This was really interesting. I am thinking bout various work environments of people who had to show up during the lockdowns and thinking about how to both protect them and factor in the co-morbities they may bring to the table. That research is likely farther off.
Also, as I work with meat production plants with high levels of ammonia, chlorine and organic particulate matter...how do those higher levels interact with GUV and change the smog calculations
You are right that those are important additional questions:
- E.g. for nursing home, GUV disinfection will reduce many more deaths than in elementary school classrooms. But the people there are also far more susceptible to the health effects of PM. We may do a calculation in a followup paper.
- Special situations like the meat plants are again very interesting. As the composition of air is changed, there could be surprises. The ammonia and the chlorine in particular could lead to surprises. Do you know of papers or reports where that composition is quantified?
I've been following you for quite some time on Twitter, and I love your effort on promoting the airborne nature of this virus, and the ways to mitigate transmission.
Our government (The Netherlands) has _decreased_ the ventilation requirements for restaurants, bars and clubs (I think by a factor of 5!) last year, in the midst of the pandemic, and they are proud of it...
I can't afford a far-UVC installation, so have been using 254 and HEPA in our kitchen, where we "entertain" visitors (they might disagree). Also use a co2.click monitor.
I would risk more UV, especially if my filter removes other contaminants that UV might interact with.
However, near me in France I have never seen any IAQ efforts - nada.
So I also use N95 Auras when indoors away from home.
It depends on the wavelength and intensity of the light. E.g. if you have 2 uw cm-2 of light at 222 nm on average in a room, it does NOT matter whether an LED or KrCl lamp generated that light.
thank you for the great write up. When you refer to "low ventilation" (typical of residential) can you tell us what approximate physical ACH that assumes?
It is low for places with mechanical ventilation, but not for e.g. homes. E.g. our house is 0.25 ACH (working on improving that). And homes are often around that range, I think more typically 0.5 in the US, with large variability
wow -- i'm not conversant with what is typical in homes, and that is very interesting to hear. no wonder there is so much need for improvement in residential settings.
Thanks for this analysis. GUV feels like a political nonstarter. No matter how safe it actually is, "the skin damage and smog levels are tolerable" is just not going to fly with people. There will be enough objectors that no building manager is going to want to install them.
I would much rather have a CO2 monitor in every room. It's a useful proxy, nobody will be bothered, and what gets measured gets fixed.
Hmmm. I also think that we should have a CO2 monitor in every room. Effectively favoring ventilation. And then also filtration as the first supplement, which CO2 does not measure. But we could have low cost PM instruments monitoring that.
- For GUV, it seems that (other than the smog issue), it would be politically appealing: effective at disinfection, not extremely costly if done at scale, and unlikely to create opposition like masks. But indeed the smog issue changes the picture some.
Wonderful post, thank you. Can I ask what quantity of GUV is used in the calculations? I would estimate that the higher the GUV power output the higher the smog (for a set wavelength)
Yes, this is specified in the paper. We used the CDC specified amount for GUV254. Then we scaled GUV222 to produce the same disinfection rate as GUV254 (it ends up requiring less photons than for GUV254, as discussed in the paper). So the amount of GUV should be realistic of good applications in the real world.
You are right that the higher the GUV power, the higher the smog.
Thank you Jose-Luis, if my calculations are correct then your photons per second per cm-2 from the paper gives an irradiance of 3.5 uWcm-2 which over an 8 hour day would be a UV dose of 100 mJcm-2. This is within the current ACGIH limits (161 mJcm-2 for the eye, 469 mJcm-2 for skin) which apply in USA but exceeds the ICNIRP limits which apply in other countries (23 mJcm-2). Maybe it could be recalculated at the GUV exposure limits?
- Do you have links to the documents that contain those exposure limits?
- Assuming it is 222 nm, where is the exposure calculated? The light beam diverges quickly with distance, so that the mJ cm-2 are a strong function of the distance between the exposed person and the light.
- I would expect the effects to be relatively linear vs. the amount of light, but we could do a couple of extra runs. Not sure where to document it though. We are working towards other papers, but they are not close to submission. Perhaps as an appendix to the original post? Suggestions?
222 nm. We have a paper we are working on which models the UV exposure of the S aureus from the Leeds chamber Scientific Reports. It shows a spread of UV doses, maybe this could be used? I’ve emailed it to your co-author Prof Miller as I had their email address. I think we could combine our CFD/MCRT with your photochemistry. Hopefully chat offline. Cheers
This was really interesting. I am thinking bout various work environments of people who had to show up during the lockdowns and thinking about how to both protect them and factor in the co-morbities they may bring to the table. That research is likely farther off.
Also, as I work with meat production plants with high levels of ammonia, chlorine and organic particulate matter...how do those higher levels interact with GUV and change the smog calculations
You are right that those are important additional questions:
- E.g. for nursing home, GUV disinfection will reduce many more deaths than in elementary school classrooms. But the people there are also far more susceptible to the health effects of PM. We may do a calculation in a followup paper.
- Special situations like the meat plants are again very interesting. As the composition of air is changed, there could be surprises. The ammonia and the chlorine in particular could lead to surprises. Do you know of papers or reports where that composition is quantified?
I've been following you for quite some time on Twitter, and I love your effort on promoting the airborne nature of this virus, and the ways to mitigate transmission.
Our government (The Netherlands) has _decreased_ the ventilation requirements for restaurants, bars and clubs (I think by a factor of 5!) last year, in the midst of the pandemic, and they are proud of it...
I hope they will one day listen to you.
Fantastic and clear paper, thank you.
I can't afford a far-UVC installation, so have been using 254 and HEPA in our kitchen, where we "entertain" visitors (they might disagree). Also use a co2.click monitor.
I would risk more UV, especially if my filter removes other contaminants that UV might interact with.
However, near me in France I have never seen any IAQ efforts - nada.
So I also use N95 Auras when indoors away from home.
Avoided infection so far...
Does the emissions profile of GUV depend at all on the type of light source, e.g. LEDs vs krypton excimer lamps?
It depends on the wavelength and intensity of the light. E.g. if you have 2 uw cm-2 of light at 222 nm on average in a room, it does NOT matter whether an LED or KrCl lamp generated that light.
thank you for the great write up. When you refer to "low ventilation" (typical of residential) can you tell us what approximate physical ACH that assumes?
As stated in the paper, that assumes 0.3 ACH which is typical of homes. If ventilation is 3 ACH or larger, the problem is greatly reduced.
Ah - i'm sorry i missed that. Wow, 0.3 is pretty low :)
It is low for places with mechanical ventilation, but not for e.g. homes. E.g. our house is 0.25 ACH (working on improving that). And homes are often around that range, I think more typically 0.5 in the US, with large variability
wow -- i'm not conversant with what is typical in homes, and that is very interesting to hear. no wonder there is so much need for improvement in residential settings.
Thanks for this analysis. GUV feels like a political nonstarter. No matter how safe it actually is, "the skin damage and smog levels are tolerable" is just not going to fly with people. There will be enough objectors that no building manager is going to want to install them.
I would much rather have a CO2 monitor in every room. It's a useful proxy, nobody will be bothered, and what gets measured gets fixed.
Hmmm. I also think that we should have a CO2 monitor in every room. Effectively favoring ventilation. And then also filtration as the first supplement, which CO2 does not measure. But we could have low cost PM instruments monitoring that.
- For GUV, it seems that (other than the smog issue), it would be politically appealing: effective at disinfection, not extremely costly if done at scale, and unlikely to create opposition like masks. But indeed the smog issue changes the picture some.
Wonderful post, thank you. Can I ask what quantity of GUV is used in the calculations? I would estimate that the higher the GUV power output the higher the smog (for a set wavelength)
Yes, this is specified in the paper. We used the CDC specified amount for GUV254. Then we scaled GUV222 to produce the same disinfection rate as GUV254 (it ends up requiring less photons than for GUV254, as discussed in the paper). So the amount of GUV should be realistic of good applications in the real world.
You are right that the higher the GUV power, the higher the smog.
Thank you Jose-Luis, if my calculations are correct then your photons per second per cm-2 from the paper gives an irradiance of 3.5 uWcm-2 which over an 8 hour day would be a UV dose of 100 mJcm-2. This is within the current ACGIH limits (161 mJcm-2 for the eye, 469 mJcm-2 for skin) which apply in USA but exceeds the ICNIRP limits which apply in other countries (23 mJcm-2). Maybe it could be recalculated at the GUV exposure limits?
- Are you referring to 254 or 222 nm?
- Do you have links to the documents that contain those exposure limits?
- Assuming it is 222 nm, where is the exposure calculated? The light beam diverges quickly with distance, so that the mJ cm-2 are a strong function of the distance between the exposed person and the light.
- I would expect the effects to be relatively linear vs. the amount of light, but we could do a couple of extra runs. Not sure where to document it though. We are working towards other papers, but they are not close to submission. Perhaps as an appendix to the original post? Suggestions?
222 nm. We have a paper we are working on which models the UV exposure of the S aureus from the Leeds chamber Scientific Reports. It shows a spread of UV doses, maybe this could be used? I’ve emailed it to your co-author Prof Miller as I had their email address. I think we could combine our CFD/MCRT with your photochemistry. Hopefully chat offline. Cheers
You can find my email on the web, just do a search, and let's connect that way. Thanks.