Search

The effect of iron trichloride (FeCl3) on chemical mechanical polishing (CMP) of Ge2Sb2Te5 (GST) film is investigated in this paper. The polishing rate of GST increases from 38 nm/min to 144 nm/min when the FeCl3 concentration changes from 0.01 wt% to 0.15 wt%, which is much faster than 20 nm/min for the 1 wt% H2O2-based slurry. This polishing rate trends are inversely correlated with the contact angle data of FeCl3-based slurry on the GST film surface. Thus, it is hypothesized that the hydrophilicity of the GST film surface is associated with the polishing rate during CMP. Atomic force microscope (AFM) and optical microscope (OM) are used to characterize the surface quality after CMP. The chemical mechanism is studied by potentiodynamic measurements such as Ecorr and Icorr to analyze chemical reaction between FeCl3 and GST surface. Finally, it is verified that slurry with FeCl3 has no influence on the electrical property of the post-CMP GST film by the resistivity–temperature (RT) tests. [ABSTRACT FROM AUTHOR]

We investigate the effect of cations with different valences on the chemical mechanical polishing (CMP) of silicon dioxide films. The removal rate and surface roughness of the silicon-dioxide-film post-CMP are checked for the silica-based slurry with different cation salts (NaCl, CaCl2, AlCl3). Meanwhile, the particle size and size distribution of the slurries are characterized to test their lifetimes. The result shows that the three kinds of salts can improve the polishing removal rate from around 20 nm/min to 120 nm/min without affecting the surface roughness when the polishing slurry is stable. With increasing valence of cations, the polishing slurry requires less cation concentration to be added to improve the removal rate, while keeping a superior surface topography and maintaining a longer lifetime as well. [ABSTRACT FROM AUTHOR]

In the paper, chemical mechanical planarization (CMP) of Ge2Sb2Te5 (GST) is investigated using IC1010 and Politex reg pads in acidic slurry. For the CMP with blank wafer, it is found that the removal rate (RR) of GST increases with the increase of pressure for both pads, but the RR of GST polished using IC1010 is far more than that of Politex reg. To check the surface defects, GST film is observed with an optical microscope (OM) and scanning electron microscope (SEM). For the CMP with Politex reg, many spots are observed on the surface of the blank wafer with OM, but no obvious spots are observed with SEM. With regard to the patterned wafer, a few stains are observed on the GST cell, but many residues are found on other area with OM. However, from SEM results, a few residues are observed on the GST cell, more dielectric loss is revealed about the trench structure. For the CMP with IC1010, the surface of the polished blank wafer suffers serious scratches found with both OM and SEM, which may result from a low hardness of GST, compared with those of IC1010 and abrasives. With regard to the patterned wafer, it can achieve a clean surface and almost no scratches are observed with OM, which may result from the high-hardness SiO2 film on the surface, not from the soft GST film across the whole wafer. From the SEM results, a clean interface and no residues are observed on the GST surface, and less dielectric loss is revealed. Compared with Politex reg, the patterned wafer can achieve a good performance after CMP using IC1010. [ABSTRACT FROM AUTHOR]