Effects of Zinc Oxide Nanoparticles on Properties of SAC 0307 Lead-Free Solder Paste

*is research investigates the effects of zinc oxide (ZnO) nanoparticles of varying concentrations 0.0, 0.25, 0.50, 0.75, and 1.0wt.% on the melting temperatures, wettability, printability, slump, and interfacial microstructure of the ZnO-doped Sn-0.3Ag-0.7Cu lead-free solder pastes on the copper substrate. *e results revealed that the introduction of the ZnO particles had no effect on the solidus and liquidus temperatures of the solders. *e maximum wettability was achieved with 0.25 wt.% ZnO nanoparticles, while the printability was inversely correlated with the nano-ZnO concentrations.*e findings also indicated that, at room temperature, the slumping and the nano-ZnO concentrations were positively correlated and that, under the 150°C thermal condition, the maximum slumping was achieved with 0.25wt.% ZnO. *e slumping mechanism of the SAC0307-xZnO solder pastes is also provided herein. Moreover, the experiments showed that Cu6Sn5 was the single intermetallic compound present in the interfacial layer between the solders and the copper substrate, with the maximum intermetallic layer thickness realized at the 0.25wt.% ZnO concentration.


Introduction
Recent decades have witnessed a growing number of countries around the globe rigorously regulating the use of certain toxic substances in electronic products.Effective over a decade ago, the European Union's Restriction of Hazardous Substances (RoHS) directive restricts the use of six hazardous materials in the manufacture of various types of electronic and electrical equipment, including lead (Pb), mercury (Hg), cadmium (Cd), hexavalent chromium (Cr 6+ ), polybrominated biphenyls (PBB), and polybrominated diphenyl ether (PBDE) [1].To comply with the RoHS directive, a variety of lead-free solders have thus been developed to replace the lead-bearing solders.In fact, the most commonly used lead-free solders belong to the Sn-Ag-Cu (SAC) group because of their functional electrical and mechanical properties [2].
By comparison, Sn-0.3Ag-0.7Cu(SAC0307) solder is a low-Ag SAC solder that is of lower cost and less Ag 3 Sn in the solder matrix [3,4].e SAC0307 and other SAC solders nevertheless are plagued with low fatigue strength and creep resistance [5][6][7][8].To mitigate, certain small particles were introduced into the solder to transform it into a composite material such that the movement of dislocations and the grain boundary slidings were hindered [6,7,9].Importantly, with the advancement in the nanotechnology, many new nanoparticles have been synthesized and used as the dispersed phases in the solders (i.e., the nanocomposite solders) to enhance their mechanical properties [10].However, the improvement in the fatigue strength and creep resistance comes with the alteration of the other properties of the solder, such as wettability and microstructure [11][12][13][14][15][16][17][18][19].
Moreover, in the case of solder paste, printability and slump are the additional characteristics required for further investigations [20][21][22][23][24].Although the slump of solder paste is very critical to the formation of solder bridges leading to the failure of the circuit, the slumping mechanism has not been reported in the literature.
is research investigates the e ects of the ZnO nanoparticles of variable concentrations (i.e., 0.0, 0.25, 0.50, 0.75, and 1.0 wt.%) on the melting temperatures, wettability, printability, slump, and interfacial microstructure of the SAC0307-xZnO solder pastes on the copper substrate.e slumping mechanism is also provided herein.

Materials and Methods
In this research, the SAC0307 solder paste and the ZnO nanoparticles were used as the matrix and dispersed phases of the nanocomposite solder, respectively.e SAC0307 solder paste was prepared from the Sn-0.3Ag-0.7Cusolder paste (Nihon Almit) with the metallic particle sizes of 20-38 μm. e solder's melting temperature, density, and viscosity were 217-227 °C, 4.16 g/cm 3 , and 150-300 Pa•s, respectively.Meanwhile, the concentrations of the ZnO nanoparticles (SAC0307-xZnO; NanoMaterials Technology) in the solder paste were varied between 0.0, 0.25, 0.50, 0.75, and 1.0 wt.%. e nano-ZnO particle size, the melting temperature, and the density were 20-40 nm, 1970 °C, and 5.61 g/cm 3 .Moreover, the solder paste and the ZnO nanoparticles were mechanically mixed for 30 min for uniform mixture.
e melting (solidus and liquidus) temperatures of the SAC0307-xZnO solder pastes were determined using the di erential scanning calorimetry technique (Netzsch DSC 204 F1 Phoenix) under the N 2 atmosphere and 10 °C/min heating rate condition.In addition, the SAC0307-xZnO solder pastes were individually screened on a copper substrate according to the JIS Z3198-3:2003 standard and then re ow-soldered in an LPKF re ow oven with the re ow pro le as per the JEDEC J-STD-020D.1 standard.e preheat temperature and time were 180 °C and 80 s, while the re ow temperature, re ow time, and cool down time were 267 °C, 210 s, and 60 s. e wettability, printability, slump, and intermetallic layer were subsequently examined and compared.In this research, the copper substrate was prepared from an oxygen-free high conductivity (OFHC) copper sheet with a thickness and surface roughness (R a ) of 0.35 mm and 0.06 μm. e wettability was assessed in terms of contact angle (θ), as shown in Figure 1 and (1): e printability is the ratio of the actual volume of solder paste on the substrate to the ideal volume of solder paste on the substrate [31].e printability (P%) can be calculated by (2), where V 1 is the actual volume of the solder paste and V 0 is the ideal volume of the solder paste: e slump testing was undertaken under two thermal conditions in accordance with the IPC-TM-650 testing procedure, each for 30 min: 25 °C (at room temperature) and 150 °C.e higher temperature testing (150 °C) was carried out using a Memmert hot air oven.e slump (S L %) could be calculated by (3), where D s is the diameter of the stencil and D p is the diameter of the pattern [32]: Furthermore, the re owed solder joints were sectioned and the interfacial microstructure examined using the Bruker XRD model D8-Discover and Hitachi eld emission scanning electron microscope (FE-SEM) coupled with IXRF Systems energy dispersive X-ray spectroscopy (EDS) to characterize the intermetallic layer.In this research, Image Pro ® Express was used to determine the area and length of the intermetallic layer, and the intermetallic layer thickness was calculated by dividing the area of the intermetallic layer by its length.e reported thickness was the average thickness of three specimens of identical ZnO concentrations.e nding could be explained by Lindemann's melting theory, in which the solidus and liquidus temperatures of a single material are the inherent properties that are dependent on the interatomic distance and the root mean vibration amplitude [33].It could thus be extrapolated that the presence of the ZnO nanoparticles neither alters the aforementioned parameters nor reacts with the solder.In other words, the introduction of the ZnO nanoparticles had no e ect on the solidus and liquidus temperatures of the SAC0307-xZnO solder pastes.

Wettability on the Copper Substrate.
e wettability of SAC0307-xZnO solders on the copper substrate was measured in terms of contact angle.In Figure 3, the contact angle of the initial SAC0307 solder was 22.24 ± 0.61 °and became drastically smaller with 0.25 wt.% ZnO nanoparticles.e trend was nevertheless reversed as the nano-ZnO concentrations increased.e phenomenon could be attributed to the reduced surface tensions of the molten solder in the presence of ZnO nanoparticles in which the liquid-solid and liquid-vapor surface tensions decreased [34].Meanwhile, the higher nano-ZnO concentrations elevated the viscosity of the molten solder, resulting in the upsurge in the contact angle [35][36][37][38].e nding suggested that the solder contact angles could be manipulated by varying the nano-ZnO concentrations.With 0.25 wt.% ZnO nanoparticles, the contact angle decreased dramatically before gradually increasing with increase in ZnO.Given the minimum surface tensions, 0.25 wt.% was thus the optimal ZnO concentration.Meanwhile, excessive fractions of ZnO nanoparticles did not reduce the surface tensions but increased the viscosity of the molten solder.
Interestingly, the low ZnO concentrations (0.0-0.25 wt.%) warrant further investigation to establish the relationship between the contact angle and ZnO concentrations.

Printability on the Copper Substrate.
In Figure 4, the printability of the SAC0307 solder paste on the copper substrate was 92.87 ± 0.56 % and steadily decreased as the nano-ZnO concentrations increased.Speci cally, the ZnO concentrations were inversely correlated with the printability of the solder paste on the copper substrate.is is because the higher ZnO concentrations caused the solder paste to become more viscous.e stickier solder paste in turn contributed to excessive solder paste residues clinging to the stencil and the subsequent lower printability.

Slumping on the Copper Substrate.
e slump testing of the experimental SAC0307-xZnO solder pastes on the copper substrate was carried out at room temperature (25 °C) and at 150 °C for 30 min each.Figure 5 shows the experimental results of slumping tests of SAC0307-xZnO solder pastes at both testing temperatures.Figure 6 illustrates the slumping results at room temperature, in which the initial slumping associated with the SAC0307 solder paste was 12.17 ± 0.76 % and increased with the increase in the nano-ZnO concentrations.Speci cally, the solder became more susceptible to collapse and its diameter larger as the nano-ZnO concentrations rose.
e nding was attributable to the absorption of the moisture or liquid in the solder paste by the ZnO nanoparticles because the absorbed moisture or surrounding liquid lowers interparticle friction [39].As a result, the nanoparticles became more free-owing and less coalescent with other particles.Given that the amount of moisture or  Advances in Materials Science and Engineering liquid was su cient, the solder continued to collapse providing the greater slump as the nano-ZnO concentration increased.Figure 7 depicts the slumping mechanism of the SAC0307-xZnO solders on the copper substrate at room temperature.
Under the 150 °C thermal condition, the slumping of the SAC0307-xZnO solders was greater than that at room temperature.In Figure 8, the initial slumping associated with the SAC0307 solder paste was −99.50 ± 0.50% and increased to ± 136.50 ± 1.00% with 0.25 wt.% ZnO nanoparticles.e slumping trajectory was nevertheless reversed as the nano-ZnO increased.
In fact, with 0.25 wt.% ZnO, the slumping was greater than that of the SAC0307.e nding was attributable to, despite the evaporation, the su ciently large amounts of remaining liquid in the solders which allow for the freeowing of the low-concentration nanoparticles and the subsequent larger slump.
e slumping however became smaller as the nano-ZnO concentrations increased.is was due to the liquid bridge phenomenon.
Unlike the slumping mechanism at room temperature (Figure 7), the slumping behavior under the 150 °C condition was characterized by the evaporation of liquid in the solder paste.Given the evaporation and the higher nano-ZnO concentrations, the proportion of liquid to particles decreased and the solder pastes were transformed from droplet or capillary state to funicular or pendular state.e funicular or pendular state solder pastes exhibited the greater attractive forces between particles [40], resulting in less slumping.Figure 9 illustrates the slumping mechanism of the SAC0307-xZnO solders on the copper substrate at 150 °C.

Interfacial Layer.
e interfacial layer between the SAC0307-xZnO solders and the copper substrate was of Cu 6 Sn 5 intermetallic compound of varying thicknesses.2.5 Solder alloys  e increased thickness of the Cu 6 Sn 5 layer was attributable to the ZnO nanoparticles acting as the new nucleation sites for the formation of intermetallic phase, resulting in a thicker Cu 6 Sn 5 layer, consistent with Peng et al. [28] and Xing et al. [41].However, at higher nano-ZnO concentrations, the Cu 6 Sn 5 layer thickness decreased.is was attributable to the nanoparticles, despite acting as the nucleation sites, restricting the growth of the intermetallic compound by lowering the surface energy.From the adsorption theory [34], the surface energy of Cu 6 Sn 5 crystal can be written as follows:

􏽘
where c k c is the surface energy per unit area of the kth crystal plane of Cu 6 Sn 5 with the adsorption of ZnO nanoparticles, A k is the area of the kth crystal plane, c k 0 is the surface energy per unit area of the kth crystal plane of Cu 6 Sn 5 without the adsorption of ZnO nanoparticles, R is the gas constant, T is the absolute temperature, Γ k is the amount of ZnO nanoparticles adsorbed per unit area of the kth crystal plane, and c is the total concentration of ZnO nanoparticles.
From (4), the surface energy of Cu 6 Sn 5 is minimized when erefore, the surface energy of Cu 6 Sn 5 decreased with increase in the ZnO nanoparticles adsorbed on the crystal.e lower surface energy in turn contributed to inferior Cu 6 Sn 5 grain growth.Although the nanoparticles acted as the nucleation sites, the reduced surface energy restricted the Cu 6 Sn 5 layer thickness.is was more pronounced at higher nano-ZnO concentrations.
Figures 11 and 12 illustrate FE-SEM images and EDS analysis of the solder joints between SAC0307-xZnO and the copper substrate.e experimental results showed that Cu 6 Sn 5 was the only intermetallic compound present in the scallop-shaped interfacial layer.
e Cu 6 Sn 5 intermetallic compound could be found at the interfacial zone and in the solder matrix, as shown in Figure 11(a).Ag 3 Sn intermetallic phase was also present.Despite the ZnO nanoparticles, the nanoparticles were undetectable by FE-SEM; however, they were detected by EDS and XRD in the intermetallic layer and the solder matrix (Figures 12(a)-12(d) and 13).e retained ZnO nanoparticles were positively correlated to the Zn concentration.Zn in the intermetallic layer and the solder matrix increased (i.e., 0.10 to 0.17 wt.%) in response to the increase in ZnO doped into the solder paste.
ese numbers were the ZnO nanoparticles remaining in the solder paste following the expulsion during the reflow process [28].

Conclusions
is research investigated the effects of ZnO nanoparticles on the melting temperatures, wettability, printability, slump, and interfacial microstructure of the SAC0307-xZnO leadfree solder pastes on the copper substrate.e experimental results are summarized below: (1) e addition of ZnO nanoparticles had no effect on the solidus and liquidus temperatures of the solders.(2) e ZnO nanoparticles improved the wettability of the solder pastes, and the maximum wettability was achieved with 0.25 wt.% ZnO nanoparticles.(3) e printability of the solders on the copper substrate decreased with increase in the nano-ZnO concentration.(4) At room temperature, the slumping decreased with increase in the nano-ZnO concentration.Meanwhile, at 150 °C, the maximum slumping was achieved with 0.25 wt.% ZnO concentration.(5) e maximum thickness of Cu 6 Sn 5 intermetallic layer was realized with 0.25 wt.% ZnO.

6
Figure10compares the intermetallic layer thicknesses at the various nano-ZnO concentrations.e intermetallic layer thickness of the SAC0307 solder was 3.31 ± 0.20 μm and increased with the introduction of 0.25 wt.% ZnO nanoparticles but then decreased as the nano-ZnO concentrations increased.However, the di erence in the thicknesses was marginal among the SAC0307-xZnO solders.Figures11(a) and 11(b) compare the intermetallic layers associated with SAC0307 and SAC0307-1.0ZnOsolders on the copper substrate.