Oxford ICP Recipes
The most important parameters for ICP recipes are probably gases used, their flow rate (in sccm), RF & ICP Generator Power, etc. Other parameters might have less impact?
In other words, change these first if you want to finely tune your etch process. Try not to chase ghosts.Check logged etches here: Oxford ICP Logbook
In the Recipe Name column, attach a link to a picture of ICP computer screen if possible.
As well as a link to the Reference (paper, document, etc) where this recipe is found, if possible.Etch calibration can be done using kapton tape as a hard mask (less recommended and less accurate) or by patterning your sample with larger features. Dektak or SEM are both useful tools for figuring out your etch rates. If you are using Dektak, note that the measured number for a larger etch will be 50-100nm larger than for a grating etch.
Recipe name | Gases (unit: sccm) | Power (W) | Etch-able materials Etch Rates (nm/min) | Notes |
|---|---|---|---|---|
(In Based etch Etch 20deg) | Cl2: 4.0 CH4: 9.0 H2: 6.0 N2: 3.0 BCl3:2.0 | P_RF = 150 P_ICP = 1500 | GaAs, GaSb, AlAs, AlSb, InAs, InSb and respective ternaries/quaternaries Etch rate is material dependent. Estimates (nm/min) GaSb 300 InAs/InSb/InAsSb 500 AlAs/AlAsSb 100 SiNx 80 SiO2 100 | Also etches hard masks, SiNx and SiO2 If the process seems unstable, increase the set pressure to 3.5-4, but you shouldn't have to do this. This etch includes CH4 and H2, which produce a polymer on the sidewalls while etching which helps pseudo-passivate the surface. Important: The etch rates aren't linear. This process tends to heat your sample pretty significantly, so any etch longer than 2 minutes may not be accurate and may leave indium residue behind (if you are etching indium-containing layers). It's recommended that you give a 1 minute pause after 2 minutes, but if you are having trouble, you may want to pause after every minute. It's also recommended that you calibrate this etch whenever you work with a new sample. |
In Based etch Etch 20deg o2 clean | See above | See above | See above | This etch is the same as above, but adds a 60s O2 clean to the beginning, a 5 second gas stabilization phase before etching, and a 30s O2 clean to the end. This etch is recommended instead of "In Based 20deg etch" |
In Based etch wait etch o2 clean NM | See above | See above | See above | This has repeated etching steps and a 1 minute pause for the sample to cool. |
In Based AK Cl2 Ar - 2 | Cl2: 5.0 Ar: 10.0 N2: 10.0 | RF 100W
ICP 750W | GaAs, GaSb
GaSb 200 | This etch does not prevent the formation of indium chloride residue, but is still useful for etching GaSb. Not recommended for any indium-containing layers. |
Cl2: Ar = 1:1 |
| Need to test out, AlSb anisotropic etch |
| |
NCM In-based BCl3 CH4 - Calibrated | Cl2: 2.5 CH4: 9 H2: 6 Ar: 5 BCl3: 7.5 | P_RF 150 P_ICP 1000 | ~280nm/min, material dependent. Rates not calibrated for individual materials | Good for highly anisotropic sidewalls. Noah got ~80 degrees when he calibrated it last. May result in some indium residue.
TODO: Try CH4 and H2 at 4 and 10sccm respectively |
not saved | Cl2: 27 N2: 3 | 150 W RF 500 W ICP | 1300 nm/min for GaP |
|
