Search results

This paper discusses the microstructures of solder joints and the mechanism of thermal fatigue, which is an important source of failure in electronic devices. The solder joints studied were near‐eutectic Pb‐Sn solder contacts on copper. The microstructure of the joints is described. While the fatigue life of near‐eutectic solder joints is strongly dependent on the operating conditions and on the microstructure of the joint, the metallurgical mechanisms of failure are surprisingly constant. When the cyclic load is in shear at temperatures above room temperature the shear strain is inhomogeneous, and induces a rapid coarsening of the eutectic microstructure that concentrates the deformation in well‐defined bands parallel to the joint interface. Fatigue cracks propagate along the Sn‐Sn grain boundaries and join across the Pb‐rich regions to cause ultimate failure. The failure occurs through the bulk solder unless the joint is so thin that the intermetallic layer at the interface is a significant fraction of the joint thickness, in which case failure may be accelerated by cracking through the intermetallic layer. The coarsening and subsequent failure are influenced more strongly by the number of thermal cycles than by the time of exposure to high temperature, at least for hold times up to one hour. Thermal fatigue in tension does not cause well‐defined coarsened bands, but often leads to rapid failure through cracking of the brittle intermetallic layer. Implications are drawn for the design of accelerated fatigue tests and the development of new solders with exceptional fatigue resistance.

Studies have been made of ageing effects, both at room temperature and at 125°C, on the microstructure of 60:40 tin‐lead solders. A comparison was made of the effect of ageing on slow and rapidly cooled matrices. Precipitation of tin within lead dendrites was observed to occur very rapidly after solidification of the alloy. Subsequently the precipitates coarsened markedly over a period of a few weeks. The matrix of the alloy also coarsened at room temperature over this period. At elevated temperatures a similar sequence of events occurred, but substantially faster. The microstructural origins of the known loss in mechanical strength of solders with ageing are discussed.

A mechanical and microstructural study was performed of 43/43/14 tin/lead/bismuth solder. This alloy melts lower than the commonly used tin/lead solders and therefore holds promise as a useful material in two‐step soldering processes or in processes with thermally sensitive components. Mechanical testing of 43/43/14 tin/lead/bismuth showed a strength comparable to that of tin/lead solders but increased creep rate. Microstructural analysis (scanning electron microscopy/energy dispersive X‐ray) exhibited the same mechanism of fatigue as for tin/lead solders, viz., heterogeneous coarsening. Thermocyclic fatigue demonstrated that the long‐term reliability of 43/43/14 tin/lead/bismuth is comparable to that of tin/lead solders.

A model SMD test‐piece has been developed which permits mechanical stressing of a solder joint in a similar mode to that occurring in the process of thermal fatigue. The failure mechanisms in the 60Sn40Pb alloy studied have been in agreement with those frequently observed in thermal fatigue. Further, the fatigue life of model joints was found to increase with increasing solder volume whilst, upon ageing at room temperature, fatigue resistance decreased. These effects were attributed to microstructural changes occurring within the solder.

Since the introduction of surface mounting technology (SMT), interest in solder joints has increased greatly. In this work soldered joints were examined for different kinds of components. Factors which influenced the temperature cycle most strongly were the component size and its placing on the circuit board. These factors led to significant differences in maximum temperature and cooling rate between different solder joints on one and the same circuit board. The microstructure of the solder joints could then be related to their individual cooling rates. The near ternary eutectic Sn‐Pb‐Ag solder presented large variations in microstructure for the interval of cooling rates which was investigated. Strengths of the joints are therefore expected to vary because of the relation between microstructure and strength. Dissolution of elements from metallised coatings can also influence markedly the structures and properties in the solder joints.

The deterioration of surface mounted solder and adhesive joints on different substrate materials under thermal cycling was investigated metallographically. Ceramic chip resistors and leadless chip carriers were soft‐soldered or glued onto alumina, FR‐4, aluminium or steel boards and the various cracking modes were observed. Fatigue cracking in the solder under the component (mode A) took place in the case of resistors on an Al substrate and carriers on all boards except alumina. Cracking on the outward surface near the upper and lower corners (mode B) occurred on all boards, but most notably on alumina. Adhesive joints seemed to offer the highest fatigue strength, but their electrical properties suffered continuously in the course of cycling even though cracking was not observed at all in many cases.

Fatigue crack growth (FCG) tests on lead‐containing solders and lead‐free solders have been carried out at frequencies ranging from 0.01 to 10 Hz and stress ratios in the range 0.1–0.7. The FCG resistance of lead‐free solders was found to be superior to that of lead‐containing solders. For both types of solder, cycle dependent behaviour is dominant for the tests at low stress ratios and high frequencies, while time‐dependent effects become important at high stress ratios and low frequencies. For cycle dependent testing conditions, cracks primarily propagated in a transgranular manner, while a mixed trans/intergranular mode of crack propagation was observed for testing conditions where time dependent effects were dominant. The propagation path of intergranular cracks depended on the test materials, and along interfaces. After the FCG tests, the formation of small grains was observed.

The wearout characteristics were investigated for soldered interconnections of surface mount technology (SMT) chip resistors, chip capacitors and a 44 I/O ceramic leaded chip carrier (CLCC) package. Four double‐sided test vehicles were subjected to accelerated thermal cycling in the — 10°C to + 110°C range; 30°C/min ramp rate; and 1 minute dwell time at each temperature extreme. The test was interrupted at initially 500 cycle and later at 1000 cycle intervals to perform visual inspection of all soldered interconnections, functional performance verification for the test vehicles, and resistance measurement on leaded SMT joints. Metallographic examinations and fractographic studies were also performed after 0, 4500 and 13000 cycles to characterise the micromechanisms of soldered joint strength degradation and failure. The wearout thresholds for soldered joints of chip resistors and capacitors on side 1 were respectively 2500 and 4500 cycles. The greater thermal fatigue resistance of the latter joints was attributed to a lower device‐substrate coefficient of thermal expansion (CTE) mismatch and a more favourable device geometry compared with chip resistors. These passive components on side 2, however, showed a virtually identical soldered joint wearout threshold of 6500 cycles. The constraints imposed by the applied mounting adhesive were primarily responsible for this behaviour. No correlation appeared to exist among various failure criteria used to determine the onset of failure for leaded SMT soldered connections. The concurrent monitoring of electrical resistance and the applied tensile load showed a modest relationship between the load drop and resistance increase, however. The test vehicles continued to pass the functional performance verification, even after 13000 thermal cycles. Nonetheless, the joint wearout thresholds were considered to be 2500, 4500 and 4500 cycles for chip resistor, chip capacitor and CLCC components, respectively. A 50% soldered joint strength drop was considered as the wearout threshold for the CLCC device. Metallographic examination showed limited barrel wall cracking of the vias and no evidence of cracks with the through‐hole soldered joints, even after 13000 thermal cycles.

The mechanical properties of eight solder alloys from the Pb‐Sn‐In‐Ag alloy systems were determined over the temperature range ‐200°C to 100°C, using uniaxial tensile tests, dynamic mechanical analysis (DMA), acoustic pulse methods and dilatometry. In general, the strength and elastic modulus of the alloys studied was inversely dependent on temperature. PbSn, PbIn and SnIn alloys were observed to turn superplastic with elongations over 100 per cent at temperatures of 50°C or above. The Pb‐based and In‐Sn eutectic solders possessed superplasticity at temperatures greater than 50°C. From these results, deformation and fracture processes are reviewed, and the appropriate fracture mechanism is proposed.

The unique advantages of confocal microscopy are used to explore four cases of interest: (i) voids in solder (depth and surface texture determined), (ii) steam vs ambient aged solder coupons (significant differences detected), (iii) integrated circuit construction (sub‐surface contamination by µm‐size particles observed) and (iv) circuit boards and solder pads (non‐destructive optical sectioning through no‐wash flux layers). It is shown that confocal microscopy strongly complements SEM (scanning electron microscopy); SEM alone presents an incomplete description of a solder surface and in fact can sometimes produce misleading results.