Interesting features
In an effort to measure the effects of bsa adsorption on pnc-Si, I made some samples from SC 073 (a good wafer for separations) and with the help of Graham Marsh (Waugh Lab), looked at them using AFM. We did our first studies of the samples in water, just in case that drying somehow distorted the adsorption. Immediately we noticed a bunch of little raised spots on the pnc-Si that we attributed to dirty RO water or clumpy BSA. With the help of Anant, we then looked at the same samples without bsa or any contact with water. Again there were little spots everywhere:
Was this something depositing on the wafer over time? A problem during the fabrication process? Why hadn’t we noticed this before? This prompted us to look at the TEM:
You can notice little gray spots on this image if you look carefully. They’re definitely not crystals, and they’re about the same size as the pores. Thinking that this was an anomoly, I scrounged up some other samples of from SC 086:
Ooops. Now the problem is worse. If you check out the TEM, the bumps are very obvious.
What could be causing this? It doesn’t affect separations, and it seems that we haven’t been having this problem recently.
Seems like both 073 and 086 were on Siltronix substrates. If I remember, we had quite a few problems with these wafers. Can we image a membrane on a Virginia wafer around the time these were made to rule out a substrate problem?
These look like colloids. These are commonly seen after CMP (Chemical Mechanical Polishing) process. Since these are Siitronix substrates, it is not surprising. We believe that these wafers had been processed before and were re-worked.
To address the Siltronics theory raised at NRG: Nakul went back and looked to see which 15 nm porous samples we gave him during our TEM work. We recorded giving him samples from TEM 039 and 042 – both were Virginia Semi wafers. However, he may have received additional samples through another route.
The features on SC086 look much different to me, so I’m not sure whether they are the same thing. With these samples being so old, it’s unlikely that we will be able to identify the exact problem, unless the substrates are indeed the issue. How does the height in AFM compare to the width in TEM? It looks to me like they are somewhat flattened – more like a deposit than a colloid. The relatively low contrast in TEM would point toward an organic material of some kind.
We are confident that the newer membrane material does not have this topography, correct?