DocumentCode :
2043616
Title :
Systematic Pd Cu wire bond recipe development flow for successful qualification result
Author :
Law Lai Cheng ; Wong Boh Kid ; Hiew Pey Fang ; Eu Poh Leng
Author_Institution :
Freescale Semicond., Petaling Jaya, Malaysia
fYear :
2012
fDate :
13-16 Dec. 2012
Firstpage :
1
Lastpage :
4
Abstract :
One of the key advantages to replace bare Cu wire with Pd Cu wire in semiconductor packages is to reduce corrosion risk against halogen from mold compound or substrate soldermask outgasing. However, existing wire bond recipe development methodology adopted from bare Cu wire was less effective for Pd Cu due to harder Free Air Ball (FAB) with Pd Cu wire, causing Pd Cu bond failure after biasing reliability stress such as HAST. This study is aimed to obtain a proper methodology in developing a reliable Pd Cu wire bond recipe. In this study, Step # 1 was to determine the best method to study Pd distribution in FAB and bonded ball surface. For Pd distribution on FAB surface, 2 methods were studied: 1. inspection under high power optical scope; 2. soaking into Nitric Acid and Ferric Chloride. For Pd distribution on bonded ball surface, the 2 methods explored were SEM vs EPMA after cross-sectioning of the bonded ball. Step # 2 was to conduct a DOE with different EFO current and time to obtain optimum Pd distribution surrounding the bonded ball. Step # 3 was to down select EFO current and time that provided optimum Pd distribution and bondability. Step # 4 was to proceed with wire bond parameters derivation through comprehensive design of experiments (DOE) and response surface methodology (RSM) to establish the bonding parameter window. Step # 5 was to subject the bonded and molded units for reliability stressing which has biasing stress, such as HAST or THB. Upon completion of the study, a systematic flow was established and applied to a test vehicle of C90 wafer technology in a thermally enhanced 31×31mm 689TePBGA-II. The result showed that the established flow was effective to prevent any failure in biasing stress, where 3 lots × 80 units had passed 1008hrs THB and 3 lots × 25 units has passed 2016hrs THB.
Keywords :
ball grid arrays; copper; corrosion; design of experiments; electron probe analysis; failure analysis; inspection; lead bonding; palladium; response surface methodology; scanning electron microscopy; semiconductor device packaging; semiconductor device reliability; 689TePBGA-II; C90 wafer technology; DOE; EFO current; EPMA; FAB surface; HAST; PdCu; RSM; SEM; THB; biasing reliability stress; biasing stress; bond failure; bondability; bonded ball cross-sectioning; bonded ball surface; comprehensive design of experiments; corrosion risk reduction; free air ball; halogen; high-power optical scope; inspection; mold compound; molded units; nitric acid-ferric chloride soaking; optimum palladium distribution; qualification result; reliability stressing; response surface methodology; semiconductor packages; substrate soldermask outgasing; systematic palladium copper wire bond recipe development flow; time 1008 hr; time 2016 hr; wire bond parameter derivation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electronic Materials and Packaging (EMAP), 2012 14th International Conference on
Conference_Location :
Lantau Island
Print_ISBN :
978-1-4673-4945-1
Electronic_ISBN :
978-1-4673-4943-7
Type :
conf
DOI :
10.1109/EMAP.2012.6507907
Filename :
6507907
Link To Document :
بازگشت