Nanoparticles of SnAgCu lead-free solder alloy with an equivalent melting temperature of SnPb solder alloy
Introduction
Electronic packaging is a manufacturing technology used for electronic products. Packaging provides a medium for electronic interconnections and mechanical support, and solder alloys provide the electrical and mechanical connections between the die (chip) and the bonding pads. The selection of materials for solder alloys is, therefore, critical and plays an important role in solder joint reliability. The Sn–Pb alloy has been the most widely used solder alloy system as an interconnection material in the electronic packaging industry. The Pb-containing alloys are reliable, well tested and quite inexpensive, and Sn–Pb alloys over the whole composition range can be used as solders. There are, however, disadvantages with Pb-containing solders. Apart from the undeniable toxicity of damaging human's nervous system, Pb is also harmful to the environment by causing ground water contamination. As a result, Pb-containing solder alloys are either being banned or phased out from the electronic industry. Legislations, such as the Waste from Electrical and Electronic Equipment (WEEE) and the Restriction of the use of Hazardous Substances (RoHS) have been implemented in the European Union since 2006. Therefore, an urgent necessity exists for more appropriate substitutes for the Sn–Pb solders.
It is also important to ensure that the physical and mechanical properties of the substitutes are comparable or even superior to those of Sn–Pb solder. From a manufacturing point of view, the melting temperature of the solder alloy is a crucial factor that has to be taken into account in order to achieve high package quality. A large variety of lead-free solders have already been developed, mainly involving the Sn–Cu, Sn–Ni and Sn–Ag systems [1]. These new lead-free solders have been identified as the most promising alternatives to the eutectic Sn–Pb solder. However, the higher melting temperatures of these alloys, comparing to that of the eutectic Sn–Pb solder, limit their applications in the electronic industry.
For that matter, new methods have been explored to decrease the melting temperature of solder alloys. In 1954, Takagi carried out the first experimental investigation dealing with the effect of size on the melting of very small particles [2]. Utilizing electron microscope, Sambles observed the lowering of the melting point of the gold particles [3]. In 1976, Buffat also described that the melting temperature could be depressed by decreasing the particle size to nanometer scale [4]. Later, nanocalorimetry was employed to characterize the melting temperature of a small number of nanoparticles [5], [6], and hot stage transmission electron microscopy was used to observe the melting behavior of isolated nanoparticles [7]. Wong et al. reported the size-dependent melting temperature depression of Sn nanoparticles [8] by traditional differential scanning calorimetry (DSC). They also reported size-dependent melting temperature depression of Sn–3.5Ag nanoparticles synthesized by a chemical reduction method [9]. However, the production rate of nanoparticles was limited by this method. Hsaio and Duh manufactured and studied nanoparticles of Sn–3.5Ag–xCu (x = 0.2, 0.5, 1.0), synthesized for lead-free solder applications [10]. They did not, however, observe any obvious melting temperature depression in their investigation.
When the particle size was further decreased until to infinitesimal size, e.g. in which only tens or hundreds of atoms were contained, an anomalous size dependence of the melting temperature was observed and it was deemed that not only the cluster size but also structural features governed their correlations [11]. Nevertheless, no such correlation has been observed for the particle size scale of interest for solder applications. Banhart et al. [12] observed that the melting temperature of Sn and Pb nanocrystals might be increased to a value even higher than the corresponding bulk alloy if exterior shells exist. Embedded in some matrix, some nanoparticles revealed even opposite changing tendency of the melting temperatures [13], [14], [15], [16], indicating that the exterior structure of nanoparticles played a significant role in their melting behavior.
Premelting, which occurs before the whole melting of the clusters or nanoparticles, is easily produced near crystal defects [17]. Except the large specific surface area, the presence of large interfaces in nanoparticles was another crucial factor to influence the melting temperature and should be taken into account [18]. Pusey and co-workers [19], [20] observed that premelting at a lower temperature was produced along grain boundaries before complete melting of the crystalline colloids. Riegler and Kohler [21] also thought that the interfacial properties would strongly affect the phase transition behavior of the small particles due to their large interfaces, and premelting would occur on surrounding interfaces. However, little information was available about grain boundary premelting of metals due to their high temperature and the resultant difficulty for simultaneous observations. Fortunately, the recently developed coherent X-ray diffraction imaging technique might be a possible approach to study the premelting of nanocrystal interfaces owing to its diffraction patterns and latest third-generation synchrotron radiation sources to obtain full three-dimensional images [22].
In contrast to the lower melting temperature (183 °C) of SnPb eutectic solder alloy [23], therefore, the lead-free solder alloys usually cause a raised temperature of about 30–40 °C during electronic assembling. And the increased temperature reduces the integrity, reliability and functionality of printed wiring boards, components and other attachment [24].
As a result, much effort has been put lately on finding possible ways to solve some of the issues imposed by lead-free solders. Lin et al. employed the supernatant process to obtain the Sn–3.5Ag solder alloy nanoparticles, yet no obvious melting temperature depression was observed possibly due to the aggregation of those nanoparticles [25]. Therefore, it remains an urgent necessity to conduct an in-depth study of feasible processes to lower the melting temperature of the currently used lead-free solder alloy systems.
Section snippets
Experimental procedures
The lead-free master alloy with a nominal composition of Sn–3.0Ag–0.5Cu (wt.%) was prepared by an induction melting method. Fig. 1 shows the DSC melting curve of the master alloy. The melting temperature was 217.8 °C, which equals the equilibrium melting temperature. Integration of the peak yields 67 J/g for the heat of fusion.
Several models described the size dependence of the melting temperature of nano-sized particles. Generally it is assumed that the melting is initiated by a continuous
Results and discussion
Unlike the chemical reduction method applied by Wong and co-workers [9], [10], the CDCA technique is suitable for most conducting materials, and the manufacturing process could be continuous allowing preparation of lead-free solder nanoparticles at high production rates. During the preparation of the nanoparticles, discharge and solder alloy breakdown took place inside the dielectric media, protecting the nanoparticles from serious oxidation. The XRD pattern of the as-prepared nanoparticles is
Conclusions
Broad endothermic peaks were observed from the DSC measurements (Fig. 6, Fig. 8) of nanoparticles produced from a Sn–3.0Ag–0.5Cu (wt.%) master alloy with a consumable-electrode direct current arc (CDCA) technique. A fraction of the produced nanoparticles showed a melting temperature similar to that of eutectic SnPb solder. Most interestingly, they partly survived several melting crystallization cycles, indicating that the oxidation of the nanoparticles was continued slowly but the liquid
Acknowledgements
This work was supported by the National Natural Science Foundation of China (Grant No.50571057), AM Foundation of STCSM (Grant No. 08520740500) and Robert Bosch Foundation (Grant No. 32.5.8003.0025.0/MA01). Liu acknowledges the Swedish National Science Foundation support (Grant No. 621-2007-4660).
References (28)
- et al.
Mater. Sci. Eng. R
(2005) - et al.
Chem. Phys. Lett.
(2006) - et al.
Acta Mater.
(2004) - et al.
Acta Mater.
(1998) - et al.
Acta Mater.
(2001) - et al.
Mater. Sci. Eng. R
(2000) - et al.
J. Alloys Compd.
(2008) - et al.
J. Alloys Compd.
(2009) - et al.
Thermochim. Acta
(2007) - et al.
Prog. Mater. Sci.
(2007)
J. Phys. Soc. Jpn.
Proc. R. Soc. Lond. A
Phys. Rev. A
Phys. Rev. Lett.
Cited by (68)
Preparation and characterization of nano-solder paste with high nanoparticle loading and their thermal and printing properties
2023, Materials Chemistry and PhysicsSymmetries in the properties of mixing of AgCu liquid alloys at 1423 K
2021, Materials Today: ProceedingsEffect of micromorphology on corrosion and mechanical properties of SAC305 lead-free solders
2020, Microelectronics ReliabilityA modified constitutive model of Ag nanoparticle-modified graphene/Sn–Ag–Cu/Cu solder joints
2020, Materials Science and Engineering: AMicrostructure evolution and properties of rapidly solidified Au-20Sn eutectic solder prepared by single-roll technology
2019, Journal of Alloys and CompoundsSynthesis, characterization and thermal stability of SnAg and SnAgCu nanoparticles
2018, Journal of Alloys and CompoundsCitation Excerpt :Indeed, after DSC, the XRD patterns of citric acid treated samples, i.e. SAC_5 and SAC_6, show a higher intensity of the SnO peaks in comparison with XRD of untreated SAC samples. A quantitative evaluation of the SAC NPs melting enthalpy indicates a value of 50 ± 2 J/g, in comparison with the value of 67.0 J/g reported for the Sn3.0Ag0.5Cu (mass%) bulk alloy [52]. The morphological aspect of the SAC_2 sample after two DSC heating cycles is shown in Fig. 9.