Low temp/ramp rate anneal
Last week I showed some interesting images on the effects of annealing at a low temperature (700 C) with the susceptor. Below are images from a wafer that was annealed at 700 C with a ramp rate of 10 C/s outside the susceptor.
Immediately following RTP, I saw that the outer part of the wafer had not changed color (the amorphous to crystalline transition causes a shift in film color). My suspicions are confirmed by these two images from the inside and outside of the wafer. The inside sample shows a crystalline background with large, irregularly shaped pores. The outside sample has zones of crystallize and amorphous material. Again, the pores seem to grow along the edges of the crystalline pockets (although not always the case, see below).
The largest pores always seem to be “unbounded” in at least one direction i.e. not bordering a crystalline zone. The smaller pores are surrounded by crystalline material (the growth stop).
I would like to see what happens when a film is annealed at 700 C and subsequently RTPed again at a higher temperature. Could this be the way to open even larger holes?
The difference in crystallization from the center and the outside of the wafer makes me think that the RTP isn’t as uniform. Is this the case or is something else going on here?
Yes, we think that outside of the susceptor the temperature on the wafer is not as uniform. This could be because the bare wafer comes in contact with two thermocouples which creates a heat sink at the contact points. There might also be non-uniformity in the lamps.
What was the measured non-uniformity from the center to the edge of the wafer? What does the vendor guarantee the tool performance to be? What does the process require?
We’re not sure what the exact non-uniformity is. We need a calibration wafer with thermocouples bonded to it in order to get that information (something we are planning on doing).
Ideally, we want uniform heating across the entire wafer; which is why we use the susceptor. The problem is that we are limited in the ramp rate if we use the susceptor.
The RTP is powered by three sets of lamps that we have control over independently. We might be able to tune the settings so that we can pull in the uniformity…