3.1. Reaction Mechanisms
The fundamental principle of this process was the reduction of silver, where silver was either reduced by copper or by potassium sodium tartrate.
When a silver–ammonia solution is added to the suspension formed by copper powder and the reducing agent, two reactions occur simultaneously. The first reaction involves the displacement of silver ions in the solution by copper atoms on the surface of the copper powder. The second reaction involves the reduction of silver–ammonia complex ions by potassium sodium tartrate. Both reactions proceed concurrently without a specific order, leading to the formation of silver atoms. These silver atoms may gradually deposit on the surface of the copper powder, forming a silver coating, or they may remain suspended in the solution and grow into elemental silver particles. The two reactions can be expressed by the following chemical equations:
According to previous study conclusionsAccording to previous study conclusions [
8], the silver atoms produced by the displacement reaction (Equation (1)) nucleate on the surfaces of the copper particles, forming the foundation of the silver coating. The number of nucleation sites on the copper surface determines the quality of the coating. If there are numerous and evenly distributed nucleation sites, the silver atoms produced by the reduction reaction (Equation (3)) can grow uniformly on the surface of the copper particles, resulting in a coating with uniform thickness and high density. Conversely, if the number of nucleation sites is insufficient, the silver atoms will not be effectively “captured” by the nucleation points, leading to the formation of a discontinuous coating, and a significant amount of free silver particles will appear in the solution.
A clean copper powder surface is essential for the formation of silver nucleation points. Antioxidant oil films and surface oxide layers hinder the displacement reaction, ultimately causing defects in the uniformity of the coating. Therefore, it is crucial to thoroughly clean the copper powder surface with excess alkali and acid during the pretreatment stage to remove oils and copper oxide. As shown in Equations (2) and (3), an alkaline environment promotes the reduction reaction, but it inhibits the displacement reaction (Equation (1)). Hence, maintaining a balanced alkaline environment is vital to ensure the equilibrium between the nucleation process (Equation (1)) and the deposition processes (Equations (2) and (3)).
3.2. Mechanical Stirring Effects
As shown in
Figure 1, the surface morphologies of the Ag-coated Cu powders prepared at different titration rates with only mechanical stirring are illustrated. The dispersibility of the Ag-coated Cu powders prepared at varying titration rates appears similar. In
Figure 1a,d,g, it can be observed that the powder exhibits moderate dispersibility, with slight agglomeration. The scanning electron microscopy images reveal a significant number of incomplete and fractured Ag-coated Cu particles at all titration rates, as indicated by the red arrows in
Figure 1b,e,h. These Ag-coated Cu particles are characterized by surface notches accounting for approximately 10% of the area, with the internal structure being nearly hollow. This indicates the absence of Cu particles, leaving only a silver shell. The formation of this silver-shell structure is likely due to a localized and continuous displacement reaction between Cu and silver ions, leading to the complete dissolution of the Cu particles.
In
Figure 1e,f,i, it is evident that under the condition of mechanical stirring alone, the surface of the silver layer is rough and not dense, indicating uneven silver growth. This result may be attributed to the incomplete and slow distribution of the [Ag(NH
3)
2]
+ solution in the reaction medium under mechanical stirring, causing localized high silver-ion concentrations and resulting in abnormal localized growth. Additionally,
Figure 1b,e,h show that as the titration rate of the [Ag(NH
3)
2]
+ solution increases, the proportion of rough particles gradually rises.
It is also worth noting that with only mechanical stirring and without external heating, the reaction solution temperature remains the same as the bath temperature, approximately 30 °C. This low temperature is also not conducive to the rapid diffusion of [Ag(NH3)2]+.
Figure 2 shows the energy-dispersive spectroscopy (EDS) mapping of the Ag-coated Cu powders prepared under mechanical stirring conditions. From the distribution of copper and silver elements, it can be clearly identified that the small spherical particles attached to the surfaces of the copper particles are silver particles. In addition, the surfaces of the copper particles are not completely covered by silver, and some areas are still exposed.
The reason copper signals are still detected in regions covered by silver particles is that the silver layer consists of a thickness of only several Ag particles, making the coverage relatively thin and, thus, detectable by scanning equipment.
The incomplete or uneven surface coverage of silver on the Ag-coated Cu particles presents two major issues: (1) it fails to meet the requirements for oxidation resistance, and (2) during the preparation of Ag-coated Cu paste, compared with Ag-coated Cu powders with smooth surfaces, more organic solvents are required to achieve the same level of flowability.
Figure 3 presents the oxidation resistance analysis of Ag-coated Cu particles prepared under mechanical stirring conditions. At 150 °C, the weight increment of the Ag-coated Cu powders is very low, with weight increments below 0.5%. Moreover, the weight increment does not increase with prolonged exposure time, indicating that the Ag-coated Cu powders exhibit excellent oxidation resistance at this temperature.
At 300 °C and above, the weight increment of the Ag-coated Cu powders increases rapidly with rising temperature. Under the same temperature conditions, the weight increment also gradually increases with extended exposure time. However, at higher temperatures, such as 600 °C, the weight increment of the Ag-coated Cu powders quickly reaches its maximum value, after which it remains nearly constant and does not further increase over time.
Additionally, the figure reveals that under all temperature conditions, the Ag-coated Cu powders prepared with the fastest titration rate (A3) exhibit the poorest oxidation resistance. This is attributed to the increased proportion of rough Ag-coated Cu particles resulting from the fast titration rate, which, in turn, reduces the overall oxidation resistance of the powders.
3.3. Feeding Rate Effect
Figure 4 shows the Ag-coated Cu powder prepared under the combined effect of ultrasonic vibration and mechanical stirring. From
Figure 4a,c,e, it is evident that the Ag-coated Cu powder prepared with the assistance of ultrasonic vibration has better dispersion than the powder prepared without ultrasonic vibration. The Ag-coated Cu powder prepared under ultrasonic vibration conditions is able to spread uniformly in a single layer, with no agglomeration between the powder layers.
Figure 4b,d,f show that the Ag-coated Cu powder prepared under ultrasonic vibration conditions has a smooth, dense surface, with no hollow shells, surface damage, or incomplete coatings. The effect of different titration speeds on the surface morphology of the Ag-coated Cu is minimal.
Under ultrasonic vibration conditions, some satellite particles with a submicron-level size were observed on the surfaces of Ag-coated Cu particles. Elemental analysis of the satellite particles, as shown in
Figure 5a,b, revealed that their primary element was silver, indicating that these satellite silver particles were formed during the reduction process without adhering to the copper particle surfaces. The formation process of these silver particles is similar to the formation process of the rough surface of Ag-coated Cu particles observed in
Section 3.2. Under the combined effect of ultrasonic vibration and mechanical stirring, most of the silver–ammonia complex ions diffuse rapidly and grow uniformly on the copper particle surfaces. However, a small amount of high-concentration silver–ammonia complexes fail to diffuse in time, leading to the formation of localized high concentrations of silver atoms, which subsequently grow into silver particles. Under mechanical stirring alone, due to poor dispersion, a large number of suspended silver particles are formed, leading to poor nucleation on the copper surfaces. Subsequently, these suspended particles adhere to the copper particles during growth, eventually resulting in the incomplete Ag-coated Cu particles shown in
Figure 1.
Figure 5c–e display the elemental distribution on the surface of the Ag-coated Cu powder prepared using the B1 process, further confirming that the copper particles are uniformly coated with a layer of silver. The satellite particles, indicated by the red arrow in
Figure 5c, are not detected in the copper elemental distribution map but are clearly visible in the silver elemental distribution map, providing further evidence that the satellite particles are pure silver.
Figure 6 shows the oxidation resistance of the Ag-coated Cu powder prepared under the combined influence of ultrasonic vibration and mechanical stirring. Similar to the Ag-coated Cu powder prepared without ultrasonic vibration, the weight increment of the Ag-coated Cu powder at 150 °C is relatively low. In the temperature range of 300 °C to 600 °C, the weight increment increases gradually with both the extension of time and the increase in temperature. When the temperature exceeds 600 °C, the weight increment of the Ag-coated Cu rapidly reaches its maximum within 10 min, and the weight increment does not increase further with additional time. Notably, under the same oxidation conditions, the Ag-coated Cu powder prepared at a low feeding rate exhibits superior oxidation resistance compared with the powder prepared at a high feeding rate.
In order to further analyze the degree of oxidation of the Ag-coated Cu powder at different temperatures, XRD analysis was performed on sample B3 under different oxidation conditions (
Figure 7). Only Ag and Cu elements are detected in the unoxidized Ag-coated Cu powder, and the peaks are very strong and clean. After oxidation treatment at 150 °C for 1 h, no phase change is observed, indicating that the Ag-coated Cu was basically not oxidized at this temperature. Cuprous oxide is detected after oxidation treatment at 300 °C for 1 h, and the peak intensity of copper is significantly reduced, indicating that copper began to oxidize. The phase compositions of the powder after oxidation treatment at 450 °C and 600 °C for 1 h are similar, the peaks of copper and silver are significantly weakened, and the peak of copper oxide is stronger, indicating that a large amount of copper was oxidized. The peak intensity of silver is weakened because it migrated and coalesced at high temperatures to form a discontinuous island structure on the surfaces of the copper particles.
The surface morphology of sample B3 after oxidation at different temperatures for 1 h is shown in
Figure 8. After oxidation at 150 °C for 1 h, as depicted in
Figure 8a,b, the Ag-coated Cu powder exhibits good dispersibility, with no agglomeration observed between particles. The particles show minimal changes compared with their pre-oxidation state. This observation is corroborated by the uniform elemental distribution shown in
Figure 9a.
When the temperature exceeded 300 °C, the Ag-coated Cu powder began to sinter into bulk structures, as shown in
Figure 8c,e,g. With increasing temperature, the sintering necks between particles grow larger, and the porosity of the sintered powder decreases significantly. Notably, the surface morphology of the silver coating on the copper particles starts to change at 300 °C and becomes more pronounced with increasing temperature. At 300 °C, the silver on the copper particles changes from smooth to angular, as shown in
Figure 8d, suggesting an initial tendency for delamination. This is further confirmed by
Figure 9b, which reveals that the particles are no longer spherical while copper remains distributed uniformly across the test area. Combined with the observations of silver at higher temperatures, it can be inferred that the silver layer begins to migrate and coalesce at this temperature, exposing parts of the copper surface.
At 450 °C and 600 °C, high temperatures cause the severe migration and coalescence of the silver coating, which separates into discontinuous clusters attached to the copper particle surfaces.
Figure 9c,d clearly show the discrete distribution of silver. Regions covered by silver exhibit a negligible presence of copper and oxygen, while areas devoid of silver undergo intense oxidation.
Figure 10 presents a statistical analysis of the elemental composition within the region shown in
Figure 9. The curves represent the signal intensity of each element, while the inserted table provides the atomic percentage of each element. Consistent with the previous analysis, the oxidation of the Ag-coated Cu powder is negligible at 150 °C. Slight oxidation is observed at 300 °C, while severe oxidation occurs at temperatures exceeding 450 °C. At these elevated temperatures, the protective function of the silver layer is essentially lost.
3.5. Silver Growth Mechanism
The growth of silver on the surfaces of copper particles is essentially a reduction process. The chemical reaction system investigated in this study includes two reducing agents, potassium tartrate and copper, and one oxidizing agent, silver ions. Ideally, silver ions react exclusively with tartrate ions, resulting in silver particles that attach to the copper surface and gradually grow, while copper remains unreacted.
However, when a system contains two reducing agents, the sequence of redox reactions and the participation of each reducing agent depend on external factors. Typically, the oxidizing agent reacts preferentially with the stronger reducing agent, although side reactions are often unavoidable. Among the two reducing agents in this study, potassium tartrate has stronger reducing properties than copper [
11]. Theoretically, potassium tartrate should primarily engage in redox reactions. Additionally, an excess amount of potassium tartrate was used in the experiments to minimize the involvement of copper. Nevertheless, copper participated in all reactions, as evidenced by the blue color consistently observed in the final solution.
Under the condition of mechanical stirring alone, the formation of a significant number of hollow silver shells indicates extensive participation of copper in the reduction of silver ions, which contradicts the theoretical assumptions. As shown in
Figure 13, the growth of silver on copper particles at different stages and the formation of various surface morphologies are detailed.
In the first stage, when the concentration of tartrate ions is high, the addition of the silver–ammonia solution to the potassium tartrate–copper suspension results in the reaction of silver ions with tartrate ions. The produced silver particles are adsorbed onto the copper particles, forming island-like silver nucleation points on the copper surfaces [
12].
In the second stage, as the reaction progresses, the number of nucleation points increases, and the island-like silver particles grow larger.
In the third stage, the concentration of tartrate ions decreases sharply and becomes insufficient to react quickly with the newly added silver–ammonia solution. Consequently, some silver ions are captured by copper particles and react with the copper, leading to its gradual consumption. This ultimately results in the formation of hollow shell structures, as shown in
Figure 14a.
For copper particles with more surface nucleation points formed in the first and second stages, their surfaces are more completely covered with silver, leaving almost no exposed copper in the solution. These particles are not consumed during the third stage. Conversely, for copper particles with an insufficient silver particle supply, the island-like structures fail to fully overlap and cover the surface, resulting in a rough surface morphology, as shown in
Figure 14b. In contrast, for copper particles with a sufficient supply of silver ions, a smooth and dense silver layer forms on the surface, as shown in
Figure 14c.
When ultrasonic vibration is applied, especially from the bottom of the reaction vessel, it prevents copper powder from accumulating at the bottom for prolonged periods. Under mechanical stirring alone, copper powder tends to aggregate at the bottom, hindering sufficient contact with the silver–ammonia solution introduced from the top, ultimately leading to defective Ag-coated Cu particles.
Ultrasonication improves the dispersion of copper powder in the solution and inhibits the dendritic growth of silver on the copper surface. This reduces the prominence of island-like structures and results in a smoother surface. Additionally, ultrasonication enhances silver nucleation, thereby improving surface quality.