[LeifCarrotson]

> many many tons of particulate matter every day

Let's do the math.

Guidelines state that the limit on PM2.5 particulates is approximately 10 micrograms per cubic meter for an annual average. Xian, where the tower was installed, is currently (http://aqicn.org/city/xian/) under 231 ug/m^3 of pollution. The tower is supposed to process 10 million cubic meters of air per day.

    10^7 m^3 x 231 x 10^-6 g/m^3 x 10^-3 g/kg x 0.15 = 231 x 0.15 x 10^-2 =>

0.34 kg of particulate contamination removed per day. Not many tons. But someone check my math, please - that seems impossibly low, unless combustion is cleaner and generates lower quantities of PM2.5 particulates than I'm imagining. I did assume that it only removes PM2.5 contaminants, ignoring larger dust particles and PM10 pollution.

Let's also check the amount of air it should be processing. It's about 100 meters tall, and intended to cover 10 square kilometers. We'll conservatively assume that this volume represents the total quantity of air it needs to process. Ihe volume is:

   10 km^2 x 10^3 m/km x 10^3 m/km x 10 ^ 2 m = 10^8 m^3

or 100 million cubic meters, so it's intended to process one tenth of that volume per day.

The math still seems low. Micrograms per cubic meter are hard to intuit.

[kragen]

0.35 kg of particulate contamination removed per day; you rounded incorrectly, but the rest of your math is right.

There's the issue that it's probably removing nearly all of the particulates from the air it actually processes, say over 90%, which works out to 2.1 kg. It doesn't reduce the PM2.5 level in the area by 90%, but only by 15%, because it's only able to process some 16% of the air in its area during whatever the relevant time interval is, which seems like a good match to your 10%-per-day.

However, you got the volume of that area wrong: it's 10⁹ m³, not 10⁸ m³. That suggests that the natural lifetime of PM2.5 particulates is closer to a week than a day.

As for jhugg's question about how the filters are cleaned, a common way to remove particulates from flue gas or indoor spaces is using electrostatic precipitators, which are flat plates at a high voltage. The voltage sticks the particulates to the plates, and when the coating is thick enough, you close the valves to stop the gas flow and vibrate the plates to unstick the dust, and it falls out the bottom of the filter into a bucket (or, say, a truck). Other systems clean the plates with water (either spray or immersion) and possibly soap or lye.

[LeifCarrotson]

Thanks for verifying my result and catching my mistake. You're right, it's a volume of 10 x 10⁸ cubic meters, which is 10⁹.

That's what I get for interweaving a mantissa of 10 with scientific notation! Too late to edit now, but people should see this comment thread.

[nullbyte]

I love HN comment like this. Seems like there's an expert in everything here!

[kragen]

I am not an expert in air pollution control systems, or even an amateur. I don't even have an air filter installed in my house. I read some Wikipedia articles over the last couple of years, Googled, and watched some videos on YouTube (including "how-to" videos on the care and feeding of electronic air cleaners.) That doesn't make me an expert. That makes me some sort of approximation of a journalist — and if you've ever read those people writing about something you know about, you know how much they get wrong.

[AstralStorm]

Considering how a lot of China electricity is generated by coal plants, electroscrubbing the filters might not be the best way.

Three are alternative chemical processes to clean up such filters.

[OskarS]

Using your numbers, I get the same thing: http://www.wolframalpha.com/input/?i=(10,000,000+m3)+*+(0.15...

The microgram is a very small unit :) I mean, a gram is already pretty dinky, a millionth of it is even tinier! The whole point of these particles is that they're miniscule and can pass through tissue other particles can't, so maybe it's not so surprising that it's not a massive amount when you pile them up.

[jhugg]

That's really interesting. Maybe the pm2.5 is less visible than the PM10, but maybe I'm just way way off.

As for the volume of air, I'm most curious where the 15% measurement was taken and how consistent it was. How far from the tower? How many different places were measured? How many days are in the sample? What's the P-Value?

But still, thanks. Makes this interesting to think about.

[OskarS]

I think PM10 rates are in similar orders of magnitude, in the "micrograms per cubic meter" range, so i don't think it would be that different. Maybe an order of magnitude larger, but not much more.

I agree, this is very surprising. Before I saw LeifCarrotson's calculation, I would have thought "tons" was the right unit as well. There's so much polluted air, you'd think that it would add up to some significant amount of matter, but apparently not.

[jhugg]

Well, 10km^2 is pretty tiny in a country the size of China, and I'm not sure how high up the pollution goes (higher than buildings but lower than planes).

So it's still a lot of pollution, just much less actual mass than I would have guessed.

[staplers]

Puts into perspective how much PM burning coal puts into the atmosphere..

EDIT: In 2016, the United States burned roughly 728 million tons of coal, enough to fill a typical railroad car every 4 seconds