Chemical Mechanical Planarization (CMP) has become an essential and enabling technique for the manufacturing of advanced semiconductor devices. Looking ahead, will CMP process be still needed when carbon nanotube is implemented as interconnect material? How would a moisture sensitive surface be CMPed? As the future technology node approaches molecular scale, what would be the desired or functional characteristics of CMP consumables such as abrasive particles and pad surface? Questions like these represent a small fraction of challenging issues faced by CMP community today. Possible answers to these questions might prepare us to meet the ever changing demand in future technology node implementation and embrace the tremendous new opportunities as CMP spread into other related technology areas that require surface preparation and fabrication. In this lecture series, a brief overview of these issues will be given from a chemist’s point of view. Some representative research activities carried out at Clarkson University in collaboration with industrial partners will be discussed.
More specifically, the discussion will be first focused on the true functions of abrasive particles as their sizes loosely follow the trend of critical dimensions in the technology nodes. Can the function of sub-20 nm colloidal silica still be accurately described as abrasives in a copper CMP slurry? How would a soft organic “abrasive” function in such an application? What would be the desired physical and chemical characteristics of abrasive particles to be used in 32 nm technology node and beyond?
The discussion will then turn to the applications of CMP in computer hard drive manufacturing related issues. More specifically, how would CMP be most useful when the industry is making the transition from LMR (Longitudinal Magnetic Recording) to PMR (Perpendicular Magnetic Recording)? The discussion will also address the issues related to the rapid growing needs in the manufacturing of MEMS (Micro-Electro-Mechanical Systems and NEMS (Nano-Electro-Mechanical Systems). The discuss will then shift to the pad surface at molecular and microscopic levels. For examples, will a metal CMP process be able to take advantages of a reactive pad? Will a so called non-cell pad be able to lower defectivity count and cost?
From processing parameters point of view, the pending increase of wafer size from 300 mm to 450 mm will bring tremendous challenges to tool manufacturers as well as consumable community. One such example is the progressive increase in temperature during polishing. This has also been seen during the transitions from 100 mm to 300 mm wafers. It is anticipated that the change from 300 mm to 450 mm will be even more dramatic. Will the future CMP consumables and processes be able to meet this challenge and possibly take advantage of the differences? |
