Search

In this paper, the coupling effect in Sn–3.5Ag–3.0Bi and Sn–8.0Zn–3.0Bi solder joint with sandwich structure by long time reflow soldering was studied. It was found that the interfacial compound at the Cu substrate was binary Cu–Sn compound in Sn–Ag–Bi solder joint and Cu5Zn8 phase in Sn–Zn–Bi solder joint. The thickness of the Cu–Zn compound layer formed at the Cu substrate was greater than or equal to that of Cu–Sn compound layer, although the reflow soldering temperature of Sn–Zn–Bi (240 °C) was lower than that of Sn–Ag–Bi (250 °C). The stable Cu–Zn compound was the absolute preferential phase in the interfacial layer between Sn–Zn–Bi and the Cu substrate. The ternary (Cu, Ni)6Sn5 compound was formed at the Sn–Ag–Bi/Ni(P)–Cu metallization interface, and a complex alloy Sn–Ni–Cu–Zn was formed at the Sn–Zn–Bi/Ni(P)–Cu metallization interface. It was noted that Cu atoms could diffuse from the Cu substrate through the solder matrix to the Ni(P)–Cu metallization within 1 min reflow soldering time for both solder systems, indicating that just 30 s was long enough for Cu to go through 250 μm diffusion length in the Sn–Ag–Bi solder joint at 250 °C. The coupling effect between Ni(P)/Cu metallization and Cu substrate was confirmed as the type of IMCs at Ni(P) layer had been changed from Ni–Sn system to Cu–Sn system apparently by the diffusion effect of Cu atoms. The (Cu, Ni)6Sn5 layer at the Ni(P)/Cu metallization grew significantly and its thickness was even greater than that of the Cu–Sn compound on the opposite side, however the growth of the complex alloy including Sn, Ni, Cu and Zn on the Ni(P)/Cu metallization was suppressed.

The Sn-3.5Ag and Sn-4.0Ag-0.5Cu solders on electroless Ni-immersion Au metallization exhibited different interfacial morphology after high temperature storage (HTS) testing at 150/spl deg/C. Ni/sub 3/Sn/sub 4/ intermetallic compounds (IMCs) were found in the Sn-Ag system, while in the Sn-Ag-Cu system, spalling grains of (Cu,Ni)/sub 6/Sn/sub 5/ IMC were observed after long aging, along with a thick (Ni,Cu)/sub 3/Sn/sub 4/ IMC layer which adheres to the P rich electroless Ni coating. Kirkendall voids were found between the Sn-Ag and electroless Ni interface after 168 and 500 hour thermal aging, while a clear gap existed at the interface after 1000 hour aging, as a result of the growth of the micro-voids. In the Sn-Ag-Cu system, Kirkendall voids were found inside grains of the Sn-Ni-Cu ternary interfacial compound after 500 and 1000 hour aging.

For double-sided assemblies, the solder joints on the topside of the board are inverted and reflowed again. During the second reflow, the components are held in place by the surface tension, which may prevent the components from falling off under the gravitational force. A method is needed to determine a component’s candidacy for bottom-side attachment based on the component weight and total pad area. In this paper, a theoretical model was introduced to determine the critical value for component fall-off during the second reflow. Design of Experiments (DOE) and ANOVA analysis for lead-free solder boards were performed to examine the main process factors which have different effects on the component fall-off for different components, and comparison was made between lead-free and SnPb solders. Optical inspection and cross-sectioning were carried out for further investigation. The test results indicated no significant difference of Cg=Pa value between SnPb and lead-free solders. [doi:10.2320/matertrans.47.1577]

The wetting of a molten solder on metallic surfaces is a rather complex phenomenon. In addition to physical spreading due to surface tension reduction, there are interfacial metallurgical and flux chemical reactions with the metallic substrate surface. Substrate dissolution and intermetallic formation take place rapidly during soldering. Since lead-free soldering requires substantially higher soldering temperatures (around 250 � C), the rates of intermetallic growth and substrate dissolution for lead-free solders are expected to be significantly greater than those for the current Sn–Pb eutectic solder. This study systematically investigates the metallurgy of the solid–liquid interface reactions and intermetallic growth kinetics for three lead-free solders: Sn–Ag eutectic (96.5%Sn–3.5%Ag), Sn–Cu eutectic (99.3%Sn–0.7%Cu) and Sn–Ag– Cu eutectic (Sn–3.8Ag–0.7Cu, SAC 387) with three metallic substrates: Cu, Ni, and Alloy 42 (42%Ni–52%Fe) over temperatures ranging from 225 to 280 � C for reaction time from 10 s to 16 h. Wetting behavior of these three alloys on PCBs with OSP, immersion Sn, and Ni/Au finishes, was also examined from 220 � C up to 260 � C. A thorough understanding of lead-free solder/substrate interfacial reactions should give guidance to the optimum lead-free soldering processes and to the optimum lead-free coating thicknesses for component and PCB terminal finishes, as well as for under-chip metallurgical coatings for flip-chip and BGA applications.

Reliability of lead-free solder joints has been a hot topic widely debated in the electronic industry. Most published data indicate that a change to lead-free soldering has the potential benefit of more reliable solder joints than the current Sn-Pb eutectic solder joints. However, in reality many mechanical, metallurgical, thermal, and environmental factors affect the service reliability of solder joints. This paper tries to shed some light on the effects of mechanical loading and thermal conditions on solder joint reliability. These conditions are determined not only by external environments but also by the solder alloy itself and the joint geometry. Analyses with first principles are carried out on solder joints of both areal array and peripheral packages. Effects on fatigue life of solder joint geometry, thermal and mechanical characteristics of components and substrate materials, and application conditions are discussed. The analysis helps explain why lead-free solder joints may not be more reliable in certain application conditions than the current Sn-Pb eutectic solder joints.

The wetting of I-Ag (immersion silver) and I-Sn (immersion tin) printed-circuit-board (PCB) finishes by Sn/Ag/Cu and eutectic Sn/Pb solders was studied in this work with Ni/Au (electroless nickel/immersion gold) and organic solderability preservative (OSP) finishes as baselines. Wetting tests were performed on fresh boards and boards subjected to different preconditioning treatments that simulated the effects of aging, storage, and multiple reflow cycles. When the boards are fresh, the wetting of the I-Sn and Ni/Au finishes is better than that on the I-Ag and OSP finishes. However, after the preconditioning treatments, the wetting of the I-Sn finish degrades the fastest, whereas the wetting of the I-Ag and OSP finishes degrade less through the different preconditioning treatments. The wetting of the Ni/Au finish remains excellent through all the preconditioning treatments. The chemical and microstructural changes in the finishes during aging treatments were evaluated using electron spectroscopy chemical analysis (ESCA), x-ray diffractometry (XRD), and cross-sectioning followed by scanning electron microscopy (SEM). The results indicate that a single lead-free reflow cycle consumes the I-Sn layer faster than a Sn/Pb reflow cycle because of the formation of the Sn/Cu intermetallic compound (IMC). Consequently, I-Sn finished boards having an original Sn thickness of ∼1 µm will not withstand multiple lead-free reflow cycles without significant degradation in wetting but up to two Sn/Pb reflow cycles are still feasible. The minimum thickness of I-Sn required for adequate wetting was evaluated by comparing the wetting after different aging treatments. The exposure of I-Sn samples to 85°C/85% relative humidity (RH) conditions increases the thickness of the Sn-oxide layer, which, above a certain thickness, can degrade wetting. Oxidized copper areas formed on top of the I-Ag surface after exposure to 85°C/85% RH treatment, and this was considered a major factor influencing wetting. The formation of sulfides on I-Ag was detected, but their overall quantity remained too small to have a detectable impact on the wetting.

Solder balling in Sn/Ag/Cu solder pastes was studied in this work. Three different solder pastes, several different reflow profiles and conditions, and two stencil thicknesses were used in the investigation. During the first phase, called the verification phase, the solder pastes were checked to ensure they met the minimum requirements. In the process-screening phase, the reflow profile was varied. Results show that besides flux chemistry, reflow atmosphere plays the major role in solder balling. The average number of solder balls with the best paste was one fifth of that with the worst paste. Furthermore, with all the pastes, the number of solder balls dropped close to zero when nitrogen atmosphere was used. Another finding during the reflow process screening was the influence of the stencil thickness on the solder-balling result. With a thinner stencil, two of the pastes exhibited significant solder balling. This is assumed to be caused by the different ability of fluxes to withstand oxidation during the preheating in the reflow process. In the last phase, the effect of the solder-paste particle size on solder balling was studied more closely. The flux chemistry was kept unchanged, and the solder particle size was varied between type 3 and type 4. The results show that, with type 4 paste, significantly more solder balls are formed compared to type 3 paste. It was also confirmed that, regarding the reflow profile, the ramp-up rate from 150°C to 217°C and the reflow atmosphere were the most significant factors that determine the solder-ball formation for both types of paste.

This paper presents the drop test reliability of 0.5 mm pitch lead-free chip scale packages (CSPs). Fifteen 0.5 mm pitch CSPs were assembled on a standard JEDEC drop reliability test board with Sn3.0Ag0.5Cu lead-free solder. Eight boards were edge-bonded with a UV-cured acrylic; eight boards were edge- bonded with a thermal-cured epoxy; and twelve boards were assembled without edge bonding. Half of the edge-bonded test boards were subjected to drop tests at a peak acceleration of 1500 G with a pulse duration of 0.5 ms, and the other half subjected to drop tests at a peak acceleration of 2900 G with a pulse duration of 0.3 ms. Half of the test boards without edge bonding were subjected to drop tests at a peak acceleration of 900 G with a pulse duration of 0.7 ms, and the other half subjected to drop tests at a peak acceleration of 1500 G with a pulse duration of 0.5 ms. Two drop test failure detection systems were used in this study to monitor the failure of solder joints: a high-speed resistance measurement system and a post-drop static resistance measurement system. The high-speed resistance measurement system, which has a scan frequency of 50 KHz and a 16-bit signal width, is able to detect intermittent failures during the short drop impact duration. Statistics of the number of drops to failure for the 15 component locations on each test board are reported. The effect of component position on drop test reliability is discussed. The test results show that the drop test performance of edge-bonded CSPs is five to eight times better than the CSPs without edge bonding. However, the drop test reliability of edge-bonded CSPs with the thermal-cured epoxy is different from that with edge-bonded CSPs with the UV-cured acrylic. The solder crack location and crack area are characterized with the dye penetrant method. The fracture surfaces are studied using scanning electron microscopy (SEM).