Reliability of ACA Interconnects

Experimental Evaluation. With regard to the reliability of conductive adhesive joints, contact resistance of single joints is the most important feature. The reliability performance of ACA flip-chip joints on FR4 boards was investigated (Ref 29). In the study, nine types of ACAs (A-1) and one pure adhesive without any particles (J) were evaluated and some relevant technical data are shown in Table 1. As shown in Fig. 7, the test chip had a pitch of 100 pm, containing 

18 single joints and two daisy-chains (18 joints for each). The substrate used was a 0.8 mm (0.03 in.) thick FR4 rigid board. Therefore, the work was particularly focused on the reliability of ultra-fine pitch ACA flip-chip interconnects on low-cost substrates. In total, 954 joints (53 chips) with various ACAs were subjected to temperature cycling between —40 and 125 °C ( — 40 and 257 F) with a dwell time of 15 min. and a ramp rate of 110 °C/min. (230 °F/min.). The contact resistance of 36 joints (two chips) with ACA A was measured in situ during testing up to 3000 cycles. Other joints were taken out from the chamber every several hundreds of cycles and measured manually at room temperature. 

As shown in Fig. 9, the reUabflity data of ACA A with the various failure criteria have also been analyzed with the three-parameter 

Weibull distribution. In the case of 20% increase, the Weibull analysis indicated that there were always some joints that would reach this value even before the first cycle was finished (y = 0 in Fig. 9a). However, if the 50 mQ failure criterion was used, it could be guaranteed that no failure occurred within 1757 cycles (Fig. 9b). Similarly, the first failure would occur after 2347 

Cumulative failures of ACA joints measured manually at room temperature are shown in Fig. 10. The MTF ( 20%) of joints with ACAs A, C, and J was 2500 cycles. Although other joints. 

---