The wafers were prepared through the following steps (certain steps were not performed on specific control wafers):
The measurement wafers and the control wafers are described in the table below.
||0.8 um||30 cycles|
||0.09 um||30 cycles|
||Measurement||SiO 300-1||0.09 um||20 cycles
||Measurement||SiO 350-1||0.09 um||20 cycles|
||Measurement||SiN 300-1||0.09 um||20 cycles|
||Measurement||SiN 350-1||0.09 um||20 cycles|
1. Film deposition
Test wafers with 100nm thermal oxide and 1um Al were used.
The films were deposited with the recipes on the table below. The specific thickness of each wafer is given on the table of wafer IDs.
|Recipe||Substrate Temp||Heat Exchanger Temp|
| SiO 300-1
2. Xactix Etch
The wafers were etched according with cycles of 30 at 3Torr of XeF2. The number of cycles of each wafer is given on the table of wafer IDs.
3. Visual Analysis
The wafers were analyzed with the assistance of a microscope.
The results of the first Xactix etch (wafers 41, 42, 43, 44)
demonstrated a higher presence of pinholes on the nitride films (wafers 43,
44), indicating that they present pinholes for films thinner than 0.09 µm. The
results for the oxide films (wafers 41, 42) with 0.8 µm thickness indicated
there was not a relevant amount of pinholes in the film due to the CCP
deposition. This motivated the second Xactix etch.
On the second etch (wafers 50, 51, 52, 54), the nitride films presented significantly less etched spots compared to the amount from the first etch. The SiO 300-1 film still presented no relevant amount of pinholes. However, the SiO 350-1 film presented a very large amount of pinholes across the whole surface.
The experiment points to the idea that the structure for nitride films is less stable, since there was a great variance between the results from each etch. The second etch (wafers 50, 51, 52, 54) demonstrated that there may be irregularities on the films deposited for the first etch, which caused the high area of undercut. This could have happened for different reasons, which include an irregularity of the machine in depositing films that resulted in films with different characteristics; the presence of alien substances inside the CCP deposition chamber during the deposition of the films for the first etch, which could have fallen on the wafers and created an anomalous amount of pinholes; different handling and ambient exposure of the wafers from the first etch compared to the second etch, since the wafers on the first run were stored for a longer time between they were cleaned and had the films deposited.
Moreover, the experiment demonstrated that oxide films may only present pinholes for films thinner than 0.09 µm. For nitride films, on the other hand, it is not safe to conclude a film thickness that would be free from pinholes due to the limited data collected.