Question: Which NO2 Sensor for low-concentration measurement?

I can only make a guess as to why they start at 0.2 ppm.

With most analytical measurements, you want a ten to one signal to noise ratio to be sure the numbers are correct. So if the limit is 2 ppm, the instrument needs to go down to at least 0.2 ppm.

Sounds like what is going on here.

CC @guolivar in case they might be able to share from their in depth knowledge of various types of air pollution sensors.

Thanks for the heads up @liz

I've been working in air quality research for 20 years and I am yet to find a reliable "low-cost" NO2 sensor. There are a few that make decent attempts like Aeroqual (https://www.aeroqual.com), SPEC (https://www.spec-sensors.com) and 2B (https://twobtech.com) but all of them have caveats, high detection limits, low resolution and most commonly, very specific operating parameters (temperature) that make them challenging to be used for ambient measurements.

If you have the budget, then you should go for the regulatory grade instruments (ThermoFisher, Ecotech, Emerson, etc), they are very sensitive and extremely reliable but they are EXPENSIVE ... think about US$20000 each unit. You can connect those to an arduino but they are also BIG ... they're meant to go on 20" wide racks.

So, assuming now that you didn't really mean "money is no obstacle" then here are my thoughts on the three devices that I have used over the past 10 years:

1.- Aeroqual (https://www.aeroqual.com/product/nitrogen-oxide-analyzer-module) . Well constructed, "plug and play" and I assume a lot better than when I last used them. The downside is their performance in low-concentration environments. They're great in road tunnels but they are a bit noisy in "clean-ish" outdoor environments. They offer an operating mode with an O3 sensor that helps with noise and they have had some good results using them in the "wild". Their cost is not cheap but not $20000 ... more like $10000

2.- "2B" (https://twobtech.com/model-405-nm-nox-monitor.html). They used to have a cheaper/smaller device that measured NO and that with an catalytic converter could measure NOx (and calculate NO2 as NOx-NO) but they apparently discontinued it ... a shame because it was easy to use BUT it was not as robust. I don't have direct experience with their new device but the measuring principle is sound and I would trust their results. The price is on the high side. They're competing with the chemiluminescence boxes so they're cheaper than those but I don't expect them to be that much cheaper.

3.- SPEC (https://www.spec-sensors.com/product/digital-gas-sensor-module-no2/). These little sensors are great but they are very temperamental. I ran some of them in a road tunnel and they compared extremely well (R2>90%) with our big chemiluminescence boxes. I've also used them indoors and their response is very good... true to their specifications. HOWEVER, once I put them outdoors, they were not happy. All electrochemical sensors have a response that changes with temperature so outdoors, with large and fast temperature changes (nighttime ~15C, daytime ~30C) their response was completely dominated by the temperature, to the point that when I corrected for temperature, I ended up with just noise. On the other hand, they're small and really easy to integrate with a DYI project (serial communication for the digital module and they also have an analogue version) and at US$75 each they're really really attractive if you don't have $100000 in your budget.

Please, please, if you do find a NO2 sensor that operates in the range 0-200ppb, is small and is not that bothered by temperature, let the air quality research community know, we've been looking for something like that since the acid rain days!