Microporous Magnesium Fluoride Nanomembrane Becomes Stronger as the Film Thickness Increases

The title really says it all. I have been attempting to make a better cell culture substrate with the Magnesium Fluoride Films. Previous posts have shown the overall structure of the film, and now I am trying to quantify the strength with a burst pressure measurement. In order to make a direct comparison, I needed to utilize the same window/pore geometry, and only vary the film’s thickness. I used wafer PFI02 (120 nm silicon nitride, 0.4 micropores, 1:1 pitch, five 0.1×3 mm slots).

Here, I’m comparing a 50 nm thin film process (actually came out about 45 nm) to a 200 nm thin film process.

50 nm Process

• Evaporate MgF2 onto 0.4 micron pore, 12o nm thick Si3N4 substrates
  • 250 C
  • 2-3 Angstroms/sec
  • Platen Rotation
• Etch with RIE, backside to remove Si3N4
  • CHF3:O2 (90%:10%)
  • 100 W Net Power
  • 100 mTorr etch pressure
  • Etch for 120 sec

200 nm Process

• Evaporate MgF2 onto 0.4 micron pore, 12o nm thick Si3N4 substrates
  • 150 C
    • The different temperature is to mitigate the larger overall amount of film stress generated by the thicker film
  • 2-3 Angstroms/sec
  • Platen Rotation
• Etch with RIE, backside to remove Si3N4
  • CHF3:O2 (90%:10%)
  • 100 W Net Power
  • 100 mTorr etch pressure
  • Etch for 120 sec

The burst pressures then are (slowly ramping up with N2 gas, ~1 PSI every 5 sec):

Increasing the thickness 4-5x improved the burst pressure strength by ~8x. There are yield improvements too; many of the broken chips for the 200 nm process had more windows than that of the 50 nm process. Here I only got 3 chips for each treatment, but I processed 9 chips for the 200 nm process and about 20 chips for the 50 nm process. This data suggests that I am on the right path to make more robust cell culture membranes.