization using LDI-TOF mass spectrometry. Corros Rev 2017; 35: 473–481.10.1515/corrrev-2017-0065Search in Google Scholar

Vvedenskii AV, Grushevskaya SN. Kinetic peculiarities of anodic dissolution of copper and its gold alloys accompanied by the formation of insoluble Cu(I) products. Corr Sci 2003; 45: 2391–2413.10.1016/S0010-938X(03)00064-7Search in Google Scholar

Wang H, Xue S. Effect of Ag on the properties of solders and brazing filler metals. J Mater Sci: Mater Electron 2016; 27: 1–13.10.1007/s10854-015-3747-zSearch in Google Scholar

Wang CB, Deo G, Wachs IE. Interaction of polycrystalline silver with oxygen, water, carbon dioxide, ethylene, and methanol: in situ Raman and catalytic studies. J Phys Chem B 1999; 103: 5654–5656.10.1021/jp984363lSearch in Google Scholar

Xiong W, Hao L, Li Y, Tang D, Cui Q, Feng Z, Yan C. Effect of selective laser melting parameters on morphology, microstructure, densification and mechanical properties of supersaturated silver alloy. Mater Des 2019; 170: 107697.10.1016/j.matdes.2019.107697Search in Google Scholar

Xue P, Zou Y, He P, Pei Y, Sun H, Ma C, Luo J. Development of low silver AgCuZnSn filler metal for Cu/steel dissimilar metal joining metals. Metals 2019; 9: 1–10.10.3390/met9020198Search in Google Scholar

Ye D, Xiong W, Zhang X, Qu J, Yao Z. Microstructure and shear strength of the brazed joint of Ti(C,N)-based cermet to steel. Rare Met 2010; 29: 72–77.10.1007/s12598-010-0013-7Search in Google Scholar

Zaharinie T, Yusof F, Fadzil M, Hamdi M, Ariga T. Microstructural analysis of brazing sapphire and Inconel 600 for sensor applications. Mater Res Innov 2014; 18: 68–72.10.1179/1432891714Z.000000000934Search in Google Scholar

Zaky AM. Role of Cl− in breakdown of Cu-Ag alloys passivity in aqueous carbonate solutions. Electrochim Acta 2006; 51: 2057–2062.10.1016/j.electacta.2005.07.013Search in Google Scholar

Zemen Y, Teusch HS, Kropke G, Pingel S, Held S. The impact of busbar surface topology and solar cells soldering process. Proceedings of 27th European Photovoltaic Solar Energy Conference and Exhibition, Frankfurt, Germany, 2012: 24–28.Search in Google Scholar

Zhang X, Wallinder IO, Leygraf C. Mechanistic studies of corrosion product flaking on copper and copper-based alloys in marine environments. Corr Sci 2014; 85: 15–25.10.1016/j.corsci.2014.03.028Search in Google Scholar

Zugic B, Wang L, Heine C, Zakharov DN, Lechner BA, Stach EA, Biener J, Salmeron M, Madix RJ, Friend CM. Dynamic restructuring drives catalytic activity on nanoporous gold-silver alloy catalysts. Nat Mater 2017; 88: 558–564.10.1038/nmat4824Search in Google Scholar PubMed

Received: 2019-08-09
Accepted: 2019-10-19
Published Online: 2020-01-11
Published in Print: 2020-03-26

©2020 Walter de Gruyter GmbH, Berlin/Boston

Articles in the same Issue

  1. Frontmatter
  2. Review
  3. Corrosion protection of copper and copper alloys in different corrosive medium using environmentally friendly corrosion inhibitors
  4. Original articles
  5. Complementary methods for characterization of the corrosion products on the surface of Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 brazing alloys
  6. Influence of temperature and potential range on Zn-Ni deposition properties formed by cyclic voltammetry electrodeposition in chloride bath solution
  7. Evaluation of corrosion inhibition and adsorption behavior of Thymuszygis subsp. gracilis volatile compounds on mild steel surface in 1 m HCl
  8. Non-ionic surfactant loaded on gel capsules to protect downhole tubes from produced water in acidizing oil wells
  9. Towards a better understanding of the oxide film growth mechanism in E110 zirconium alloy under high-temperature oxidation in steam
  10. Environmentally assisted cracking of T91 ferritic-martensitic steel in heavy liquid metals
https://doi.org/10.1515/corrrev-2019-0067
Keywords for this article
brazing silver alloys; characterization; corrosion film; LDI; mass spectrometry

Articles in the same Issue

  1. Frontmatter
  2. Review
  3. Corrosion protection of copper and copper alloys in different corrosive medium using environmentally friendly corrosion inhibitors
  4. Original articles
  5. Complementary methods for characterization of the corrosion products on the surface of Ag60Cu26Zn14 and Ag58.5Cu31.5Pd10 brazing alloys
  6. Influence of temperature and potential range on Zn-Ni deposition properties formed by cyclic voltammetry electrodeposition in chloride bath solution
  7. Evaluation of corrosion inhibition and adsorption behavior of Thymuszygis subsp. gracilis volatile compounds on mild steel surface in 1 m HCl
  8. Non-ionic surfactant loaded on gel capsules to protect downhole tubes from produced water in acidizing oil wells
  9. Towards a better understanding of the oxide film growth mechanism in E110 zirconium alloy under high-temperature oxidation in steam
  10. Environmentally assisted cracking of T91 ferritic-martensitic steel in heavy liquid metals
Sign up now to receive a 20% welcome discount
Institutional Access
How does access work?
Have an idea on how to improve our website?
Please write us.
© 2025 De Gruyter Brill
Downloaded on 22.9.2025 from https://www.degruyterbrill.com/document/doi/10.1515/corrrev-2019-0067/html
Scroll to top button