COVID-19 Emergency - Using UV Lamps to sanitize Face Masks overnight

[Zero999]

Alcohol unfortunately can not be used to disinfect masks.  It destroys a charged layer in the N95 masks that attracts virus particles.  I wrote up a proposal for building a UV sterilizer for mask a couple days ago.  Anyone who is interested please E-mail me:

https://drive.google.com/open?id=1wF4yYgSE1n0Jqewb8JdwosTRhx0X89wF

This is a Google Drive link.

It is a difficult problem because UV can also potentially destroy the elastomers in the mask so that the mask no longer fits well on the wearer.   There is also cumulative damage to the filtration capability.  But it presently seems the most promising approach.   A paper in 2015 studied this:

W.G. Lindsley, S. B. Martin Jr., R. E. Thewlis, K. Sarkisian, J. O. Nwoko, K. R. Mead, and J. D. Noti,  “Effects of Ultraviolet Germicidal Irradiation (UVGI) on N95 Respirator Filtration Performance and Structural Integrity,” Journal of Occupational and Environmental Hygiene, v 12., pp. 509-517 (2015).
https://www.researchgate.net/publication/274090116_Effects_of_Ultraviolet_Germicidal_Irradiation_UVGI_on_N95_Respirator_Filtration_Performance_and_Structural_Integrity

There's a crowdsourced actively maintained document of resources at http://ppereuse.com/ .

Best,

Dan

The lamp mentioned in your proposal produces ozone, which is good because it will not just sanitise the surface, but penetrate more deeply. The downside is ozone can be more damaging to fabrics and elastomers.
https://www.amazon.com/AGAWA-Ultraviolet-Disinfection-Sterilization-Germicidal/dp/B085CGGQFR/

I wonder how good the cheap LED germicidal lamps are? The blurb for the one linked below says it doesn't produce any ozone, so no good for masks, as it'll only sterilise the surface. It's inexpensive, considering the price of discrete UVC LEDs from reputable distributors. I accept distributors have lots of overheads, but I'm cynical of really cheap things.
https://www.amazon.com/dp/B085VWPT9C/ref=psdc_2314207011_t1_B085CGGQFR

[NiHaoMike]

Doctors scramble for best practices on reusing medical masks during shortage: "Referring to a paper published in the journal Annals of Occupational Hygiene in 2009, among others, the researchers compared and contrasted ... different methods for sterilizing N95 masks, many of which were ineffective."

Two were, however: "70 C / 158 F heating in a kitchen-type of oven for 30 min, or hot water vapor from boiling water for 10 min...."

Wouldn't a kitchen oven, which is designed for much higher temperatures, overshoot badly? (That's a common problem seen when trying to use a kitchen oven for drying 3D printing filament.) I suppose if you get a "dumb" oven designed for 240V and run it on 120V it would work a lot better.

Perhaps the simplest solution is to just put them in a box and just leave them for the 9 days or so (which is turning out to be quite conservative) it takes for the virus to degrade on its own. It would also be interesting to see how long it takes to destroy the virus at lower temperatures like 120F. (The 9 day figure was under near ideal conditions, at room temperature it's more like 3 days.)

[james_s]

Some kitchen ovens are better than others at regulating low temperatures. If you were willing to modify the oven you could install a PID controller to precisely control the temperature at any level you want.

[Mechatrommer]

overshoot badly?

the simple trick is to undershoot the thermostat setting by a bit. but i dont think covid will be happy with this overshoot behaviour.

[Zero999]

overshoot badly?

the simple trick is to undershoot the thermostat setting by a bit. but i dont think covid will be happy with this overshoot behaviour.

If a kitchen oven does overshoot, I've never had a problem with it. I frequently use my oven on the 60^oC setting to dry my shoes with no problems. I'd hope a decent oven will be able to leaven bread, on its lowest setting, without killing the yeast.

[jeremy]

See https://www.nebraskamed.com/sites/default/files/documents/covid-19/n-95-decon-process.pdf

And https://m.box.com/shared_item/https%3A%2F%2Fstanfordmedicine.app.box.com%2Fv%2Fcovid19-PPE-1-1

Stanford research article very clearly and explicitly states: “DO NOT use alcohol or chlorine based disinfection methods” for mask disinfection.

[NiHaoMike]

I'd hope a decent oven will be able to leaven bread, on its lowest setting, without killing the yeast.

Bread dough contains a significant amount of water that allows it to survive a lot of overshoot just fine. Plastics, especially thin plastics, overheat easily as often seen happening from those trying to use a kitchen oven to dry 3D printing filament.

[Stray Electron]

I wonder how good the cheap LED germicidal lamps are? The blurb for the one linked below says it doesn't produce any ozone, so no good for masks, as it'll only sterilise the surface. It's inexpensive, considering the price of discrete UVC LEDs from reputable distributors. I accept distributors have lots of overheads, but I'm cynical of really cheap things.
https://www.amazon.com/dp/B085VWPT9C/ref=psdc_2314207011_t1_B085CGGQFR

    FYI.  Germicidal lamps emit 254nm UVC radiation. Usually they are mercury lamps and mercury emits both 254nm and 185nm radiation. the 185nm radiation is what generates ozone and the 254nm is what directly kills viruses. However some germicidal lamps use a doped envelope (aka bulb) that block 185nm so those lamps do not generate ozone. A moral of the story is that if you want ozone then make sure that you use a lamp that doesn't have a doped envelope.   Depending on the doping agent, the envelope on a doped lamp may have a greenish cast to it if you look at it in daylight.

[Stray Electron]

Alcohol unfortunately can not be used to disinfect masks.  It destroys a charged layer in the N95 masks that attracts virus particles.

   Dan,  a few questions. Is that true of all types of alcohol?  Also couldn't "alcohol" still be used to disinfect the non-N95 masks?  The charged layer is activated charcoal isn't it?  If so could the alcohol be baked or live steamed out of the charcoal layer? (Live steam is what is used to "activate" charcoal.)

[Mechatrommer]

Plastics, especially thin plastics, overheat easily as often seen happening from those trying to use a kitchen oven to dry 3D printing filament.

have you done that? looks like you experienced. i did it and yes the old filament stock i tried to dry sagged quite badly but not it become unusable (setting is something like 100degC iirc). the next roll i dried i reduced setting to 60degC and it works. filament brittleness gone and become usable again. for things like paper, tissue, mask etc... i dont think they will combust at 100degC, even if overshooted to say like 150degC.

[Zero999]

I wonder how good the cheap LED germicidal lamps are? The blurb for the one linked below says it doesn't produce any ozone, so no good for masks, as it'll only sterilise the surface. It's inexpensive, considering the price of discrete UVC LEDs from reputable distributors. I accept distributors have lots of overheads, but I'm cynical of really cheap things.
https://www.amazon.com/dp/B085VWPT9C/ref=psdc_2314207011_t1_B085CGGQFR

    FYI.  Germicidal lamps emit 254nm UVC radiation. Usually they are mercury lamps and mercury emits both 254nm and 185nm radiation. the 185nm radiation is what generates ozone and the 254nm is what directly kills viruses. However some germicidal lamps use a doped envelope (aka bulb) that block 185nm so those lamps do not generate ozone. A moral of the story is that if you want ozone then make sure that you use a lamp that doesn't have a doped envelope.   Depending on the doping agent, the envelope on a doped lamp may have a greenish cast to it if you look at it in daylight.

Yes, that's true, germicidal lamps are made the same as ordinary fluorescent tubes, but with quartz and no phosphor, so the UV passes through.

I was talking about newer UVC LED-based germicidal lamps. The ones I linked to looked very cheap. I also question their efficiency. The data sheets for UVC LEDs I've found have an efficiency of around 2%, which is miserable compared to the 30% offered by traditional mercury lamps.
http://www.farnell.com/datasheets/2774752.pdf
https://www.ushio.com/files/specifications/germicidal-low-pressure-mercury-arc.pdf

[jeremy]

FYI in that video he is using alcohol to clean surgical masks, not N95 respirators. The respirators are what is damaged by the alcohol.

[jfiresto]

Another sterilization method: using vaporized H₂O₂ to sterilize N95 masks

[Stray Electron]

I wonder how good the cheap LED germicidal lamps are? The blurb for the one linked below says it doesn't produce any ozone, so no good for masks, as it'll only sterilise the surface. It's inexpensive, considering the price of discrete UVC LEDs from reputable distributors. I accept distributors have lots of overheads, but I'm cynical of really cheap things.
https://www.amazon.com/dp/B085VWPT9C/ref=psdc_2314207011_t1_B085CGGQFR

Yes, that's true, germicidal lamps are made the same as ordinary fluorescent tubes, but with quartz and no phosphor, so the UV passes through.

   I understand but my point is that some 254nm lamps use doped envelopes so they won't pass 185nm and are incapable of creating ozone. If you want ozone then you must use lamps that have undoped quartz envelopes.

     The plain glass lamps with phosphor are 366nm lamps that are used for generating black light for cosmetic effects.  They're useless for killing bacteria or virus and also don't create ozone.  They have a SLIGHT effect on bacteria and virus but their killing power is literally about 0.0001% of what a 254uM lamps is.  I think they cover that in here https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3831650/

[Stray Electron]

  If you really want to play with a kickass UVC source then go find one of the old Mercury vapor street lamps.  We played with them when I was a kid and the ones that we used were 400 Watt lamps and had a small mercury vapor lamp with an undoped quartz envelope inside of a larger plain glass envelope that contained phosphors.  We broke open the outer envelope and removed it and just ran the inner lamps by itself.  They emitted tons of 254um radiation and any florescent material without about 20 feet would glow BRIGHTLY. They would also give us sun burns if we run one for more than about an hour and they also generated a lot of ozone.

   FYI  They have a negative resistance so they are supposed to be used with a ballast.

[NiHaoMike]

Found this study linked in the comments:
https://www.ara.com/sites/default/files/MitigateShortageofRespiratoryProtectionDevices_2.pdf

have you done that? looks like you experienced. i did it and yes the old filament stock i tried to dry sagged quite badly but not it become unusable (setting is something like 100degC iirc). the next roll i dried i reduced setting to 60degC and it works. filament brittleness gone and become usable again. for things like paper, tissue, mask etc... i dont think they will combust at 100degC, even if overshooted to say like 150degC.

Results vary dramatically depending on the oven, based on what I have read in 3D printing groups. Some suggestions to improve chances of success include letting the oven cycle off before putting in filament and some even suggested turning the oven off but leaving the light on.

I personally keep filament dry by putting it in a bag (mostly sealed but not completely) and leaving it behind my PC.

I was talking about newer UVC LED-based germicidal lamps. The ones I linked to looked very cheap. I also question their efficiency. The data sheets for UVC LEDs I've found have an efficiency of around 2%, which is miserable compared to the 30% offered by traditional mercury lamps.
http://www.farnell.com/datasheets/2774752.pdf
https://www.ushio.com/files/specifications/germicidal-low-pressure-mercury-arc.pdf

I think those LEDs are best used for small devices like self cleaning toothbrushes. They're also RoHS compliant although I recall lamps not for general illumination purposes are exempt since many such specialty lamps (UVC being one) have no feasible RoHS compliant alternative.

[Zero999]

I think those LEDs are best used for small devices like self cleaning toothbrushes. They're also RoHS compliant although I recall lamps not for general illumination purposes are exempt since many such specialty lamps (UVC being one) have no feasible RoHS compliant alternative.

I don't see why UVC LEDs would be better for small devices. 3W E17 lamps are widely available and it should be possible to make small mercury lamps the size of tiny neon indicator lamps, just by changing the gas fill and using quartz for the bulb. The amount of mercury used is tiny. I remember people being concerned about the environmental damage from mercury from compact fluorescents lamps, but it was later found out that the amount of mercury emitted from the extra coal burned to power incandescent lamps, exceeded that by many times.

I can see some advantages of UVC LEDs, over mercury lamps, such as not requiring a warm up time, more mechanically robust, longer life and an unlimited number of on/off cycles, but efficiency doesn't seem to be one of them.

[X.G.]

I  had  used  an oven to be a face mask restorer . 


http://www.crystalradio.cn/forum.php?mod=viewthread&tid=1838490&extra=

[Siwastaja]

I did look at some papers on this, and one advantage identified for the LEDs is that they seem to sanitize better per mW of output, compared to the 254nm mercury vapor bulbs. The reason is likely that the most effective sanitization wavelength range appears to be somewhere around 270-280nm, and 254nm is outside the optimum area, so the LEDs emitting at 275 or 285nm are a better match. Some numbers I remember seeing suggest a difference of maybe 2-3 times, so the final efficiency is still worse than the mercury vapor lamps, but not very much. Is 30% a realistic efficiency number for classical germicidal lamps? I remember seeing lower numbers. OTOH, with no heat-up time, any application requiring intermittent operation (like water or air filtration based on flow rate) benefits from the LEDs.

Looking at the rate of UVC LED development, efficiency and price, maybe they break through in the next 5-10 years. It's likely this crisis puts extra funding on the UVC LED development.

I found quite different requirements for the energy of killing viruses on surfaces with UVC. One study suggested 9J/m^2 or 14 J/m^2 for killing 90%, or 98% of the viruses, respectively. The latter being 1.4 mJ/cm^2, those pesky 10-20mW UVC LEDs might be appropriate for disinfecting more than just toothbrushes. They are still quite expensive, though; one 10mW LED clearly isn't enough to disinfect large area of surfaces or large volumes of air. But I guess there is a lot of difference based on how much material there is for the light to penetrate; for masks, it's useless, as discussed, because the UVC light would need to penetrate many millimeters of the material. A lot of energy is needed for this, likely damaging the material. Heat works better.

I'm personally looking at using UVC LEDs in disinfecting air after proper HEPA filtration. In this use case, there are no large particles that "shade" the UVC light and protect the viruses from the light.

[KaneTW]

Germicidal efficiency of UVC leds at optimum wavelengths is maybe 30% more effective than 254nm. Efficiency is still an order of magnitude worse.

The requirements vary and also are not super easy to calculate, but you want a 254nm dose of ~130mJ/cm2 to reach the target for a 99.99% inactivation of even resistant virii. Standards require 40mJ/cm2.
Specifically, this is not the power emitted from the bulb, but actually reaching the microorganism. This isn't a huge issue when disinfecting air, but gets more complicated when disinfecting water.

[profdc9]

If there are any references for UV mercury germicidal lamp efficiency, please send them may way.  The study I was referring to is

Hao & Han, Qiuyi & Zhang, Shanduan. (2013). “254 nm Radiant Efficiency of High Output Low Pressure Mercury Discharge Lamps with Neon-Argon Buffer Gas” Applied Mechanics and Materials. 325-326. 409-412. 10.4028/www.scientific.net/AMM.325-326.409.

That is where I obtained by 30% estimate from.

[LaserSteve]

This is in my State, using H2O2 aka Peroxide.  They did the research four years ago. Their machine handles whole shipping containers of masks.

https://www.battelle.org/newsroom/news-details/battelle-cleared-to-sterilize-n95-masks-at-max-capacity-operate-in-other-states-to-fight-coronavirus-ppe-shortage.

Process is Vapor Phase Hydrogen Peroxide.

Steve

[Caliaxy]

I'm not sure I've seen it mentioned here, but 70% isopropyl alcohol (rubbing alcohol) is more effective at disinfection than 80%+.  Although I vaguely remember this being the case for bacteria, unsure about viruses.

A very plausible explanation I heard for that is that alcohol evaporates faster at higher concentrations. Even though 95% alcohol is more effective per se than 70% (in destroying bacteria and viruses), it stays on surfaces for a shorter time, thus its overall efficiency is reduced. 70% seems to be the sweet spot. Of course, this only applies to disinfecting surfaces, it doesn't applies to disinfecting objects immersed in alcohol.

[pipe2null]

I have not been able to find COVID-19 specific engineering data, but some medical professionals have used data from SARS-CoV studies for sterilization guidelines.
https://www.eevblog.com/forum/projects/covid-19-related-engineering-data-ftlog-data-only-please!/msg2992562/#msg2992562

For SARS (similar to coronavirus):
Heating less than 75 degree C is NOT sufficient to kill all the viruses.
"...it was completely inactivated at 75 ◦C in 45 min (Fig. 2C). Surprisingly, at both 56 and 65 ◦C the virus was inactivated at early time points but at 60 min a small amount of virus was detected."

UVC (254nm) is effective at killing the virus much faster than heat, 6 - 15 min.
UVA (365nm) has no effect.
NOTE: If you are reading a study that has to do with UV, if they do not specify the wavelength used (nm), then you should greatly discount their UV related results...

As far as reusing masks go, profdc9 posted a collection of info:
https://www.eevblog.com/forum/projects/covid-19-related-engineering-data-ftlog-data-only-please!/msg2992912/#msg2992912

...
I found quite different requirements for the energy of killing viruses on surfaces with UVC. One study suggested 9J/m^2 or 14 J/m^2 for killing 90%, or 98% of the viruses, respectively. The latter being 1.4 mJ/cm^2, those pesky 10-20mW UVC LEDs might be appropriate for disinfecting more than just toothbrushes. They are still quite expensive, though; one 10mW LED clearly isn't enough to disinfect large area of surfaces or large volumes of air. But I guess there is a lot of difference based on how much material there is for the light to penetrate; for masks, it's useless, as discussed, because the UVC light would need to penetrate many millimeters of the material. A lot of energy is needed for this, likely damaging the material. Heat works better.
...

If you found professional or academic studies on this, would you mind please posting them to the above-linked thread?

[Siwastaja]

If you found professional or academic studies on this, would you mind please posting them to the above-linked thread?

I found a few by simple googling. Just made my own napkin memo & calculations, didn't save the references, sorry. You should be able to repeat the googling and find even more than I did, though.