Microporous MgF2 Burst Pressure Fabrication and Results

ByGreg Madejski
Summary Statement:
We have doubled our MgF2 yields (compared to nanoporous MgF2) by moving to thicker microporous membranes. While the increase in strength is more modest than initially expected, we have plenty of substrates for use in cell culture.
Statistics:
1. Burst Pressure (most colorful)
2. Yield after Etching
3. Yield after Evaporation
4. Yield after Dicing (least colorful)
Window [mm] Average Std Dev Count
0.1 21.02 8.10 14
0.3 8.18 5.62 17
0.5 4.33 1.66 10
0.7 2.87 0.94 11
0.9 1.23 0.23 3
1.1 1.32 0.33 6
Discussion

Our burst pressure wafer (Wafer #4632) has a number of window sizes (0.1, 0.3, 0.5, 0.7, 0.9, 1.1 mm square windows) and varying porosity (low = edge, high = center). I was targeting  200 nm thick MgF2.

I began by splitting the wafer into 4 pieces (quads) so that they would coat evenly on our evaporator.

Microporous Wafer after dicing (black edges). The central blue square demarcates the change between low (out of square) and high (inside square) porosity. The blue colored squares are broken chips

Each quad had a separate deposition, and the film thicknesses (measured with profilometer) varied a little:

Deposition settings

  • 6KVOxide Recipe
  • 1 mA beam current
  • 2-4 Angstroms/sec
  • 200 C
  • Platen Rotation on
  • sub 5-e5 torr base vacuum
Top Left (1) 165 nm
Top Right (3) 174 nm
Bottom Right (4) 173 nm
Bottom Left (2) 210 nm
Microporous wafer after evaporation step. Each of the blue colored chips are broken.

These chips are then ready to be etched

 

Etching Settings

  • 2.5 mTorr Oxygen (8.9 Setpoint)
  • 47.5 mTorr CHF3 (26.9 Setpoint)
  • 110/10 W Fwd/Ref
  • ~100 mTorr Etching Pressure
  • 5.5e-5 torr base pressure
  • Samples face down on carrier wafer
  • 120 second etching time

Each Quad was etched with these settings. The chamber was first cleaned with a 5 min, 200 W, 100 mTorr Argon plasma, then seasoned with the etch recipe for 5 minutes.

Yield after Etching. The blue colored squares are broken. Large area windows broke more than small area windows.

Inspection of the chips revealed some interesting cracking structure. These were seen across the whole wafer, revealing a difference between central and edge chips (porosity difference). Maybe this explains the noticeably lower yield in the central region?

Overall Yield was 51%, where 89% made it through evaporation and 57% made it through the etch. As with the first burst pressure wafer, there was strong dependance on window size, but also the higher porosity material was less likely to survive (about 2-3x poorer yield).

This compares very favorably to the thinner 50 nm nanoporous process which had 22% yield of mostly smaller windows.

Window Edge Chips Start Edge chips after Evap Edge chips after Etch Yield After Evap Yield After Etch Etch Process Yield
0.1mm 30 30 26 100% 87% 87%
0.3 39 39 38 100% 97% 97%
0.5 30 29 24 97% 80% 83%
0.7 32 29 23 91% 72% 79%
0.9 31 9 3 29% 10% 33%
1.1 36 31 9 86% 25% 29%
Sum 198 167 123 84% 62% 74%
Window Center Chips Start Center Chips after Evap Center chips after Etch Yield After Evap Yield After Etch Etch Process Yield
0.1 35 35 30 100% 86% 86%
0.3 28 28 17 100% 61% 61%
0.5 36 35 17 97% 47% 49%
0.7 34 31 9 91% 26% 29%
0.9 34 31 2 91% 6% 6%
1.1 28 24 2 86% 7% 8%
Sum 195 184 77 94% 39% 42%
Total Chips Chips Left after Evap Chips Left after Etch Yield after Evap Yield after Etch Etch Process Yield
Total Sum 393 351 200 89% 51% 57%

 

After this quantification, we had to see how strong the membranes actually were. The chips were clamped into a burst pressure rig, and then destroyed using a slow nitrogen flow, ramped up by hand (0.2 PSI/sec).

 

1.1 mm MgF2 microporous edge chip in burst pressure clamp.

 

 

Burst Pressure Results. Everything yellow-green colored is above 5 PSI, which is the apocryphal threshold for a practical chip. Again there is a window size dependance, as larger windows have smaller burst pressures

I tabulated the burst pressures for each porosity. There were no significant differences (t-test, 2-tails, 95% confidence level, unequal variances).

 

 

Center
Window [mm] 0.1 0.3 0.5 0.7 0.9 1.1
Average 21.1 7.2 3.7 3.0 1.1 1.0
Std. Dev 10.3 4.1 1.5 1.2 0.4
n 9 5 5 7 1 2
p-value (Different?) 0.951 0.587 0.289 0.494 0.122

 

Edge
Window [mm] 0.1 0.3 0.5 0.7 0.9 1.1
Average 20.9 8.6 4.9 2.7 1.3 2.1
Std. Dev 0.8 6.3 1.8 0.4 0.3 1.2
n 5 12 5 4 2 5

 

Finally, combining all chips:

Window [mm] Average Std Dev Count
0.1 21.02 8.10 14
0.3 8.18 5.62 17
0.5 4.33 1.66 10
0.7 2.87 0.94 11
0.9 1.23 0.23 3
1.1 1.32 0.33 6

 

Error bar is 1 standard deviation of microporous MgF2 material. The new material is stronger than the 50 nm thick nanoporous material and matches much more closely to pnc-si.

 

 

Similar Posts