Reliability Comparison between Lead and Lead-Free Solder Joints

Affect of Solder Joints' Internal Microscopic Structure on Reliability

Microscopic structure of internal solder joints and IMC (intermetallic compound) structure at the interface of solder and PCB base determine mechanical properties of solder joints. Soldering techniques and later aging of solid phase, together with thermal cycle further determine original microscopic structure and its evolution. Optimal IMC is expected to be generated at the interface to implement wetting and metallurgical interconnection so that satisfying solder joint intensity and reliability can be achieved. Solder joints' internal microscopic structure demonstrates micro characteristics of material and available microscopes and technologies can be used to obtain the information.

• Lead solder joints

When it comes to SnPb, its micro structure is composed by Sn-rich phase and Pb-rich phase.

• Lead-free solder joints

In SAC alloy, metallurgical reaction between Sn and secondary element of Ag and Cu is the leading element determining its application temperature, curing mechanism and mechanical performance.

In accordance with binary-phase figure, three types of binary eutectic reactions are available between the above three types of elements:

a). Reaction between Ag and Sn takes place at a temperature of 221°C with eutectic structure at Sn base phase and ε IMC (Ag3Sn) phase formed.

b). Reaction between Cu and Sn takes place at a temperature of 227°C with eutectic structure at Sn base phase and η IMC (Cu6Sn5) phase formed.

c). Reaction between Ag and Cu also takes place at a temperature of 779°C with eutectic alloy formed of Ag-rich α phase and Cu-rich α phase.

Material ingredient determines microscopic structure that further determines failure mode. During the application of products, microscopic structure encourages the generation of tiny sediment. Particle dispersion, even distribution and granulation are beneficial to fatigue resistance improvement. Fatigue life will be reduced, however, when aci-form and brittleness phase and excessive cavities take place and stress concentrates. Even distribution of plastic deformation improvement within a small range through microscopic structure control is an effective measure to increase fatigue intensity.

Affect of Solder Joints' Interface IMC Microscopic Structure on Reliability

• Microscopic Structure of Interface IMC

a). Shape and Figure

Layer η- Cu6Sn5 features three types of shapes and figures:

1). Rough cellular layer. It features a section area containing dendrites between which such large space is available that a rough interface contacting with solder is achieved which is not a compact structure.

2). Compact layer on scalloped interface. Similar with dendrite crystal particles, this layer features a similar shape but with compact chemical compound. The interface contacting with solder is like a scalloped shape.

3). Compact layer on flat interface. As Pb content, temperature and reaction time rise, shape and figure of η layer begins to transform from rough cellular layer to compact layer on scalloped interface.

b). Affecting Elements

1). High cooling rate will lead to the generation of flat η-phase layer while a low cooling rate will lead to that of small-tumor η-phase layer.

2). Short reflow soldering time leads to flat η-phase layer while long reflow soldering time to small-tumor or scalloped η-phase layer.

c). Peel off

IMC that is originally generated between pad and liquid solder sometimes will separate from interface as reflow soldering time or reflow soldering times rise. This phenomenon is usually correlated with Ni. For example, it tends to take place more on Ni plating layer of ENIG.

1). IMC goes through peeling off at the interface of ENIG Ni plating layer at different content of phosphorus. Peel off is determined by phosphorus content improvement and reflow soldering time prolonging.

2). After some lead-free solder (Sn3.5Ag, Sn3.5Ag3.0Bi and SAC387) and some types of plated base [Cu, Ni(P)/Au and Ni(P)Pd/Au] go through reflow soldering for 20 minutes at the temperature below 250°C, interface IMC and most IMC layers formed with previous two types of solders will deviate or peel off from the interface with only a thin IMC left on the interface. When it comes to SAC387 on [Ni(P)/Au and Ni(P)/Pd/Au] base, IMC of (Cu, Ni)6Sn5 can be well connected with interface. as far as plated Ni base is concerned, however, three types of lead-free solders can be well connected with Ni3Su4 IMC.

d). Effect of Au on IMC between SAC solder and Cu base

IMC formed by Cu and SAC solder perform as pebbles. After Au of 0.1 to 5wt% is added to SAC387, eutectic phase generated at the temperature of 204.5°C contains 4 composites (AuSn4, Au3Sn, β-Sn and Cu6Sn5). As Au-Cu-Sn ternary metal compound is generated, most Au in solder will flow out and move towards interface. In interface reaction, participation of Au will be transformed from ordinary scalloped to compound type which is composed by (Au, Cu) 6Sn5 crystal particles and island-shape β-Sn with excellent distribution.

• Growing of Interface IMC Layer

Growing of interface IMC layer features extremely large influence on the reliability of solder joints. It's studied that no living rules are available between IMC thickness and time. Liquid phase condensation stops IMC from growing, leading to uneven growing.

As far as components with lead plated on pins, a rough linear ratio relation exists between IMC growing and square root of time, which can be regarded as being compatible with dispersion rule. However, for components whose pins are SnPb plated, SAC solder IMC growing features an obvious trend.

• Element Distribution on Solder Joint Interface

Based on high and low temperature impact and high-temperature test, it can be seen that a slight reduction takes place on Ag3Sn net structure during high temperature test and obvious change takes place towards granulous Ag3Sn phase without soldering intensity influenced. High temperature is used to implement interface alloy layer accelerating growing test. For components whose pins are lead plated, rough linear ratio relationship does occur between alloy growing and square root of time. Growing takes place under a certain dispersion control rate. Nevertheless, formed chemical compound is able to definitely reduce solder joints' intensity for either high-low temperature impact test or high-temperature test.

Lead-free solder joints feature higher hardness and intensity than SnPb solder joints and smaller deformation, which doesn't refer to high reliability of lead-free solder joints. Because lead-free solder alloy features worse wettability, more defects tend to take place such as cavities, displacement and tomb standing and cavity size tends to become large.

• Brittleness and Its Mechanism

1). Between plated Ni(P) layer and SnPb solder, long-time reaction will take place with Kirkendall cavities generated around Ni surface. When available copper is supplied to solder, brittleness will become more complex. Ternary metal compound (Cu, Ni)6Sn5 accumulates on Ni3Sn4 formed at Ni surface, cavities will be generated on Ni3Sn4/(Cu, Ni)6Sn5 at the time of aging. Similar issues will also occur when SAC solder is applied to connect with Ni because this type of solder alloy contains copper source.

2). Black pad is a unique phenomenon related with brittleness with massive agreement, which is especially so concerning ENIG Ni/Au. Obvious brittleness takes place at pad or around it due to insufficient solderability of Ni(P) surface, which will finally decrease mechanical fatigue intensity. Black pad is correlated with phenomenon concerning solder joint cracks. Anyway, harmful black pad effect is related with another brittleness that optimal metal alloy structure will degenerate as time goes by.

3). SAC solder plays a more important role than SnPb solder during black pad effect and aging process when IMC structure suffers from brittleness on ENIG Ni/Au pad. Lead-free soldering should avoid or decrease brittleness processes as a result of thickening of Au in Ni/Au coating.

4). Even the commonest thermal cycle usually requires solder joints to be capable of withstanding creep load aroused in each thermal cycle. Accordingly, the structure of IMC on pad has to withstand the load brought by solder creep. External mechanical load, especially that aroused by system mechanical impact, creep of solder is usually very high because deformation carried out by creep on solder joints is too large. As a result, even if it is capable of fully withstanding IMC structure in thermal cycle, the most fragile connections will be as well generated during shearing force or tensile force test.

5). Au with SnPb solder added in reflow soldering process will gradually return to Ni surface in later aging process, leading (Ni, Au)3Sn4 to be accumulated on Ni3Sn4 IMC. The interface generated as such is unstable and will decrease with the improvement of (Ni, Au)3Sn4 thickness.