So you have your die perfectly clean. But how do you keep it clean?


Skin is a large component of dust. One way to determine if you have skin contamination is to confirm that is responds to polarized light. Example unpolarized:



Some materials tend to be cleaner than others. Glass and teflon tend to be very clean materials. Plastics on the other hand you often have to be more careful with.

Case study

I had issues with streaks showing up on samples that for a long time I couldn't figure out the source. Example:

I verified this wasn't from the dies by taking each and putting a few drops on a clean glass slide using a fresh pipette. I got the same result using hardware store acetone or Duda Diesel (fairly pure) although the hardware store was a little worse. Originally I thought it was small particles and thought maybe it was dust or pollen. I took some pollen from the yard and discovered it has more of a spiky structure:

Which does not match what I was seeing. Just to be sure it wasn't related to the acetone I soaked it. It loses its color but otherwise remains the same. Polarized version looks pretty nifty:

Additionally pollen responds more to polarized light and this doesn't. Upon closer examination I also noticed it had more of a residue than a solid appearance. I tried taking acetone from the bottle using a glass pipette and found that it did not leave a mark. This seemed to indicate it was from the pipette.

Upon looking at a fresh the pipette under the microscope I noticed it was actually quite dirty. I got them used and so they may have been contaminated before I got them (and that's why there were ditched to eBay) or during storage. However, the pipette I had been using initially had been used many times and I had been very careful to rinse it out with clean acetone. So, the contamination should have gone down after the pipette had been rinsed out.

I now believe it may also be related to the pipettes being made of polypropylene. While polypropylene itself appears to be very resistant to acetone it is sometimes alloyed with other plastics. I believe it may have had some impurities that were being slightly dissolved. When the acetone evaporated it would leave little dots from where the plastic solidified.


Plastic antistatic tweezers have things embedded in them to conduct. These can flake off, especially if partially dissolved by things like acetone or acid. Additionally, if the die is still wet it could slightly dissolve the tweezers and contaminate the sample. This die was still slightly wet with acetone and took some tweezers with it:

You will probably not notice the material missing from the tweezers but may feel the die stick. If a die sticks to tweezers there is a good chance something is going to come off.

Air quality

Industrial setups use cleanrooms for quality control. Dust can get in the way of imaging but is unlikely to harm the device. Therefore, you don't need to go nuts like for fabrication but its not a bad idea to try to keep things reasonably clean. Closing windows should be sufficient for most purposes. In the end you'll have to experiment with your particular setup to see how much air quality control you actually need.

Air filtering

The air in my room varies from about 300k to 1M 0.5 um particles per / ft^3. I have recently been experimenting with a Honeywell 50250 air filter. This has been shown to get my room down to 5k in 15 or 20 minutes despite still having dusty corners and my desktop on. This level of control isn't necessary for small runs but can become an annoyance for long imaging runs. For example, GPUs tend to be large, on the order of 1 cm X 1 cm. At 50X these can take 12 hours to complete which actually accumulates a fair amount of dust.

My desktop computer makes the air quality go from about 30k when off to about 600k when on. It settles down after a while but is still very bad for cleanroom conditions.


Low budget operations can use a Dylos DC1100 air quality monitor. The pro model is calibrated for 0.5 um particles which makes it nice for checking your capabilities against a cleanroom. For example, a class 10,000 cleanroom refers to < 10,000 0.5 um particles per ft**3 (who's bright idea was it to mix metric and english units?). The serial version seems to just be a bring out that you could probably do yourself. Mine wasn't attached very well and looks like it was hastily Dremeled out, something I could have done at home. I don't know how to change the calibration though and I would probably go with the pro model unless you need a truckload of them and would rather calibrate it yourself.

contamination_control.txt · Last modified: 2013/10/20 10:59 (external edit)
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution 4.0 International
Recent changes RSS feed Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki