February_EDFA_Digital

edfas.org 39 ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 21 NO. 1 T he FIB User Group meeting was held concurrent with the Contactless & Nanoprobing User Group during the two-hour lunch break on Thursday. Overall attendance was up from last year, topping 100. This year, a new format was initiated, presenting a list of the top 30 topics collected by audience interviews upfront as discussion suggestions. A fewmicro-talks (three slides/ three minutes) were prepared as conversation starters, should they be needed to ensure robust discussion topics were available. The organizers wish to thank Thermo Fisher Scientific as the sponsor of this user group event. We began the discussion on FIB tools, uptime, and keeping them running. Brian Tracy, EAG, opened with a comment that he loves all-inclusive coverage as it maxi- mizes his uptime. Valery Ray, PBS&T, MEO Engineering/ UCONN, then presented current service structures for lab tools and noted the pros and cons of each option: OEM support/third party support/self-service with diversified support. Users noted that while OEM support provided the lowest downtime due to contractual commitments, the cost of full-service contracts can be expensive and makes the most sense when the lab has several tools. Some users noted that they have had full service contracts since their first tool. Ray also commented on the recent “Right to Repair” legislation, which allows individuals to service their own systems. While the law provides users with additional flexibility and options formaintaining their systems, users noted concerns over how self-servicing would affect the systemwarranty. One user stated a need for “super users” to have basic triage training from the factory, so they can best recognize problems and report them to field service engineers. Vendor audience members also agreed that accurate reporting will help reduce tool downtime. The next topic addressed the benefits of TEM/STEM imaging. Providers noted that too often end-customers do not necessarily understand their actual needs and often request services that are muchmore advanced and labor intensive thanneeded. It was noted that labs need towork more closelywith customers to understand requirements andprovide the right level of support and turnaround time to meet success rate expectations. It was also mentioned that at times TEM analysis is not necessary, and instead STEM analysis will be sufficient to understand the failure ISTFA 2018 FOCUSED ION BEAM (FIB) USER GROUP Chair/Co-Chairs: Steven Herschbein and Michael Wong steven.herschbein@globalfoundries.com , mike.wong@thermofisher.com mechanisms. Members noted that when purchasing a dual beamsystem, the STEMoption is worth considering. Adam Steele, ZeroK, and Rick Livengood, Intel, both provided a prepared overview of the LoTIS Cesium (Cs+) technology. Some attributes of the Cs+ source include its high brightness, long source life, high resolution (2 nm) at low beam energy, high secondary ion yield (whichpromotes excellent SIMS results), andhigher depo- sition yield for materials like tungsten (but with 2× higher resistivity due to higher carbon and organic residue). This topic concluded with sharing the successful edit results (electrical data and cross sections) using a ring oscillator fabricated in a 10 nm process. Another topic on tool options and accessories was introduced. One user noted that SIMS was a good option to have for lithium research as it has 10× sensitivity over EDS. Another interesting and useful option is the chamber backscatter detector. Guy Burg, Marvell, asked, “Who does your FIB? Was it more cost/time effective to have an in-house FIB lab or to outsource the work to a FIB house?” While his main inter- est was chip edit, the discussion addressed all aspects of lab FA work. The audience discussed the pros and cons of each option. For FIB houses, issues such as turnaround time, IP constraints, process technology knowledge, and access todesignerswereareas of concern. For the in-house FIB, providing good justification tomanagement to invest in such large capital and staffwas themaindiscussion. The audience provided examples of how their FA/CE allowed products tobe first tomarket before competitors. Users all agreed that FA/CE, while sometimes difficult toquantify its return on investment, is actually extremely cost effective and provides a great return. The final topic was on cross sectional curtaining reduction when using large beams with big tails (typical “AT TIMES TEM ANALYSIS IS NOT NECESSARY, AND INSTEAD STEM ANALYSIS WILL BE SUFFICIENT TO UNDERSTAND THE FAILURE MECHANISMS.”