We investigated a Cu-loss problem after dual-damascene patterning during manufacturing; that is, more than a dozen wafers were stored in a FOUP. We found that a decreased yield due to the Cu-loss strongly depended on the wafer position in a FOUP and on the queue time between etching and wet cleaning. We developed a Cu-oxidation model to explain what happens in a FOUP during the queue time; that is, the F content, which catalyze Cu-oxidation, in the post-etch residue gradually evaporate into the slot-to-slot space. The ambient gas analysis in a FOUP showed that F-containing gas evaporates from the post-etch wafers, and that the evaporation is gradual, which is consistent with our model. On the basis of our model, we controlled ambient gas in the slot-to-slot space. The increased yield showed that the Cu-loss problem was successfully suppressed.
No takes yet. Share an insight, caveat, or question.
Kamoshima et al. (2008) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: