PM 2.5 fine dust / particle measurements status?

[Matthias Kauer]

Hi,
I am wondering what the status is regarding the measurement of PM2.5 or PM10. My application would be to measure general air quality with respect to industrial pollution, driving, forest fires (as in Indonesia -> pollution hundreds of km away in Singapore)

The wiki claims you can measure PM 2.5 in one point. With the caveat that someone adds what you're actually measuring only correlates to that.
Then I find the discussion page: http://airqualityegg.wikispaces.com/Dust+Sensor+design
But that doesn't offer any conclusion.

I find plenty of promising threads discussing various sensors. In particular good posts from Gustavo Olivares.
Could someone guide me a bit as to where this all lead? Is there now a definite recommendation for a sensor that I can hook up to, e.g., an Arduino and get somewhat decent results or is still impossible and I absolutely require a 5000$ equipment?

Thanks for your help,
Matthias

[Matthias Kauer]

Actually this seems to be the most recent thing.
https://groups.google.com/forum/#!topic/airqualityegg/L5j7CO1dZQA

So the Shinyei sensor is currently the best choice, although David Holstius says that the measurements need to be taken with a grain of salt.

Am I missing something or is this currently the best option?

[David Holstius]

It's a good option, and I would encourage anyone who has one to try and calibrate it, using a good reference method, with your aerosol of interest. One option may be to locate people in your area who can rent or loan you some pre-calibrated $5000 instruments for a couple of days.

Why does it matter (to use the aerosol of interest)?

Remember the Sony Bravia commercial?

http://www.youtube.com/watch?v=0_bx8bnCoiU

Imagine that but with basketballs, golf balls, and BB pellets. (In real life, there's more variation in particle sizes. But, there really are three main "sizes": ultrafine, fine, and coarse. For practical purposes, particles tend to lump into one category or another, and they differ in diameter by about an order of magnitude each.)

Different aerosols have different proportions of basketballs, golf balls, and BB pellets. For example, soot is going to be coarser (more basketballs), and fresh diesel exhaust will have lots of ultrafines (BB pellets).

Now imagine it's your job to estimate the concentration (either mass or count, your choice) of the balls bouncing by.

Rule #1: You have to stand on the side of the street.
Rule #2: You have to look straight ahead, through a cardboard tube.
Rule #3: All you can do is count. You can't differentiate between balls with different colors, diameters, or compositions.
Rule #4: Did I mention you can't see the BB pellets? They're too small.
Rule #5: You have to ignore the basketballs. (That's the "2.5" in "PM2.5".)

That's the game that particle counters play. With a light-scattering device, like a PPD42NS, you are shining a flashlight rather than looking directly through a tube, so you're limited in slightly different ways.

If you want to estimate masses from counts, you're going to have to make assumptions about the relative proportion of heavier particles, which is different for different aerosols (cigarette smoke vs wildfire smoke, for example). Hence the need to calibrate against your aerosol of interest.

Even if you want to just estimate counts, you're going to have to be careful about *which* particles you claim to be counting. The most numerous (BB pellets) are also not directly visible to optical instruments. (Condensation counters are a special case; they add an alcohol mist, which causes ultrafine particles to aggregate into visible clumps.) So your lower limit of detection (in terms of particle diameter) really matters.

Again, I completely encourage everyone who can to experiment with the Shinyei, Dylos, Sharp, and any other cheap optical sensors. It is really important to get more data on PM concentrations. We just need to be clear about what kind of PM we're gathering data on!

David

[Michael Heimbinder]

Hi David. Thanks for the excellent description of the different types of aerosols. I definitely learned a few things. As you're recommending, I've been running comparison tests between the Shinyei and Dylos. Do you have recommendations for generating aersols that are predominately composed of ultra fine, fine, or coarse particles using readily available materials. I'd like to test how the different instruments respond to each mix of particles. Currently I'm only testing with cooking oil+meat and cigarette smoke.

-Michael-