Specific Process Knowledge/Thin film deposition/Deposition of TEOS/Deposition of TEOS using LPCVD
Feedback to this page: click here
This page is written by DTU Nanolab internal
DTU Nanolab has an LPCVD furnace for deposition of TEOS. TEOS is a silicon dioxide based on tetraethoxy-silane. The furnace has been installed in 1995, and it is a Tempress horisontal furnace.
On the LPCVD TEOS furnace there are two recipes that can be used for deposition of TEOS, dependent on how thick a layer that has to be deposited. The process parameters are the same for both recipes, but the speed at which the furnace is opened after the deposition is different. When using the recipe "TEOSSLOW" the furnace is opened very slow, and it is therefore a good idea to use this recipe for deposition of thick TEOS layers (1000 nm - 2000 nm) to prevent cracks in the TEOS film and particles in the furnace, if the wafers are cooled too fast. For deposition of thin layers of TEOS (0 nm - 1000 nm) the recipe "TEOSPNE" can be used. TEOS layers thicker than 2000 nm have to be deposited over more runs.
Process parameters for the standard recipes on the LPCVD TEOS furnace:
| Recipe name | Wafer size and number of wafers | TEOS thickness [nm] | Temperature [oC] | Pressure [mTorr] | TEOS gas flow [sccm] | O2 gas flow [sccm] |
|---|---|---|---|---|---|---|
| "TEOS" | 4" wafers
1-15 wafers in a run Only 7 wafers recommended if a good uniformity is needed |
0 - 1000
Open furnace using the "STANDBY" recipe 1000 - 2000 Open furnace slowly using the "STB-SLOW" recipe |
725 | 175 | 50 | 30 |
Expected results when using the standard recipes on the LPCVD TEOS furnace:
| Recipe name | Deposition rate [Å/min] | Refractive index@630nm | Thickness variation on wafer | Thickness variation along the boat | Stress |
|---|---|---|---|---|---|
| "TEOS" |
|
1.43 - 1.44 | Center wafer: 0.44 %
Christian Østergaard, DTU Nanotech |
Center point: 8.40 %
Christian Østergaard, DTU Nanotech |
~ 60 MPa (compressive) |
The deposition rate changes a little over time, but generally it is fairly good from run to run. The last measured values of the TEOS thicknesses can be found on the TEOS page in the furnace logbook.