Selective Dissolution Of Cu Research Articles

Early Stages of Metal Corrosion in Coastal Archaeological Sites: Effects of Chemical Composition in Silver and Copper Alloys.

In this study, metal disks with different chemical composition (two Ag-based alloys and three Cu-based alloys) were buried in the soil of coastal archaeological sites for a period of 15 years. The aim was to naturally induce the growth of corrosion patinas to obtain a deeper insight into the role of alloying elements in the formation of the patinas and into the degradation mechanisms occurring in the very early stages of burial. To reach the aim, the morphological, compositional and structural features of the patinas grown over 15 years were extensively characterized by optical microscopy, field emission scanning electron microscopy coupled with energy dispersive spectrometry, X-ray diffraction and micro-Raman spectroscopy. Results showed that the Cu amount in Ag-based alloys strongly affected the final appearance, as well as the composition and structure of the patinas. Corrosion mechanisms typical of archaeological finds, such as the selective dissolution of Cu, Pb and Zn and internal oxidation of Sn, occurred in the Cu-based alloys, even if areas enriched in Zn and Pb compounds were also detected and attributed to an early stage of degradation. In addition, some unusual and rare compounds were detected in the patinas developed on the Cu-based disks.

Employing magnesium-lead melt for synergetic and selective extraction of copper from copper-cobalt alloy

This research aims to extract Cu from Cu-Co alloy with high efficiency and selectivity by employing binary Mg-Pb melt. The optimal conditions for the extraction of Cu were determined. The results showed under optimal conditions, 96.5% of Cu in the Cu-Co alloy could be selectively extracted after treatment at 800 °C for 1 h, with the extraction rates of only 0.2% Fe, 0.6% Co, and 1.4% Si. The dissolution mechanism involved the counter diffusion of Mg/Pb and Cu across the diffusion zone of the Cu-Co alloy, and Mg in the binary Mg-Pb melt played a major role in the selective dissolution of Cu, especially at the dissolution forefront. The rate-controlling step of the extraction was dominated by the interfacial reaction.

Dissolution and the Consequent Surface Enrichment of Platinum in Pt–Cu Binary Alloys under Potential Cycling

In this study, electrochemistry was combined with mass spectroscopy and surface analysis to investigate the selective dissolution of Cu and the consequent surface enrichment of Pt in Pt–Cu binary alloys. Two different compositions of Pt–50at% Cu (Pt–50Cu) and Pt–75at% Cu (Pt–75Cu) underwent potential cycling. Cu from both alloys was selectively dissolved, although trivial Pt dissolution occurred, causing the subsequent formation of a Pt-enriched layer on the alloys’ surfaces. The morphology of the layer was quite different between the two alloys. The Pt-enriched layer formed on the Pt–50Cu surface was extremely thin and compact, further suppressing Cu dissolution from the alloy in the later stage of potential cycling. The Pt-enriched layer formed on the Pt–75Cu surface contained numerous pits, mainly because a lot of Cu was dissolved from Pt–75Cu during potential cycling. Thus, Cu continued to dissolve from the bottom of the pits at the later stage, causing the thickness of the Pt-enriched layer to gradually grow with an increase in the cycle number. The correlation between Pt and Cu dissolution and surface morphological changes of Pt–Cu alloys was discussed.

Dissolution and the Consequent Surface Morphology of Pt-Cu Binary Alloys Under Potential Cycling

Polymer electrolyte fuel cell (PEFC) is regarded as clean energy conversion system, however its high cost prohibits the progress of PEFC. Thus, a reduction in the amount of Pt used as cathode catalyst is required. In order to overcome this problem, Pt transition-metal (Pt-M) alloy catalysts have been developed instead of Pt catalysts, however a deterioration of Pt-M alloy catalysts is concerned because both Pt and M are dissolved from Pt-M alloy catalysts under PEFC operating conditions. In this study, potential cycling tests and inductively coupled plasma mass spectroscopy (ICP-MS), field emission-scanning electron microscopy (FE-SEM), and field emission auger electron spectroscopy (FE-AES) were combined, and dissolution behavior of Pt-Cu alloys has been investigated under potential cycling by using these techniques. Pt50-Cu50 (Pt-50 at% Cu) and Pt25-Cu75 were fabricated by arc-melting under an Ar atmosphere. These alloys were subjected to potential cycling tests at 298 K in 0.5 M H2SO4 solution with a two-compartment stationary cell. A double junction silver / silver-chloride electrode was used as reference electrode, and Au coil was used as the counter electrode. A total of 1,000 cycles of potential cycling tests were employed between 0.05 V and 1.4 V vs. SHE at 100 mV s-1. Tests was interrupted every 100 cycles, and the test solution was sampled for quantitative ICP-MS analysis of both Pt and Cu ions dissolved in it. In addition, surface structure and depth profile after potential cycling tests were evaluated by FE-SEM and FE-AES, respectively. FE-AES depth profiles before and after 100 cycle potential cycling confirmed that Pt-enriched layer formed on Pt25-Cu75 surface after potential cycling. Moreover, the amount of dissolved Pt and Cu ions during 1st sequential order of 100-cycles-potential cycling were 1 monolayer (ML) and 2370 ML, respectively. Hence, Pt-enriched layer formed by the large amount of Cu selectively dissolved from Pt25-Cu75. FE-SEM image of Pt25-Cu75 after 100 cycle potential cycling showed numerous pits (d pit = 2 – 5 nm) formed on its surface, and the large amount of Cu dissolution occurs from 2nd to 10th sequential order of 100-cycles-potential cycling. Thus, selective dissolution of Cu seems not to be suppressed in the latter cycles because Cu dissolution continuously occurs through the pits. On the other hand, FE-SEM image of Pt50-Cu50 after 100 cycle potential cycling was extremely flat. Accordingly, dense Pt-enriched layer formed on its surface, and it suppressed further Cu dissolution from the alloy. The authors acknowledge Ookayama Materials Analysis Division, Tokyo Institute of Technology, for assistance with the ICP-MS analysis.

Preparation of Pt-Cu Catalyst Particles Using Electrodeposition Method and Evaluation of Durability and ORR Activity

Introduction Although Proton exchange membrane fuel cells (PEMFCs) are clean and efficient energy converters, the amount of platinum used as the catalyst for the oxygen reduction reaction (ORR) at the cathodic electrode is still large and has to be reduced to drive down cell manufacturing costs. Core-shell structured Pt alloys with relatively abundant transition metals, such as Cu, Co, and Ni, have demonstrated significant improvements for the ORR activities, suggesting one solution to tackle this problem1). However, it has been reported that Pt-M alloys catalysts degrade more easily than pure Pt because of selective dissolution of M, leading to severely reduced performance in long run operation2). Taylor et al. have shown that Pt/C-catalyzed GDL fabrication using the electrodeposition method enables the high Pt utilization ratio in the catalyst layer because it ensures that Pt is deposited only on the electrochemically active area3). Woo et al. have reported considerable single cell performance of Pt-Co catalysts prepared on unloaded GDL using the electrodeposition method4). However, more investigations are needed to clarify their corrosion and degradation behaviors. In this work, we present the preparation of Pt-Cu catalyst particles using the electrodeposition method, and the evaluation of durability and ORR activity using electrochemical measurements such as immersion test and ORR measurement. Experimental Glassy Carbon was used as a substrate on which Pt-Cu nanoparticles were deposited. Direct current electrodeposition of Pt-Cu nanoparticles was conducted in a conventional three-electrode cell. A double junction KCl-saturated Ag / Ag-Cl electrode was used as a reference electrode, and a carbon rod was used as the counter electrode. Electrodeposition was carried out in a plating bath containing a solution of 6 mM K2PtCl4 and various concentrations(1-40 mM) of CuSO4・5H2O dissolved in 0.5M Na2SO4. The parameters for direct current electrodeposition were an electrodeposition potential of -0.05 V vs. SHE and a total charge density of 0.11 C cm−2. Chemical compositions and particle diameters were determined by SEM-EDX. The duration of the immersion test was 3 h in 0.5 M H2SO4, followed by the investigation of the dissolved amount of Cu using ICP-MS. ORR activities were determined by measuring the potentiostatic polarization curve in 0.5 M H2SO4 from OCP to 0.6 V vs. SHE at a scan rate of 0.2 mV s-1. All electrodeposition processes and electrochemical measurements were conducted open to the atmosphere and at 25 oC. Results and Discussion SEM-EDX observations confirmed that various compositions of Pt-Cu were prepared and homogeneous nanoparticles were obtained in each sample. Chemical compositions were controlled from 13 to 85 at. % Cu by changing the concentration of precursor in a plating bath. The amounts of dissolved Cu ions during the immersion test drastically increased with atomic ratio of Cu as shown in Fig. 1(a). The corrosion potential monitored during the immersion test, all Pt-Cu catalysts formed the Pt enriched layer on the surface by selective dissolution of Cu. However, Cu-rich alloys did not suppress the dissolution of Cu. In addition, SEM observations after the immersion test displayed that the surface morphology of Cu-rich Pt-Cu nanoparticles completely changed. Fig. 1(b) shows specific activities at 0.9 V initial against Cu composition. Pt58Cu42 indicated the highest ORR activity, 2.6 times higher than pure Pt. Fe group metal content enhances Oxygen adsorption and weakens O-O bond5) and thus ORR activity is enhanced by the effect of alloying when Cu content increases. From our results, however, the Pt enriched layer which weakens the effect of alloying becomes thicker as Cu content increases because the amount of dissolved Cu ions also increases. As a result, optimum composition is determined by the balance between the effect of alloying and the thickness of the Pt enriched layer. Pulse current electrodeposition of Pt-Cu nanoparticles on GDL will be further discussed in the session. References 1) Mehtap Oezaslana, Peter Strasser, Journal of Power Sources 196 (2011) 5240–5249. 2) S. Mukerjee, S. Srinivasan, Handbook of Fuel Cells, vol2 (2003) 502. 3) E.J. Talyor, E.B. Anderson, N.R.K. Vilambi, J. Electrochem. Soc. 139 (1992) L45. 4) Seunghee Woo, In Kim, Jae Kwang Lee, Sungyool Bong, Jaeyoung Lee, Hasuck Kim, Electrochim. Acta 56 (2011) 3036–3041. 5) T. Toda, H. Igarashi, H. Uchida, and M. Watanabe, J. Electrochem. Soc. 146 (1999) 3750. Figure 1

Cerium addition on pitting corrosion of (Cu50Zr50)100–2xCe2x(x=0, 1, 2 and 3) metallic glasses in seawater

The industrial application of metallic glass is a longstanding challenge for researchers in the field. Toward this objective, the electrochemical performance in sea water of Cu-Zr-(Ce) metallic glass with various Ce content was investigated. Cu-Zr-(Ce) metallic glass was fabricated by melt-spinning technique and characterized by X-ray diffraction. The corrosion resistance in seawater was then investigated by potentiodynamic polarization, immersion test, scanning electron microscopy, electron probe microanalysis, and energy dispersive spectrometer analysis. The results showed that Ce addition lowered the corrosion current density of Ce-containing Cu-Zr alloy system. The attack type changed from uniform corrosion of Cu50Zr50 metallic glasses to local one of the Ce-containing alloys. Appropriate content of Ce inhibited the selective dissolution of Cu in the pits and thus improved the corrosion resistance of the alloys.

Electrochemically fabricated hierarchical porous Ni(OH)2/NiCu electrodes for hydrogen evolution reaction

A hierarchical porous Ni(OH)2/NiCu electrode was fabricated by a novel electrochemical method, which included brush plating, cyclic voltammetry (CV) treatment, electrochemical dealloying and hydrolysis. We firstly discussed the feasibility of using brushing plating to produce NiCu precursors. Then we found that the CV treatment can be used to yield a roughened NiCu surface; thereafter, the selective dissolution of Cu can be used to further improve the surface area. Through controlling the cycle number and time of dealloying, we can get a hierarchical porous NiCu which exhibit highest hydrogen evolution reaction activity. Subsequently, the obtained hierarchical porous NiCu immersed in 0.1M NiCl2 to get a hierarchical porous Ni(OH)2/NiCu electrode. The influence of immersing time was also discussed. The fabricated hierarchical porous Ni(OH)2/NiCu electrode exhibits higher hydrogen evolution reaction (HER) activity than those of the brush-plated NiCu and porous NiCu electrodes, which can be attributed to the unique hierarchical porous structure and the synergy between the Ni(OH)2 and NiCu surfaces.

Rational designing of nanoporous nanopattern arrays of Au, Pt and SiO2: synthesis using lithography, sputtering and selective dissolution

Nanoporous materials are of great importance in a variety of applications such as heterogeneous catalysts because of their large specific surface area and tuneable porosity. Herein, we proposed an approach for rationally designing nanoporous materials on a substrate by lithography, co-sputtering and selective dissolution. The physical processes (i.e., lithography and sputtering) were used to design shapes and arrangements of nanostructures, and the chemical dissolution process was used to form nanopores inside the nano-patterns. In order to investigate the compatibility of the sputtering and selective etching with the lithography process, we formed patterned nanoporous Au by co-sputtering of AuCu followed by selective dissolution of Cu with acid. Interestingly, using this process, dish-shaped Au nanostructure arrays formed through lithography processes were successfully decorated with nanoporous Au, resulting in the enhancement of SERS intensity by 10 times. Then, in order to show the versatility of our approach, we synthesized nanoporous SiO2 patterns. In this case, nanoporous SiO2 was synthesized by co-sputtering of SiO2 with Cu followed by selective dissolution of Cu with acid. Because of removal of the thermal process, this SiO2 synthesis appears to be compatible with a variety of nanofabrication procedures. Finally, in order to investigate further merits of the use of sputtering, we demonstrated the synthesis of dendritic Pt nanoporous particle arrays by co-sputtering and acid treatment. Interestingly, nanoporous particle arrays, i.e., hierarchical structure, were formed from PtCu crystal grains without coagulation, showing a possibility that crystal grains can play a role in the formation of hierarchical nanostructures.

Dealloying of Cu3Pd Single Crystal Surfaces

Nanoporous metallic structures formed from dealloying have recently attracted considerable interest due to a wide range of potential applications in areas such as catalysis, actuators, biomedical sensors and fuel cell electrodes. The Cu-Pd system is interesting as a model system for corrosion as well as a potential catalyst material. We report a combined in-situ X-ray diffraction (XRD) and ex-situ atomic force microscopy (AFM) study of the initial structural evolution during the selective dissolution of Cu from Cu3Pd (111) and Cu3Pd (100). The experiments were performed under potential control in 0.1 M H2SO4 solution. Below the critical potential, no distinct diffraction signal is evidenced by in-situ XRD but AFM reveals the existence of islands on the surfaces. At the critical potential, on both single crystal surfaces the formation of an epitaxial Pd layer exhibiting substrate orientation is evidenced by XRD. Ex-situ AFM imaging clearly showed at comparable stages the formation of larger nanoscale islands. We further compare and contrast the behavior of the dealloyed Cu-Pd surfaces with that of Cu-Au alloys.

Improved pseudocapacitive performance and cycle life of cobalt hydroxide on an electrochemically derived nano-porous Ni framework

The pseudocapacitance and morphology of an electrodeposited cobalt hydroxide (Co(OH) 2) significantly depends on the architecture of the electrode substrate. The nano-porous Ni framework, derived from the selective dissolution of Cu from a Ni–Cu alloy, effectively promotes the electrochemical utilization of deposited Co(OH) 2 even at a high loading amount condition. The great electronic and ionic conduction within the nano-structured electrode improves the energy storage performance of Co(OH) 2 as compared to that for a conventional flat Ni substrate. In this work, the Co(OH) 2 mass specific capacitance, evaluated using cyclic voltammetry (CV), only slightly decreases from 2650 to 2470 F g −1 when the potential sweep rate is substantially increased from 5 to 200 mV s −1. The developed Ni(OH) 2/NiOOH (from the nano-porous framework) incorporates with the deposited Co(OH) 2 upon CV cycling; the mixed hydroxide shows a noticeably synergistic capacitance. Furthermore, the dissolution of Co(OH) 2 in KOH electrolyte is greatly suppressed due to the incorporation of Ni(OH) 2/NiOOH, consequently prolonging the electrode cycle life.

In situx-ray diffraction study of the initial dealloying and passivation ofCu3Au(111)during anodic dissolution

We present an in situ x-ray diffraction study of the initial steps of the potential-controlled, selective dissolution of Cu from a ${\text{Cu}}_{3}\text{Au}(111)$ single-crystal surface immersed in 0.1 M ${\text{H}}_{2}{\text{SO}}_{4}$. We followed the structural evolution of an ultrathin Au-rich metallic passivation layer at low overpotentials toward thicker, pure Au islands at elevated overpotentials below the critical potential. The thickness of the ultrathin layer grows from 2 hcp monolayers to several fcc-stacked monolayers. The epitaxial Au-rich layer, which exhibits a reversed stacking with regard to the single-crystal substrate, is passivating the ${\text{Cu}}_{3}\text{Au}(111)$ surface. Complementary ex situ atomic force microscopy (AFM) images are also presented and show at elevated overpotentials a surface completely covered with islands of a homogenous size distribution, which agrees well with the findings from the x-ray experiments. The Au islands are found to be weakly hexagonally correlated.

The influence of chloride on the initial anodic dissolution of Cu 3Au(1 1 1)

We report a detailed in situ X-ray diffraction study of the influence of chloride on the atomic structure evolution at the solid-electrolyte interface during the selective dissolution of Cu from a Cu 3Au(1 1 1) surface immersed in 0.1 M H 2SO 4. We disclose that the formation of the initial ultrathin Au-rich (1 1 1) with an inverted stacking sequence, as recently observed at Cu 3Au(1 1 1) in contact with pure 0.1 M H 2SO 4, is strongly influenced by adding 5 mM HCl. The main finding is a negative shift of about 150 mV of the critical potential at which the ultra-thin Au-rich layer transforms into thicker Au islands. The presented results support the view that it is not a thermodynamic driving force, but rather the rate of surface diffusion that dominates the formation of the structures of the metallic layer.

Composition and electrochemical properties of natural patinas of outdoor bronze monuments

The present study concerns the chemical–physical and electrochemical characterizations of the corrosion compounds formed on outdoor bronzes exposed in urban conditions. Here, we focused on the pale green patinas of surfaces directly exposed to the rainfall leaching. The investigation has been performed from scraped-off patina products of a well representative outdoor bronze monument (Monument to Francis Garnier, 1898) situated in Paris (France). The composition of the patina was analysed by energy dispersive spectrometry (EDS), coupled with SEM, and Raman spectroscopy. Structural characterization was conducted through high-resolution TEM (EDS coupled). The electrochemical reactivity of patina was investigated by cyclic voltammetry and EIS with a microcavity electrode in sulphate solution at pH 2 and 5.6 with time of immersion (up to 10 h). The bronze patina corresponds to a two-layer structure. The external layer is characterised by a marked selective dissolution of Cu and Zn of the alloy, which has been determined from the calculation of dissolution factors. The same phenomenon with the same amplitude has been evidenced on other bronze monuments used for comparison. The pale green patina is a complex mixture of crystalline copper hydroxisulphate and of amorphous/nanocrystal Sn-containing compounds. This patina is stable at pH 5.6 and reactive at pH 2 in relation with the dissolution of copper sulphate compounds in the acidic solution.

Formation of nanoporous platinum by selective dissolution of Cu from Cu0.75Pt0.25

This paper gives results demonstrating the production of nanoporous platinum through the de-alloying of Cu0.75Pt0.25alloy in 1 M H2SO4. Both field emission scanning electron microscopy and small angle neutron scattering confirm the presence of porosity with a diameter of approximately 3.4 nm. This is the smallest porosity quantitatively reported from a de-alloying process to date. The small size is attributed to the extremely small values of surface diffusivity expected for Pt at room temperature, effectively eliminating room-temperature coarsening processes. The results also show that larger length scales can be achieved through coarsening at elevated temperatures. The ease of production of porous platinum makes it attractive for possible applications, such as high surface area electrodes for biomedical devices or as catalyst materials.

Selective Dissolution below the Critical Potential and Back Alloying in Copper-Gold Alloy

Below the critical potential, selective dissolution of Cu from Cu-18 atom % Au proceeds via Cu transport through an essentially planar Au-rich product layer, which over time increases to hundreds of atom layers in thickness. Transmission electron micrographs (TEM) taken after the selective dissolution (at and subsequent (current-off) back alloying processes reveal nearly uniform Moiré patterns characteristic of the superposition of two layers with similar crystal structures, i.e., the Cu-depleted (Au-rich) product layer and the parent alloy. The Moiré fringe spacing decreases with the time of Cu dissolution (for ∼60 s), indicating solid-solution Au enrichment in the product layer during dissolution. The Moiré patterns give no indication of discrete islands of residual Au. The decreasing Cu dissolution rate follows the parabolic law for the first 60 s with an effective diffusivity of Cu in the product layer of at 23°C. The TEMs also show a few pits, whose diameter is in the <10 nm range and whose number increases with time, which may account for the higher rate of Cu dissolution (than given by the parabolic law) after 1 min. © 2001 The Electrochemical Society. All rights reserved.

Comparative investigation on copper oxides by depth profiling using XPS, RBS and GDOES

Depth profiling has been performed by using X-ray photoelectron spectrometry (XPS) in combination with Ar-ion sputtering, Rutherford backscattering spectrometry (RBS) and glow discharge optical emission spectrometry (GDOES). The data obtained by XPS have been subjected to factor analysis in order to determine the compositional layering of the copper oxides. This leads to two or three relevant components within the oxide layers consisting of Cu(2)O or CuO dependent on the sample preparation. GDOES measurements show sputtering profiles which are seriously influenced by a varying sputter rate. To ensure the results obtained so far, RBS measurements of the oxide layers have been carried out in order to discover artefacts of the other methods used and to demonstrate the excellent suitability of RBS for quantitative analysis of these layers. Chemical analysis consisting of (1) carrier-gas fusion analysis (CGFA) and (2) selective dissolution of Cu(2)O/CuO allows the determination of the total amount of oxygen and copper, respectively, and can serve as a cornerstone of quantitative analysis.

Selective dissolution of copper from Au-rich Cu-Au alloys: an electro chemical STM study

The selective dissolution of Cu from CuAu 3 alloys has been studied by STM imaging in situ as a function of anodic polarization. The experiments were performed on (111) oriented films grown on mica that exhibit large terraces separated by steps. Dissolution of Cu in the subcritical region produces vacancies that aggregate to form two-dimensional voids in the terraces. The voids are one atom deep and grow in size as a function of dealloying time. Voids are not formed in the exposed second layer. The layer-by-layer removal of material becomes three-dimensional with profuse surface roughening above the critical potential. A model for the dealloying based on these results is proposed.

I n s i t u scanning tunneling microscopy of inhibited Cu and Cu–Au electrodissolution in aqueous media

I n situ scanning tunneling microscopy (STM) of copper dissolution in the Tafel region, shows that the initial mechanically polished surface (0.05-μm Al2O3 ) roughens by forming large scale (μm) plateaus that appear to consist of low index hkl facets, while the plateaus, themselves, become smoother (&amp;lt;100 nm scale) than the as-prepared surface. The tendencies for both plateau development and smoothing of the plateaus are slowed by the addition of inhibitor to the solution and this inhibiting effect is particularly strong in some regions suggesting the preferential adsorption of the inhibitor on the surface. Real-time STM of the profuse surface roughening (10–102 nm scale) that occurs during selective dissolution of Cu from the Cu–Au alloy is illustrated. As in post (ex situ) STM examination, the scale of the roughness in the plane of the surface is found to be in good agreement with scanning electron microscopy results, but the depth of the roughness (pits) is severely underestimated except at the very initial stages of surface roughening.

Selective dissolution of AuCu and AuAg alloys in vitro and in vivo

AuCu and AuAg alloys (30 at.% Au) have been immersed in aqueous solutions of chloride and sulphide with additions of protein (mucin). The surfaces of the alloys were subsequently analysed by means of ESCA and compared with previous and similar tests without addition of mucin. No noticeable effect of the mucin addition was observed on the surface composition of the alloys after exposure to the various solutions. Cu in contrast to Ag was found to be selectively dissolved from the 2 or 3 uppermost atomic layers in chloride as well as sulphide solutions (pS > 4), both with and without additions of mucin, leaving behind a surface enriched from 10 to 25 at.% in Au. Testing of the same alloys in vivo for 8 d revealed the same tendencies to selective dissolution of Cu. The AuAg alloy was tarnished in the sulphide solutions and in vivo.

Tarnishing in vivo and in vitro of a low-gold alloy related to its structure.

Studies on tarnishing of a low-gold alloy, both as-cast and solid-solution-annealed, have been carried out in vivo and compared with that of a type III gold alloy. Small polished samples were partly embedded in complete dentures for nine months. Generally, the low-gold alloy in the as-cast state was found to be more prone to tarnishing than were the other two materials, followed by the same alloy, solid-solution-annealed. These results are in agreement with both previous and present in vitro studies using 2% solutions of Na2S. The tarnish attacks followed the pattern of Cu segregations in vitro, and frequently also in vivo. Previous findings in vitro, on the other hand, show that tarnishing occurs predominantly on Ag-rich phases. The link is to be sought in the structural details. In the Cu-rich regions along grain boundaries and in interdendritic positions, eutectic particles with thin lamellae (approximately equal to 0.05 micron) of alternating Ag and Cu enrichments occurred. In these areas, we found a more unstable matrix with increased tendency to splitting into similar, but even thinner, lamellae. Most likely, these lamellae of alternating compositions act as galvanic cells. The cathodic reduction of oxygen probably takes place mainly on the Cu-rich lamellae, which are Au-enriched on the surface due to initial selective dissolution of Cu. With increasing time, attacks up to several micron in depth comprising both kinds of lamellae tend to occur at the grain boundary regions in an in vivo test, probably by a fretting corrosion mechanism.