Hi,

I'm doing some research into fabrication of silicon microlenses (~400
microns diameter, ~15 microns high) by photoresist reflow followed by
transferring the resist profile into the silicon substrate with a dry
etch using CF4 / O2 plasma.

Everything is working fine, except that the surface of my lenses looks
like the surface of the moon. They are covered with circular pits of
different sizes, up to about 4 microns diameter and 100 nm deep.

I have been unable to track down the source of this problem in the
literature. If anyone knows of the origin of this surface damage, advice
would be greatly appreciated.

Attempting to isolate the origin of this damage, I have drawn the
following conclusions that may help diagnose the problem:

- Photoresist surface is fine before dry etch

- Pits are occurring in the photoresist during the dry etch and are
subsequently transferred into the silicon substrate

- Damage is independent of RF power over the range 50W-200W and ICP
power 300W-750W

- Damage still occurs when temperature of resist surface does not exceed
85 degrees Celsius during dry etch

- Damage still occurs if resist lenses are left for 24 hours before dry
etch

- If the resist lenses undergo a further high-temperature bake after
reflow (5 mins bake, temperatures of 165, 185, 205 and 225 degrees),
then the damage gets worse as the temperature of this bake is increased.

My process parameters:

AZ4562 resist

Reflow at 145 degrees Celsius for 10 minutes

CF4 conc. 20sccm

O2 conc. 3sccm

RIE powers: 50W, 100W, 200W

ICP powers: 300W, 500W, 750W

Pressure 10mTorr

I also have photos of the silicon surface, if these would help shed
light on the situation then please email me.


Thanks!

Jason Milne
Microelectronics Research Group
University of Western Australia
Perth, Western Australia