Wetting and the reaction of multiwalled carbon nanotube-reinforced composite solder with a copper substrate

1 Introduction

In electronics technology, lead-free solder research has increased recently as a result of concerns about toxicity and health hazards posed by the lead present in commonly used lead-bearing solders coupled with environmentally friendly market trends for green products, leading to the search for alternative solders that are lead free [1, 2]. With the development and the basic functions of solders having changed, not only do solders have to function in their traditional role as reliable electrical interconnects, lead-free solders must also meet mechanical requirements as well as be reliable [3–7].

In recent years, composite solders have become lead free with improved comprehensive properties [8–10]. Various approaches have been used to fabricate these composite solders, such as in situ methods [11–13], gas atomization [14, 15], powder blending [16–18], and mechanical alloying [19–22]. The reinforcement particles used in the composite approach usually included pure metals and micro- or nanosized metallic, intermetallic, or oxide particles [23–25]. The particles in the composite solders strongly affected the diffusion behavior of Sn in the solder and thus also altered the formation of intermetallic compound (IMC) at the solder/substrate interface.

With the influence of recent nanotechnology development, multiwalled carbon nanotube (MWCNT)-reinforced composite solders have been developed for the first time in this research. The goal of the study was to compare the morphology of the IMC formed around the MWCNT reinforcements in the Sn-Ag-Cu solder so as to gain a better understanding of parameters that contribute to various IMC morphologies. In addition, a detailed study to correlate thermal aging effects on the IMC growth behavior for lead-free solder and composite solder was also done. From the point of view of electronic package reliability, the IMC layer growth should be analyzed when high-temperature storage or extended aging time creates thick IMC layers.

2 Experimental procedure

An Sn-3 wt% Ag-0.5 wt% Cu (SAC305) solder powder was used as the matrix of the composite solder; the nominal size of particles was 25–45 μm. The reinforcement was MWCNT; the volume fraction of the composite solder varied from 0.05% to 1%. Defects on MWCNT surfaces were reduced by ultrasonic-assisted hydrochloric acid treatment for 1 h and then letting the solder rest in an ultrasonic bath for 2 h. Then, an aqueous solution of 5% Triton X-100 surfactant was mixed with the MWCNTs and dispersed by magnetic stirring for 2 h at room temperature. These were subsequently treated with 4 M HNO₃ for 48 h under magnetic stirring and then cleaned with deionized (DI) water to remove the nitric acid residues. Later, they were transferred into a H₂SO₄/H₂O₂ solution (volume ratio of H₂SO₄/H₂O₂=4:1) to oxidize the surface for 0.5 h. Finally, the modified MWCNTs were washed with DI water and then dried at 100°C in an oven for 48 h.

After drying, MWCNTs and solder powder were compacted in a die (cold pressing at room temperature) at 200 MPa to form cylindrical bars 20 mm in diameter; then they with a die directly into a furnace. Subsequently, argon gas was introduced at a flow rate of 2 l/min, and the temperature in the furnace was controlled at 350°C for 1.5 h to achieve melting. After a certain reaction time, the specimen was removed from the furnace and cooled in ambient air. Small composite solder cubic beads (2 mm³) were cut from the cylindrical bars. The beads were cleaned with acetone in an ultrasonic bath. The substrate consisted of Cu plates of 99.9% purity and area 10 mm². A soldering paste composition flux, consisting of 80–90 wt% vaseline, 4–6 wt% paraffin, 6–9 wt% zinc chloride, 1–3 wt% ammonium chloride, and 2–4 wt% water, was formed. Wetting tests were conducted on a hot plate with the temperature maintained between 210°C and 270°C for 10 s. The contact angles and spreading area were measured on solidified samples at room temperature, as shown in the schematic diagram of Figure 1. High-temperature storage aging treatment was performed in an oven at 150°C for periods of 0, 25, 50, and 100 h. Upon completion of the aging step, all the specimens were cut, cold-mounted in epoxy, polished for subsequent preparation of cross sections, and etched with a 94% CH₃OH-4% HNO₃-2% HCl (in vol%) solution. The morphology of the IMCs was observed by means of scanning electron microscopy (SEM). Energy-dispersive X-ray spectroscopy (EDS) analysis of each intermetallic phase was performed.

Figure 1

Schematic drawing of a method for measuring contact angle and spreading area.

3 Results and discussion

3.1 Measurement of contact angle and spreading area

The contact angle and spreading area of the solder alloy on the substrate are important in practical applications. Figure 2 shows the results of contact angle tests in normal atmospheric air for the four types of composite lead-free solder as well as SAC305 lead-free solder. Contact angles ranging from 70° to 129° were observed without use of flux and were significantly higher, ranging from 16° to 23°, than those obtained when flux paste was used. As can be seen, the contact angle decreased as temperature increased from 210°C to 240°C. Both with and without flux paste, a slightly lower contact angle was achieved at 240°C. However, higher contact angles were obtained at 240°C and 270°C. The MWCNT-reinforced composite solder had the lowest contact angle and hence provided better wetting. In addition, either with or without flux paste, the wettability increased with the increase in MWCNT volume fraction, but was not significantly affected when the MWCNT volume fraction was increased from 0.05% to 0.5%. Furthermore, the MWCNT-reinforced composite solder at 0.1 vol% had better wettability than the SAC305 lead-free solder.

Figure 2

Contact angle as a function of temperature for MWCNT-reinforced composite solders on copper substrate.

The spreading areas of SAC305 lead-free solder and composite solders with four different MWCNT additions and temperatures are shown in Figure 3. For all solder alloys, a higher spreading area was observed when flux was used. At 240°C, the MWCNT-reinforced composite solder at 0.1 vol% also had a high spreading area. At the lowest volume fraction of 0.05 vol%, the spreading area was 24.96 mm², whereas at 1 vol% the spreading area decreased slightly to 24.37 mm². The variation in contact angle was almost dependent of temperature for all solder alloys analyzed. It was expected that in all cases the increase in temperature (210–240°C) would result in greater spreading. Moreover, the spreading area decreased as temperature increased from 240°C to 270°C. As noted previously, the MWCNT-reinforced composite solder had the highest spreading area. The spreading area of composite solder was significantly affected when the effective reinforcement volume fraction was between 0.1 and 0.5 vol%.

Figure 3

Spreading area as a function of temperature for MWCNT-reinforced composite solders on copper substrate.

3.2 Microstructural analysis of the interfacial intermetallic layer

Figure 4 shows the SEM images of the cross section of the MWCNT-reinforced composite solder copper interface obtained at a temperature of 240°C. These show the presence of IMCs at the interface. Two types of intermetallics were observed: a darker phase adjacent to the copper substrate (Mark A) and some island-like (Mark B) bigger grains with size 2–3 μm. SEM/EDS analysis revealed that the IMC layer contained C. The intermetallic phase of the two regions could be expressed as (Cu,C)₆Sn₅, reflecting the compositional ratio (6:5) between Cu+C and Sn. The formation of intermetallic layers was caused by the diffusion of copper into the liquid solder. When the boundary layer of the molten solder adjacent to the copper metal became saturated with Cu, the IMCs started to form at the copper substrate.

Figure 4

SEM micrographs of as-solidified solder joints made from (A) SAC305, (B) SAC305 – 0.05 vol% MWCNT, (C) SAC305 – 0.1 vol% MWCNT, (D) SAC305 SAC305 – 0.5 vol% CNT, and (E) SAC305 SAC305 – 1.0 vol% CNT.

3.3 Intermetallic compound growth during isothermal aging

Figure 5 shows the thickness of the reaction layer as a function of aging time at 240°C. In all solder metals, EDS analysis on the reaction layers indicated that they were in the (Cu,C)₆Sn₅ phase. As in the solder joint on the copper substrate, the thickness of the intermetallic increased as the aging time was increased. Initially, the thickness of this IMC was very thin (∼2.5 μm). The thickness increased significantly with longer aging times (100 h), and the corresponding thickness was about 4.2 μm. Figure 6 shows the microstructures obtained at the solder/substrate interface after aging for 50 h at a temperature of 240°C. It can be seen that the reaction layer of the SAC305 lead-free solder was thinner compared with that of the MWCNT-reinforced composite solder. In other words, the copper substrate containing MWCNT-reinforced composite solders generally formed thicker (Cu,C)₆Sn₅ layers.

Figure 5

IMC thickness of Sn-Ag-Cu-MWCNT systems with aging time at 240°C.

Figure 6

SEM micrographs of the interfacial IMC formed at 250°C for 50 h. (A) SAC305, (B) SAC305 – 0.05 vol% MWCNT, (C) SAC305 – 0.1 vol% MWCNT, (D) SAC305 SAC305 – 0.5 vol% MWCNT, and (E) SAC305 SAC305 – 1.0 vol% MWCNT.

Figure 7 represents the SEM micrographs for the interfaces of MWCNT-reinforced composite solders on a Cu substrate during the soldering operation from 230°C to 270°C for 50 h. At this stage, a very thin layer of Cu₆Sn₅ compounds appeared between the solder and the copper substrate. The intermetallic compound thickness gradually increased with the increase in aging time; for MWCNT-reinforced composite solders the growth rate was much higher.

Figure 7

SEM micrographs of the interfacial IMC formed between the SAC305 – 0.1 vol% MWCNT solder and the copper substrate at different aging temperature for 500 h: (A) 230°C, (B) 250°C, (C) 260°C, and (D) 270°C.

Figure 8 shows plots of the average thickness of IMCs at the solder/substrate metallization interface at various temperatures for different aging times. The initial growth rate of the IMCs was high for all the solder systems. For the Sn-Ag3.0-Cu0.5 lead-free solder, the average IMC thickness was less than that of MWCNT-reinforced composite solder. However, from the trend of the curves at four different aging times, it was clear that at higher temperature the IMC growth rate was higher.

Figure 8

The variation of IMC thickness of Sn-Ag-Cu-MWCNT systems with soldering temperature.

4 Conclusion

1. The wetting property improved by 0.05–0.5 vol% with the addition of MWCNT into the SAC305 lead-free solder alloy, whereas excessive addition of MWCNT up to 1 vol% reduced the beneficial influence.

2. MWCNT-reinforced composite solders had a better wetting ability compared to the SAC305 lead-free solder, indicating that the addition of a small amount of MWCNT into the SAC305 lead-free solder helped spreading over the copper substrate surfaces.

3. The flux provided adequate wetting for the copper substrate. Raising the temperature caused a slight improvement in solderability; it deteriorated above 240°C.

4. From the microstructural examination, the reaction between the solder and copper substrate resulted in the formation of a (Cu,C)₆Sn₅ IMC at the interface. During thermal aging, the IMC layer thickness increased with increasing time and temperature.

5. A small amount of MWCNT was added into the Sn-Ag3.0-Cu0.5 lead-free solder to increase the growth rate of intermetallics; meanwhile, the microstructure of a MWCNT-reinforced composite solder showed a thick (Cu,C)₆Sn₅ IMC layer.