Conformity - November 2008 - (Page 40) However, the ionization safely removes the charge well before the upcoming “baseline” robot arm discharge (#6). 6. Robot arm does not discharge wafer: The wafer is not charged by the time the robot arm makes contact. However, the “solution” was not yet perfected. When the ionization was in constant operation in step #5 above, the airflow from it caused a potential contamination problem in the process itself. Specifically, the normal fluid spin-off was disrupted by the airflow from the ionizer, and insufficient wafer cleanliness (water spots, essentially) was observed as a result. To correct this phenomenon, the ionizer airflow was shut off just before the step #5 charging mechanism, and turned back on after a few seconds delay. This was necessary to complete the fluid spin-off step, but still allow for complete discharge during the remainder of the process. Floating Metal and FOUP Issues The single wafer chamber (evaluated above) processes wafers with active metal structures on them. Even with the wafer potentially grounded in the chamber fixture, it is noted that simply grounding the wafer in the chamber does not ensure that all electrical damage possibilities to wafers are eliminated. As an example (even if most of its metal structures are grounded), there are many wafers that may have additional “floating metal” structures internally insulated electrically from the “grounded” structures. It is still possible for CDM discharges to occur when the wafer is “grounded.” Floating structures can become charged inductively and then discharge internally to the grounded structures. Many case studies exist where “floating metal” structures have been an integral part of CDM damage. Implementation of ionization inside (chamber) mini-environments is often necessary to guard against ESD and ESA. There is another example of why simply grounding the wafer in the tool chamber is not reliable enough to eliminate all ESD and ESA issues. Charging mechanisms can occur to wafers in a FOUP, even though the wafer is “connected to ground” internally in the FOUP itself. FOUP designs incorporate an inner (skeletal) fixture that is designed to “keep the wafers at ground potential” during normal operations. However, recent studies22 have conclusively proven that native oxides on wafer surfaces can prevent good grounding contact of the wafer by this fixture, and that the wafer can be essentially a “floating” conductor. As the rest of the FOUP container is constructed from plastics that can charge greatly during normal handling and which is in close proximity to the wafers, the resulting large fields couple into the “floating” wafer and cause it to charge inductively. Charged wafers consequently attract more particles, and run the risk of electrical ESD damage when contacted by subsequent conductors, such as robot arm assemblies. Recent studies are available that document substantially increased particle counts on wafers transferred in and out of FOUPs when ionization equipment is not implemented22. Summary Wafer charging hazards can exist in semiconductor tool chamber minienvironments, even when ionization resides in other areas of the tool (equipment front-end, or EFEM, and process modules). It is important to evaluate the process steps involved in real time to properly characterize and eliminate potential ESD, ESA, and EMI failure modes. Real time monitoring of the wafer is a prerequisite to characterize the risks in the chamber thoroughly. Eliminating potential ESD, ESA, and EMI failure modes in the chamber mini-environment and inside the semiconductor tool is an important part of an overall yield improvement program, and should be considered an important addition to standard E78 evaluations. Figure 4: Selective ionization throughout process 0 Conformity november 2008
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.