Effects of Sn addition in W-doped Ag paste against electrochemical corrosion and sulfurization

Abstract

Purpose

This study aims to address these challenges by enhancing the resistance of Ag-based pastes to corrosion and sulfurization, thereby improving their performance and weatherability in high-power and high-frequency electronic applications.

Design/methodology/approach

This study investigates the influence of Sn doping in W-doped Ag paste to enhance resistance against electrochemical corrosion and sulfurization. A systematic examination was conducted using transient liquid phase sintering and solid–liquid inter-diffusion techniques to understand the microstructural and electrochemical properties.

Findings

This study found that Sn addition in W-doped Ag paste significantly improves its resistance to electrochemical corrosion and sulfurization. The sintering process at 600°C led to the formation of an Ag₂WO₄ phase at the grain boundaries, which, along with the presence of Sn, effectively inhibited the growth of Ag₂WO₄ grains. The 0.5% Sn-doped samples exhibited optimal anti-corrosion properties, demonstrating a longer grain boundary length and a passivation effect that significantly reduced the corrosion rate. No Ag₂S phase was detected in the weatherability tests, confirming the enhanced durability of the doped samples.

Originality/value

The findings of this study highlight the potential of Sn-doped Ag-W composites as a promising material for electronic components, particularly in environments prone to sulfurization and corrosion. By improving the anti-corrosion properties and reducing the grain size, this study offers a new approach to extending the lifespan and reliability of electronic devices, making a significant contribution to the development of advanced materials for high-power and high-frequency applications.

1. Introduction

High-temperature sintering silver paste finds extensive application in discrete electronic components like thermistors, varistors, multi-layer ceramic capacitors and multi-layer ceramic inductors. When contrasted with base metal system such as nickel and copper, the exceptional attributes of silver paste, its unparalleled conductively, reduced equivalent series resistance and elevated resonant frequency remain indispensable to fulfill the evolving requirements of high-power and high-frequency electronic advancements (Harris, 1998; Yin et al., 2010; Komoda et al., 2012). However, it is widely recognized that the functionality of components can be degraded because of various interactions between electrodes and the active layer. Zou et al. (2021) undertook the optimization of a CuO-doped silver paste to increase adhesion strength on AlN ceramic substrate. On the other hand, co-firing process sometimes yields adverse effects on performance because of physical mismatches, bonding energies or other thermodynamic factors. Instances include cracks generation in multi-layer ceramic capacitors and delamination in integrated circuits (Smith et al., 2008; Zhang et al., 2011). Consequently, the co-firing and sintering behaviors of pastes have garnered significant attention, involving sintering conditions, substrates, conductive pastes, particle sizes and doping materials (Wang et al., 2007; Shih et al., 2024; Yonezawa et al., 2008).

Recently, to address joint attachment and enhance the bonding of silver paste, transient liquid phase sintering (TLPS) or solid–liquid inter-diffusion bonding techniques have gained prominence (Jung et al., 2018; Yang et al., 2016). In the sintering process, the filler metal undergoes melting at the bonding temperature, leading to the formation a liquid phase. Consequently, the constituents of the filler metal permeate the base material, establishing a solid–liquid inter-diffusion interface. Tin has been used as a filler metal to enhance the bonding strength of silver paste, because of its low melting point (as low as 235°C) and comparable ionic radius to silver. Huang et al. (2022) reported TLPS Ag-Sn alloy joints exhibited improved shear strength in comparison to conventional techniques. However, it is widely recognized that intermetallic compounds (IMCs) are commonly generated within solid solution composite systems. This stems from the transient liquid Sn phase at melting point or attachment temperature diffusing into a higher melting point Ag matrix, which remains in a solid state. Consequently, this leads to the formation of Ag₃Sn with a higher melting point (as high as 470°C) than the initial Sn reactant (Yang et al., 2016). Notably, Ag₃Sn IMCs have demonstrated an ability to impede the migration of copper and promote the abnormal grain growth (Hsu and Ouyang, 2015). Therefore, in the silver-tin alloy system, tin exhibited a potential to dissolve into silver, thereby enhancing the physical properties of silver.

Given the escalating severity of air pollution, the resistance of silver to sulfidation and corrosion has become an important issue. Palladium and chromium have undergone extensive investigation as passivation additives to protect silver against corrosion or sulfurization (Kim et al., 2013; Jia et al., 2020; Mareci et al., 2010). With attributes like high thermal conductivity, hardness and excellent oxidation/corrosion resistance, Ag-W composites have identified significant potential for electrical contacts in circuit breakers (Kesim et al., 2018). Previous research has explored the use of Ag₂WO₄ compound generated at 600°C in a similar TLPS to fulfill the porosity of W-Ag paste and enhance the sintering effects (Longo et al., 2014; Pan et al., 2013). Because of its elevated chemical stability, the passivating surface layer of Ag₂WO₄ exhibits outstanding characteristic to sulfur-containing compounds. This property inhibits the direct reaction between sulfur species and silver. Based on these corrosion-resistant properties, this study delves into the mechanisms of grain growth and characterizes the effects of Sn addition in W-Ag paste.

2. Materials and methods

The preparation of the Ag-W-Sn pastes involved mixing silver powder, 2.5% tungsten powder and different weight percentages (0.5%, 1.5% and 3%) of tin powder through a triple roll mill. The purity of the silver, tungsten and tin powder used in this study exceeded 99.9%, with an average diameter of 1 µm. Diethylene glycol monobutyl ether and ethyl cellulose were selected as the solvent and thickener. The Sn doped Ag-W pastes were labeled as AgWSn05, AgWSn15 and AgWSn30, representing weight percentages of 0.5%, 1.5% and 3% tin addition, respectively. Following through mixing, the resulting Ag-W-Sn pastes were applied onto a 1 × 4 cm Al₂O₃ substrate using a screen-printing technique, drying in a hot-wind loop oven at a 120°C. Subsequently, the dried Ag-W-Sn samples were sintered in a high-temperature furnace at 600°C, 700°C and 800°C for 10 min. To comprehend the influence of Sn addition, the Ag-W-Sn samples were compared with the 2.5% W doped Ag paste (denoted as AgW25) and pure Ag paste in the following analyses. The thermal expansion properties of the samples were measured using a thermal mechanical analyzer across a range of temperature from room temperature to 900°C. Surface morphology was examined using a scanning electron microscope (SEM, JOEL-7600F) with an accelerated voltage of 15 kV. Crystallinity was determined through monochromatic Cu-Kα radiation (λ = 1.540598 Å) X-ray diffraction (XRD, Bruker D8 Advance), with the results being referred with the Joint Committee on Powder Diffraction Standards (JCPDS) database. Electrochemical polarization curves were conducted to measure the corrosion potential and corrosion current density. A reference electrode of AgCl and a platinum counter electrode were used. Finally, to assess weather resistance, the pure Ag samples, Ag-Pd samples and Ag-W-Sn samples were exposed to sulfurization by placing them 30 cm above a septic tank effluent in Pingtung, Taiwan. Following the sulfur corrosion test, XRD and SEM analyses were conducted again to evaluate the anti-sulfurization properties.

3. Results and discussion

The shrinkage behavior of the pure Ag, AgW25 and AgWSn30 were depicted in Figure 1 within an atmospheric environment, covering a temperature range of 25°C–900°C. Analyzing the slope of shrinkage at specific temperatures, it was evident that pure Ag paste displayed two distinctive shrinkage events. The initial shrinkage transpired at 109°C, attributed to the solvent evaporation in the paste (Park et al., 2008). The subsequent shrinkage at 391°C in the pure Ag paste could be linked to the thermal decomposition of ethyl cellulose (Li et al., 1999). Under these circumstances, particles within the silver pastes underwent a restructuring process, leading to volume reduction and consequent shrinkage. However, the pure Ag paste exhibited expansion when the temperature exceeds 600°C, which associated to the oxidation of silver and the formation of silver oxide (Tsai and Lin, 2013). Given the employment of the same solvent and thickener as used in the pure Ag paste, AgW25 and AgWSn30 pastes also exhibited one obvious shrinkage, indicating the same thermal mechanism as pure silver at 109°C and 117°C. Based on previous research (Shih et al., 2023), the absence of thermal expansion beyond 600°C suggested that the introduction of tungsten slowed down the oxidation of silver; this observation was similarly occurred in the AgWSn30 sample. Because of low melting point of Sn (231.9°C), AgWSn30 sample showed a fast shrinkage than Ag and AgW samples. Beyond this temperature, the molten Sn clad the surface of the high melting point Ag, giving rise to a TLPS. Besides, a minor expansion in the AgWSn30 sample at 270°C probably was attributed to the oxidation of Sn alloy; the phase of SnO₂ will be confirmed in XRD analysis.

To investigate the impact of W and Sn additions, the microstructure of pure Ag, AgW25, AgWSn05, AgWSn15 and AgWSn30 samples sintered at 800°C were examined using SEM as shown in Figure 2. The Ag-W samples appeared to have a larger grain size and greater density than the typical pure Ag sample. This phenomenon can be attributed to the liquefaction of α-Ag₂WO₄, resulting in the coverage of Ag particles by α-Ag₂WO₄ phase (Shih et al., 2023). Three distinct microstructures were observed on the surface of AgWSn05. The first microstructure exhibited the standard silver grain formation, bearing a resemblance to the surface morphology of both pure Ag and AgW25 samples. The other micro-structures, such as prismatic micro-rods, were observed in AgWSn05 and AgWSn15 samples, as shown in Figures 2(c) and 2(e). The third micro-clusters were present in both AgWSn15 and AgWSn30 samples, as shown in Figures 2(e) and 2(f), . To investigate the elemental compositions of the micro-structures, energy-dispersive X-ray spectroscopy (EDS) was performed. At point A in Figure 2(c), the primary elements ratio of prismatic micro-rods is Ag:W:O = 22.3%:11.3%:66.4%. At point B in Figure 2(d), the primary elements ratio of prismatic micro-rods is Ag:W:O = 30.6%:13.5%:55.8%. The compositions of prismatic micro-rods correspond to elemental ratios of Ag₂WO₄ (Andrés et al., 2014). Besides, the elemental ratios of micro-clusters are Ag:Sn:O = 24.7%:12.1%:61.3%, partially contributing by SnO₂ and would be confirmed in XRD analysis. Figure 3 indicates the EDS mapping of AgWSn05 sample sintered at 800°C. From the mapping images of Sn and W, it is evident that Sn and W are roughly localized at specific positions, corresponding to the grain boundaries. The elemental distribution at grain boundary of tungsten was also confirmed in the other study (Shih et al., 2023). This observation plays a role in the grain growth of Ag pastes and the sintering mechanisms.

Figure 4 presents the gain size distribution of the pure Ag, Ag-W and Ag-W-Sn samples at varying sintering temperatures, which were calculated from the SEM images using the “Image J” software. The average gain sizes of pure Ag and AgW paste are 7.9 µm and 13.1 µm; the increase in grain size is attributed to grain growth enhancement of Ag₂WO₄. Moreover, a notable observation is the significant reduction in grain size upon Sn addition. The average grain sizes were reduced to 3.29 µm, 3.85 µm and 4.26 µm for 0.5%, 1.5% and 3% Sn addition. This indicates that Sn addition would decrease the grain size of pastes, and the reduction values in grain size would decrease with the increase in Sn composition. This phenomenon will be discussed with XRD results.

The XRD patterns for pure Ag, Ag-W and Ag-W-Sn specimens were presented in Figure 5. The principal phases, Ag(111) and Ag(200), were verified based on JCPDS card number 04–7803. Figure 5(a) demonstrates the presence of a secondary phase in the Ag-W and Ag-W-Sn samples, corresponding to the planes of α-Ag₂WO₄ (002), (231), (400), (111), (352) and (550) as defined by JCPDS card number 34–0061 (Cavalcante et al., 2012). According to EDS analysis, these α-Ag₂WO₄ phases correspond to the composition of the flakes and prismatic rods observed on the B sites. The crystallization of α-Ag₂WO₄ was associated with the bonding of Ag, W and O atoms at temperatures below 600°C (Andrés et al., 2014). The reaction formula can be expressed in equation (1): (1) $$2\text{Ag}+\text{W}+2{\text{O}}_2\stackrel{}{\to }{\text{ Ag}}_2{\text{WO}}_4$$

Upon elevating the sintering temperature to 700°C and 800°C, the excessive heat induced the transformation of crystalline α-Ag₂WO₄ into a non-crystalline liquid state – a secondary TLPS occurrence. Simultaneously, this liquid α-Ag₂WO₄ phase initiated capillary action and wetted the contacting Ag grains, consequently prompting particle rearrangement and densification (Huang et al., 2022). This phenomenon led to the disappearance of most α-Ag₂WO₄ peaks in AgW25 sample, while an increase in these peaks was observed in all the Ag-W-Sn samples, as depicted in Figures 5(b) and 5(c), . Furthermore, the XRD patterns of AgWSn15 and AgWSn30 samples revealed the presence of SnO₂ (110) and (101) planes, in correspondence with JCPDS card number 41–1445. This result is consistent with the elemental composition of the white clusters located on the C sites. But no SnO₂ phase was observed at 600°C, indicating rapid oxidation of Sn as sintering temperature beyond 700°C (Wittmer et al., 1981; Leitner and Sedmidubský, 2019). In comparison to AgW25, the (002), (400) and (402) peaks of α-Ag₂WO₄ vanished in Ag-W-Sn samples disappeared at 600°C. However, these peaks appeared at 700°C, suggesting that the presence of emerging SnO₂ increased the required sintering temperature. This occurrence can be attributed to the interaction between liquid α-Ag₂WO₄, Ag particles and SnO₂, resulting in an unconventional TLPS process. Based on the findings from a previous study (Cavalcante et al., 2012), XRD peaks of α-Ag₂WO₄ are observed at 600°C, but they diminish at 700°C and 800°C within the Ag-W binary system. Above the melting point of α-Ag₂WO₄ (600°C), its crystalline structure transitions to an amorphous state, resulting in the disappearance of XRD peaks. However, within the Ag-W-Sn ternary system, the presence of SnO₂ partially obstructs grain growth and raises the sintering temperature. Consequently, crystalline α-Ag₂WO₄ can persist at 600°C and 700°C, but they diminish at 800°C because of amorphous transformation at high temperature. In addition, the XRD patterns did not reveal any IMC, such as Ag₃Sn (Kim et al., 2002; Henderson et al., 2002). This could be attributed to the high sintering temperatures exceeding the melting point, which probably led to the decomposition of Ag₃Sn (Sun et al., 2022).

The resistivity for pure silver, Ag-W and Ag-W-Sn samples sintered at varying temperatures are illustrated in Figure 6. Similar with ordinary commercial silver paste (Wang et al., 2018), the resistivities of pure silver samples are 5.05 µΩ-cm, 4.66 µΩ-cm and 4.49 µΩ-cm at 600°C, 700°C and 800°C, respectively. The resistivity of the Ag-W-Sn samples at 800°C are measured as 6.46 µΩ-cm, 6.61 µΩ-cm and 8.31 µΩ-cm for 0.5%, 1.5% and 3% Sn addition, respectively. The trend of the curves demonstrates an increase in resistivity with rising Sn content because of the emergence of SnO₂. As an industrial practice to prevent silver from sulfidation, Pd doped silver paste is typically used. In this regard, a comparison of resistivities is drawn between the Ag-W-Sn samples and Pd doped (0.5% and 5%) Ag paste (denoted by AgPd05 and AgPd50). The resistivities of AgPd05 and AgPd50 at 800°C give 4.5 µΩ-cm and 21.49 µΩ-cm, respectively. While the resistivities of Ag-W-Sn samples increased because of Sn oxidation, all Ag-W-Sn samples exhibit superior resistivity in comparison to the commercial Ag-Pd samples. Lower resistivity represents that Ag-W-Sn samples offer enhanced conductivity and reduced heat accumulation.

Figure 7(a) displays the Tafel chemical potential polarization curves measured in a 0.001 M Na₂S solution (pH = 2.7) for the pure Ag, Ag-Pd and the Ag-W-Sn samples. The corresponding corrosion potential (E_corr) and corrosion current density (I_corr) are systematically depicted in Figures 7(b) and 7(c), . For the pure Ag, its E_corr and I_corr stand at −620 mV and 2.87 µA/cm², respectively. The commercial AgPd50 sample shows a higher E_corr and smaller I_corr than pure Ag, recording −119 mV and 0.4 µA/cm². All the Ag-W-Sn samples exhibit E_corr surpassing those of pure Ag and AgPd samples, indicative of a superior anti-corrosion capability. At 800°C, the E_corr of Ag-W-Sn samples measure 139.5 mV, 120.1 mV and 41.5 mV for 0.5%, 1.5% and 3% Sn additions, respectively. Notably, AgW25 possesses the highest E_corr value of 160 mV. However, it also presents the highest I_corr value of 4.9 µA/cm², as shown in Figure 7(c). Previous studies concluded that secondary phase of α-Ag₂WO₄ at grain boundaries effectively counters chemical corrosion. But when the α-Ag₂WO₄ being corroded, the exposed Ag grains face rapid corrosion, consequently increasing I_corr (Kim et al., 2003). However, the incorporation of Sn and subsequent SnO₂ clusters improved this shortage. Sn addition eventually inhibited grain growth, leading to smaller grain size and extended grain boundary length. This increase in grain boundaries led to more α-Ag₂WO₄ layers, ultimately delaying the erosion of inner Ag grains. Consequently, the I_corr values for Ag-W-Sn samples sintered at 800°C experienced reductions of 0.87 µA/cm², 2.19 µA/cm² and 2.21 µA/cm² for AgWSn05, AgWSn15 and AgWSn30, respectively. Among the Ag-W-Sn samples, those sintered at 800°C demonstrated superior performance compared to the samples sintered at 700°C. Especially AgWSn05 sintered at 800°C exhibited highest E_corr and the lowest I_corr. When compared to AgW25, despite a slight 12.8% decrease in E_corr of AgWSn05, the corrosion rate I_corr significantly diminished by a factor of 4.7. Consequently, the AgWSn05 sample was selected for further analysis and experimentation.

The grain boundary length of AgWSn05 in SEM images (2.5 mm²) was calculated using Image J version 1.54d software. The grain boundary lengths, E_corr and I_corr, as a function of Sn content were summarized in Figure 8. It is evident that with a 0.5% Sn addition, the grain boundary length significantly increased. This is attributed to the presence of SnO₂, which inhibits Ag grain growth. This observation is consistent with the SEM images, showing a rise in grain boundaries length from 3.4 mm in AgW25 to 13.86 mm in AgWSn05. The curves in Figure 8 reveal a negative correlation between grain boundary length and I_corr. Basically, the generation of Ag₂WO₄ at the grain boundaries establishes a passivation layer that withstands electrochemical corrosion (Mahesh and Raman, 2014; Ralston et al., 2010). Although AgW25 sample exhibits higher E_corr value, fewer grain boundaries of Ag₂WO₄ passivation led to an increase in I_corr because of the shorter grain boundary length. When the Ag₂WO₄ surface corrodes, their interior Ag quickly undergoes corrosion. However, the growth of grains in Ag-W-Sn samples is obstructed by the presence of Sn/SnO₂ clusters; this observation will be further discussed in Figure 9. This results in longer grain boundaries and a greater number of Ag₂WO₄ passivation layers, enhancing resistance against electrochemical corrosion (Kim et al., 2003; Mahesh and Raman, 2014).

The comprehensive sintering mechanism of Ag-W-Sn samples has been delineated in Figure 9. Initially, Sn, W and Ag powders, each with an average diameter of 1 µm, were uniform. The thermal mechanical analyzer results indicated that at a temperature beyond 270°C (exceed tin’s melting point), Sn underwent liquefaction, contributing to a minor thermal expansion as depicted in Figure 1. Concurrently, the molten Sn clad the surface of Ag particles, initiating the first TLPS and promoting grain growth. As the sintering temperature was elevated to 600°C, some α-Ag₂WO₄ segregated at the grain boundaries. However, the growth of the grains was somewhat blocked by Sn or SnO₂ clusters. This led to the absence of α-Ag₂WO₄ microstructure and weak XRD intensity, as depicted in Figure 5(a). At 700°C, the rapid oxidation of Sn swiftly pinned the expansion of Ag grain. This was evident through the proliferation of SnO₂ in both SEM images and XRD patterns. The growth of crystalline α-Ag₂WO₄ consequently became anchored at the grain boundary. Moreover, SnO₂ clusters also contributed to the limitation of Ag gain size, which is in an agreement with the gain size calculations in Figure 4. According to Zener’s theory (Nes et al., 1985), which considers only interfacial energies, coherent particles, such as precipitates (Wang et al., 2017) or oxide, are more effective in pinning grain boundary migration. Furthermore, previous studies (Shih et al., 2023) on Ag-W pastes have indicated that α-Ag₂WO₄ forms at the grain boundary without any microstructure. Additionally, analysis of SEM and XRD results reveals that SnO₂ at grain boundaries partially inhibits Ag grain growth and raise the sintering temperature. Consequently, it can be inferred that SnO₂ contributes to the formation of α-Ag₂WO₄ micro-rods, potentially pinning grain boundaries and impeding grain growth in directions parallel to the grain boundaries (Gusak et al., 2021; Pei et al., 2020), as shown in the SEM image of Figure 2(c).

The resistant qualities of various samples against sulfurization were tested over a period of 96 h (Kesim et al., 2018; Braun et al., 2012). The tested samples include pure Ag, commercial AgPd05, AgPd50, AgW25, AgWSn05 and AgWSn30. The XRD results following the weatherability experiments are depicted in Figure 10. In both the pure Ag and AgPd05 samples, an Ag₂S monoclinic phase emerges, as indicated by the 14–0072 JCPDS card. The presence of this phase illustrates that not only pure Ag but also AgPd05 fail to endure the corrosion induced by sulfurization. Conversely, the absence of Ag₂S phase in AgPd50 AgW25, AgWSn05 and AgWSn30 samples exhibits their effective resistance against sulfurization. The SEM images in Figure 11 support our findings consistently. Corrosion pits with diameters ranging from 200 nm to 600 nm can be observed at the grain boundaries of pure Ag and AgPd05 samples, as indicated by the solid arrows in Figures 11(a) and 11(b), . Similar corrosion pits have been reported in W-implanted TiN coatings (40), suggesting that sulfurization initiates at the grain boundaries, where H₂S tends to accumulate at these positions. Furthermore, because of the higher surface energy at the grain boundaries, silver atoms are more prone to ionization compared to those within the grains. Consequently, some grain boundaries appear blurred because of sulfur-induced corrosion, as indicated by the hollow arrows in Figures 11(a) and 11(b), . In contrast, no such pitting is observed on the surface of AgW, AgPd50, AgWSn05 and AgWSn30 samples [Figures 11(c)–(f)]. While some pores are visible in Figures 11(c)–(f), they were naturally formed during the sintering process, as explained in results and discussion section. Pd doping in AgPd50 and the formation of SnO₂ in AgWSn30 inhibited grain growth, resulting in smaller grain sizes and a more porous surface morphology. Moreover, the diameter of these pores (2 µm –7 µm) is significantly larger than that of the corrosion pits, indicating a different formation mechanism. The grain boundaries in Figures 11(c)–(f) also appear clear and continuous, demonstrating good anti-corrosion properties. The protection is attributed to the covalent bonding of Ag₂WO₄, which resists dissociation or sulfidation of Ag grains. Moreover, the formation SnO₂ also increases grain boundaries length of Ag₂WO₄, improving sulfidation resistance. Consequently, Ag grain shielded by Ag₂WO₄ remains well-protected. Ag-W-Sn samples exhibit comparable resistance to sulfurization in comparison to commercial AgPd50 samples.

4. Conclusion

The influence of Sn addition in W doped Ag paste has been thoroughly examined and found to be significantly enhance resistance against both electrochemical corrosion and sulfurization. The sintering process facilitated the segregation of Sn, effectively coating the Ag particles upon reaching the melting point. At 600°C, a transformation from secondary phase Ag₂WO₄ to a non-crystalline liquid state occurred at the Ag grain boundary, followed by a second TLPS causing Ag₂WO₄ segregation. However, beyond 600°C, rapid Sn oxidation led to the grain boundary pinning of Ag₂WO₄. Subsequently, the obstruction of SnO₂ on Ag₂WO₄ growth resulted in smaller grain size compared to the Ag-W samples. Analysis of chemical potential polarization curves further revealed the exceptional electrochemical corrosion resistance of Ag-W-Sn samples, because of the presence of Ag₂WO₄ passivation layers. Additionally, the longer grain boundaries length of Ag₂WO₄ significantly decreased the corrosion rate. Subsequent weatherability experiments verified the absence of Ag₂S phase in XRD patterns, and SEM images displayed clear surface without any corrosion induced hole. In summary, the Ag-W-Sn samples demonstrated their superior properties for countering sulfurization and corrosion. Furthermore, Sn addition successfully improved the shortage of high corrosion rate in Ag-W samples. Among the Ag-W-Sn samples, the 0.5% Sn addition exhibited optimal anti-corrosion properties, presenting a promising way for enhancing the durability and lifespan of electronic components.

Funding: This work was supported by the National Science and Technology Council (NSTC), Taiwan through grant number 111–2221-E-110 –064 -MY3.

Declaration of competing interest: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Figure 1

Thermomechanical analysis for pure Ag, AgW25 and AgWSn30 samples, ranging from 25°C to 900°C

Figure 2

Surface morphology for (a) pure-Ag, (b) AgW25, (c) AgWSn05, (d) AgWSn15 and (e) AgWSn30 samples sintered at 800°C

Figure 3

EDS mapping for AgW paste doping with 0.5% Sn sintered at 800°C

Figure 4

The grain size as a function of Sn content for the samples sintered at 600°C, 700°C and 800°C

Figure 5

X-ray diffraction patterns for Ag, AgW25, AgWSn05, AgWSn15 and AgWSn30 sintered at (a) 600°C, (b) 700°C and (c)800°C

Figure 6

Resistivity for pure Ag, Ag-Pd and Ag-W-Sn samples sintered at (a) 600°C, (b) 700°C and (c) 800°C

Figure 7

(a) Chemical potential polarization curves for pure Ag, AgW25, Ag-Pd samples and Ag-W-Sn samples sintered at 800°C; (b) corrosion potential as a function of doped concentration; and (c) corrosion current density as a function of doped concentration of the samples

Figure 8

Grain boundary length, i_corr, e_corr as a function of Sn content for the samples sintered at 800°C

Figure 9

Schematic diagrams depict sintering mechanism of Sn addition in W-doped Ag paste

Figure 10

X-ray diffraction patterns for pure Ag, AgPd samples, AgW25 and Ag-W-Sn samples after 96 h of sulfurization test

Figure 11

Scanning electron microscope images for (a) pure Ag, (b) AgPd05, (c) AgPd50, (d) AgW (e) AgWSn05 and (f) AgWSn30 samples after 96 h of sulfurization test.