Schematic diagram of plasma etch chamber equipped with ATR-FTIRS setup.
Conceptual layout of a partially patterned film on a Ge ATR crystal. The ATR technique is used to measure the removal rate of as well as the formation rate of C–F in the sidewall passivation layer. Patterned photoresist and anti-reflective coating are stacked on top of mesoporous film. The magnified cross-sectional view outlines the etch front at the trench bottom.
Time-series of IR absorbance spectra taken during a 2 min etch process show both C–F and Si–O–Si stretching vibrational modes. Si–O–Si absorbance becomes increasingly negative, indicating its loss, whereas C–F absorbance becomes increasingly positive, indicating its gain during the etch.
Integrated absorbance of Si–O–Si is plotted vs time for (a) a sample with relatively anisotropic etch profile and (b) a sample with highly nonideal, bell-jar shaped etch profile. The corresponding cross-sectional SEM images are shown in the insets. The scale bar is 200 nm. The IR absorbance corrected for the exponentially decaying evanescent field is used to calculate the total loss of Si–O–Si bonds. A nonlinear slope of total Si–O–Si loss during plasma-on period foretells a deviation from perfect anisotropy. The nonideal profile is particularly pronounced for the profile shown in (b).
IR absorbance of C–F stretching vibrational mode near corrected for the exponentially decaying evanescent field strength is used to calculate the total number of C–F formed during etching for (a) a sample with relatively anisotropic etch profile and (b) a sample with highly nonideal, bell-jar shaped etch profile. The inset in (a) conceptually describes the shape and thickness of sidewall passivation in the beginning (B), intermediate (I), and final (F) stage of etching. Note that the inset is not drawn to scale. The C–F buildup is pronounced towards the end point, and a critical buildup is necessary to maintain the anisotropic profile.
Article metrics loading...
Full text loading...