Can you kill coronavirus with UV light? | Hacker News Reader

Can you kill coronavirus with UV light?

bbc.com

101 pointstartoran98 comments

The simple answer is it probably can, but the actual facts are more nuanced than a simple confirmation.

All bio-safety labs or hospitals or water treatment plants or special HVAC systems have UV-C lamps to sanitize pathogens because UV-C radiation destroying nucleic acids (DNAs / RNAs).[1] However, the effectiveness really depends on the amount exposure/energy level, light of sight(because UV-C doesn't penetrate deep into objects), and even dust in the air or the on the UV-C light lamps(which reduced the energy they emit) as well as how DNA/RNA can be evolved or self-repaired.

Anecdotally, Chinese hospitals are using UV-C lamp to sanitize their medical equipments, masks (when it was in short supply months ago) among other things.

And yes, more exhaustive experiments should be conducted to affirm this practice as it's the most non-toxic and energy efficient way to sanitize surfaces.

[1]: https://en.wikipedia.org/wiki/Ultraviolet_germicidal_irradia...

[2]: https://www.cnbc.com/2020/03/18/how-china-is-using-robots-an...

It's not just Chinese hospitals that are using UVC to sterilize masks, I think some US hospitals have started doing the same thing due to the shortage there.

This is being pioneered in Omaha:

https://www.unmc.edu/news.cfm?match=25283

All water treatment plants do not use UV.

It is actually fairly expensive.

Some municipalities only use chlorine. My municipality only has a 40 day supply of chlorine to treat the water.

For such reasons, I purchased a whole-home UV-C water unit.

Now, the wastewater plant uses UV-C, but its output is cleaner than the drinking water input (river).

Can also use ozone. One of Toronto's plants uses it and treats about 20% of the city's water. Nice thing is that you don't need railcars to arrive with it.

https://www.toronto.ca/services-payments/water-environment/t...

And I believe a byproduct of UV irradiation is the generation of ozone.

Any idea on cost? I just looked for a whole home system and it is several thousand dollars.

> I purchased a whole-home UV-C water unit.

Got a link for the system you use?

I just searched Amazon, ‘whole home uv water’, and it was around $200 and does 12gpm.

In your experience, what is the service life of the UV light source for your system?

The manual said 1 year, and the ac adaptor/ballast has an LED status light to indicate power and bulb failure.

There's 2 things to also note:

1: Don't try this at home! If you are directly exposed to the light you too will get burned and you will likely do serious damage to your eyes. Seriously!

2: UV-C also degrades plastics. Many of these masks and equipment are made of plastic materials and can have reduced efficiency. Specifically the bands will break quicker and parts of the mask may fuse or thin. It also isn't guaranteed that you covered the entire surface of the object. Masks are opaque and have folds in them, thus meaning there is non-uniform exposure. That's why countries are using these methods as a last resort. Even if we can say that these methods certainty kill the virus (there's no reason to believe it wouldn't), there's the question of quality control. How strong is the mask after? And how efficient is it after?

Yes AND, despite the mongering in this article, UVC wavelengths below 222nm have been found to be safe to mammalian skin.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5552051/

https://journals.plos.org/plosone/article?id=10.1371/journal...

Interesting, thanks for sharing.

Not only skin, but also eyes. You can look at Far-UVC light and it won't hurt your eyes either. Whereas regular UVC light can burn your eyes and cause temporary or permanent blindness.

Wow, that's surprising, and exciting. If true, every room in a hospital should have these lights.

Let's start with copper doorhandles and bed-frames.

Far better tested and understood.

It still takes 4hrs for copper to kill coronavirus. Why not both?

From the linked paper:

> the effect of chronic irradiation with a high dose of 222-nm UVC to mammalian cells has not been determined

Please be careful in announcing new truths based on your interpretation of a couple papers. That’s not how science is done.

You're misinterpreting that sentence. They are talking about data missing from _prior_ work there.

The title of the article is literally "Chronic irradiation with 222-nm UVC light induces neither DNA damage nor epidermal lesions" and a few sentences below the discussion of prior work they write "chronic irradiation with 222-nm UVC light was revealed not to induce mutagenic or cytotoxic effects in the epidermis".

> of rats

The tests on the SARS virus used 254nm wavelengths, which deactivated the virus in fifteen minutes. Your sources both say that that length causes severe damage, and both only tested 222nm's ability to kill bacteria. A conclusive "yes" would need testing done on a virus with the 222nm length.

Different virus, but still

https://www.nature.com/articles/s41598-018-21058-w

"We show for the first time that far-UVC efficiently inactivates airborne aerosolized viruses, with a very low dose of 2 mJ/cm2 of 222-nm light inactivating >95% of aerosolized H1N1 influenza virus."

That's enough to show it should be in serious testing currently, though I worry the difficulty in patenting any development will hinder progress.

Has anyone found anything on the health implications of mounting 220nm lamps [1] indoors, beyond the issue of damaging human DNA? In other words, if you assume those lamps to be safe to living cells due to the light not penetrating beyond the first layer of dead skin cells, what other problems might there be?

Our skin surface contains a lot of living bacteria. Different strains seem compete with each other, and for example an overgrowth of staphylococcus aureus seems to be correlated with skin issues. In that sense living bacteria protect our skin. How would an far-UVC 'antibiotic' lamp influence bacterial repopulation later? Either on skin and in spaces. Would it impair immune system development?

Viruses seem far more fragile than bacteria, so perhaps a low-power far UVC-lamp will be enough to kill viruses but leave the bacterial populations on skin and objects intact

[1] https://www.nature.com/articles/s41598-018-21058-w

You get a nasty sunburn from resonably strong UV lamps. I've seen that happen with people being very careless in the lab when using a UV bed for gel electrophoresis.

These are UV lamps strong enough to visualize ethidium bromide, I'm not even sure if they're strong enough to kill bacteria quickly, though I'd guess they might be. UV lamps are really not something you should point at humans.

It also burns your retinas and produces ozone.

No optical damage in this wavelength range (although the ozone may still be an issue.)

Suggest reading the articles people have linked to in this thread, pretty interesting stuff. I had no idea that there was a portion of the UVC spectrum that doesn't have significant harmful effects on humans.

UV definitely can damage the eyes. I suspect what the paper meant is not the wavelength, but the power. Because UV also burns skin. I've seen it happen. Both power and wavelength matter.

Far-UVC deactivates pathogens without causing damage to skin or eye tissue.

Sigh

> Viruses seem far more fragile than bacteria

This is a complicated issue, partially because viruses are not "alive" in the way bacteria are, so it can be tricky to define what "killing" them means. In the case of coronavirus, plain soap is like kryptonite to them. Whereas triclosan--which kills many bacteria--does almost nothing to coronavirus.

I have ordered a 15 W Chinese E27 UVC (200-275 nm) lamp from Amazon, and a E27 to wall socket plug with remote control. I am going to place it in the quarantine zone downstairs to zap incoming mail and packages, just for any case.

In one SARS-CoV study [1] the authors have irradiated SARS-CoV with a 254 and 365 nm UVC-UVA light source that emitted 4016 μW/cm^2. They found that "exposure of virus to UVC light resulted in partial inactivation at 1 min with increasing efficiency up to 6 min (Fig. 1A), resulting in a 400-fold decrease in infectious virus."

If my math is correct, a 15 Watt UVC light bulb will provide 4.77 W/m^2 flux within the radius of 0.5 m - assuming it emits omnidirectionally. Assuming the delivered power versus inactivation efficiency relationship is linear, it would require roughly about an hour of zapping to achieve the same levels of inactivation with a 15 W lamp.

Yes, we don't know if SARS-CoV inactivation times are comparable to that of COVID-19, but it's better than nothing. I am not a biologist, and micrographs of both viruses look somewhat similar. I wonder if they are structurally similar, should we expect similar susceptibility to incoming UV photons?

[1] https://medtradex.com/assets/Uploads/Literature-UVD-Corona.p...

Would you mind linking to the light you bought? Or a comparable one l.

Here is the one. I don't know if it performs in spec. Never bought one before.

https://www.amazon.co.uk/gp/product/B085HZK755/ref=ppx_yo_dt...

Because this is (a) NOT Far-UVC, and (b) 360-degree light that shines in all directions, the only way to use it safely is with a remote power switch or a short timer so you (and pets) can get out of the room before the light turns on. Don't re-enter the room until the light is switched off remotely or with a timer. It is hazardous to your eyes and skin.

Could it be used in a spotlight lamp that directs all the light in one direction?

I mention below I place the light bulb (using a hand-held socket) in a glass aquarium. Glass completely blocks UV-C light. The other hazard is ozone, so I use this outdoors on my patio or garage.

> Glass completely blocks UV-C light.

Erm, then why do they come in glass tubes like florescent lights?

Not all glass is the same. Pure silica glass will pass UV, the soda-lime glass your windows are made of won't.

Yeah, I got a E27 socket wall plug with wireless control. I assume it should be possible to get a remotely controlled lamp with reflector to make a spotlight out of it.

I bought a cheap glass aquarium for my UVC light bulb. Glass block UVC light completely so it becomes safe to use that for smaller items. Also you will produce a lot of ozone, so do this outdoors with a lot of ventilation.

This is a 3-watt handheld - the power may be too low.

https://www.amazon.com/Foldable-Travel-Household-Wardrobe-To...

One thing to note is that in your calculations, you are using 50cm distance, but in the study you linked, they are using 3 cm distance. Since this is inverse square law, that makes sense it would take longer to deactivate.

For another point of reference, this study[1] says it takes 1.8 J/cm^2 to deactivate the flu virus. I'm not sure which is more hearty but from my calculations, it seems as though it's similar (about 2 J/cm^2). If that's the case then if you brought the intensity of your light up (bring the bulb much closer and possibly wrap it in material that reflects UV-C if that's even possible) then it would only take seconds to deactivate the virus.

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4699414/

As far as I understand, we are talking about flux - per area at specified distance, not luminosity. I did recalculate flux for 0.5 m.

No. The study I referenced used total energy per cm^2 required to deactivate the flu virus, which is J/cm^2. You get that by multiplying the flux by the exposure time.

So whichever exposure time gets you the 1.8 J/cm^2 is enough to kill the flu virus and presumably SARS-CoV-2.

Well, like I said - I recalculated for the flux of the 15 W lamp. An hour of zapping gets you to 1.72 J/cm^2 which is very close to your 1.8 J/cm^2 figure.

We are not disagreeing here, I think your 60 min calculation is correct. My point is that your calculations are at 50 cm and you get a much lower time for exposure if you bring your light bulb much closer and figure out a way to reflect the UV light so that more light from the bulb hits the intended target.

Make sure you wear longsleeve clothes and sunglasses, so you do not burn your skin or eyes!

This needs to be up higher. I'm frequently concerned with people that read articles like these and do not understand safety precautions. The light the gp bought here will damage skin and eyes.

Slightly OT: numerous sources say that if the virus is on a surface, it eventually dies (or deactivates anyway, if you don't consider viruses alive in the first place). It can last up to 4 days on glass, 3 on steel and plastic, and 24 hours on cardboard.

What I've not seen explained and have no guess for is why it deactivates. It doesn't have a metabolism. It doesn't consume resources or use energy. It just sits there, a little spiky spheroid surrounding some fragments of DNA or RNA.

So what changes about it as it sits that causes it to deactivate?

Oxygen and the atmosphere.

https://www.marketwatch.com/story/deadly-viruses-are-no-matc...

So why are surfaces different? The virus is held together by a combination of hydrogen bonds (like those in water) and hydrophilic, or “fat-like,” interactions. The surface of fibers or wood, for instance, can form a lot of hydrogen bonds with the virus.

In contrast, steel, porcelain or Teflon do not form much of a hydrogen bond with the virus. So the virus is not strongly bound to those surfaces and is quite stable.

I linked Thordarson's Twitter thread on the subject above. Didn't know he'd written an article too. Nice!

Just guessing - molecules of oxygen flying like crazy all around (somewhere read on average 5km/s), eventually at one point knocking something crucial off?

At those scales, these things start to matter.

Just taking a shot in the dark here, but it likely just disintegrates, probably from moisture and even friction from air. That would explain why the "survival" is inverse to the moisture-holding capability of the surface.

Entropy. A virus is simple by the standards of living things, but it's still a pretty complicated system. There are lots of random perturbations that can affect it ranging from quantum to cosmic rays to chemical and thermal perturbations. All it takes to deactivate a virus is to break one of its many critical chemical bonds so that it can no longer reproduce.

[UPDATE] Wow, five replies posted within two minutes of each other! As other parallel replies have pointed out, oxygen is very likely a significant factor here as it is plentiful and very reactive.

They dry out.

https://www.usatoday.com/in-depth/news/2020/03/25/coronaviru...

> In all settings, viruses need water to survive. "Viruses can withstand a small amount of dehydration," says Dr. Paul Meechan, a former director of safety at the Centers for Disease Control and Prevention and president of the American Biological Safety Association.

I believe this is the correct answer. My daughter with the masters in biology also said this was the answer.

Interesting side effect is that you can deactivate the virus by putting it in a warm dry atmosphere. The warmth encourages evaporation and drying.

Sadly I don't have any equipment for doing viral particle counts (was tempted to bid on some here: https://www.equipnet.com/auctions/ but alas, I already have way more EE test gear than I can use, no sense adding a full bio laboratory to the mix :-).

Even though I cannot prove its viability as a mitigation solution, since I have a heat gun I spray my mail with 610 degree F heat to ensure it is dry and warm (without exceeding 452 degrees of course!)

I'm curious to see how the curve might be affected with warmer weather coming to the northern hemisphere.

One can hope, certainly in the drier climates. As challenging as the pandemic is, the science part is still fascinating to me.

the corona virus (similar to the HI virus) has an envelope which is a lipid bilayer, there are viruses without an envelope. Now you may think that a virus with an envelope is more stable than viruses without, but the contrary is the case. once the bilayer tears, the virus is dysfunctional.

Also, what is not always talked about in the media currently is, that not all viruses are replicated functional. In fact, a large fraction of viruses produced are not capable of infecting a cell in the first place due to replication errors, mutations, etc.

>what is not always talked about in the media currently is, that not all viruses are replicated functional. In fact, a large fraction of viruses produced are not capable of infecting a cell in the first place due to replication errors, mutations, etc.

I agree that people aren't writing news articles about that, but why is it noteworthy that they're not? It doesn't matter if only 1% of the virus particles on a surface are infectious if there are millions of particles.

"SARS-CoV-2 RNA was identified on a variety of surfaces in cabins of both symptomatic and asymptomatic infected passengers up to 17 days after cabins were vacated on the Diamond Princess but before disinfection procedures had been conducted"

https://www.cdc.gov/mmwr/volumes/69/wr/mm6912e3.htm?s_cid=mm...

RNA !== active virus.

"Saying that live virus is there because viral RNA was found is like saying I must be holding a meatball sub because there's a marinara stain on my pants." - https://twitter.com/kenjilopezalt/status/1242285123652825089

They disassemble, the atoms find lower energy states.

Viruses are nanomachines held together with fairly weak bonds. Rough surfaces interact with and literally pull viruses apart. Very interesting thread here:

https://twitter.com/PalliThordarson/status/12365493051895971...

For anyone who's interested, CIE recently made their UVGI standards available for free ($150 normally):

http://files.cie.co.at/cie187-2010%20(free%20copy%20March%20... http://files.cie.co.at/cie155-2003%20(free%20copy%20March%20...

Generally you don't want to be around most UVC (253nm) because these wavelengths are very good at giving you cataracts, etc. So if you buy these things online, don't stay in the room when you turn them on.

There are some newer lamps (210-220nm) that don't seem to penetrate the outer layer of the epidermis (cornea or skin) and so might be OK for occupied spaces. But companies are still being very cautious about occupied uses, because this all still has yet to be proven in real life. Also, 220nm requires excimer lamps, so 253nm is way easier to get for now.

Just to add, when I did some research on excimer lamps last week, producers are few, and the research I discovered were sometimes using the 253nm lamps, but adding filters to get to the 210-220nm range. It quite confusing, to say the least.

Edit: Indeed the article linked above is one such study:"We used a bank of three excimer lamps containing a Kr-Cl gas mixture that predominantly emits at 222 nm25,26. The exit window of each lamp was covered with a custom bandpass filter designed to remove all but the dominant emission wavelength as previously described15. Each bandpass filter (Omega Optical, Brattleboro, VT) had a center wavelength of 222 nm and a full width at half maximum (FWHM) of 25 nm and enables >20% transmission at 222 nm."

How about just heating your mail in a dry oven? Thats what Stanford tested for PPE masks.

Or just waiting 24 hours or more works as well,

hey! relevant anecdotal data here. I work in an automotive repair garage. Anyone who has to drive a vehicle dropped off by a customer has to make sure its been UV sanitized first. We drape a 40 watt CFL UV lamp from your rearview mirror, put your floor mats on your windshield and rear window, and cover the windows with scratch mats for 35 minutes. After the timer goes off, a big blue sticker goes on your window. We then do it all over again before we give the truck or car back to you.

We had 2 ozone generators we were using (normally to remove cigarette smoke from car seats) but they take hours. the bulbs are cheaper and faster!!

I'm not running out to get any sort of UV lamp now but I am mildly interested in a thing I heard about that works with forced air ducting in a home AC/heating system. Not sure if it would help with allergies.

Does anyone know how LED-based UVC lamps like [1] with >300 LEDs can sell on Amazon for less than $100 when it appears the individual LEDs are fairly expensive? On DigiKey/Mouser the cheapest price I can find is still more than $1/LED [2].

[1] https://www.amazon.com/2020-Newest-Germicidal-Lamp-Light/dp/...

[2] https://www.mouser.com/Optoelectronics/LED-Lighting/LED-Emit...

Curious, I didn't even know they made UV-C leds yet. Normal UV leds are UV-A.

Otherwise digikey/mouser prices for LEDs are usually exorbitant. Try Chinese marketplace sites instead.

Yeah I would expect a rise in HVAC installations. I was surprised that cruise ships don’t have them installed along with hepa filters.

Interesting; is this a current practice? How do you get enough exposure time to be effective in moving air?

By integrating over distance :)

Big Clive over on YouTube was complaining the other day that certain sellers of these lights had taken part of his video warning against exposing skin to them and used it to claim they were skin-safe when promoting them for use against the coronavirus. It was some way into a multi-hour livestream, so I'm not going to try and dig it out though.

This is a pretty sham conclusion. "No" if you use where people are. No mention of the utility in cleaning areas that are temporarily closed to the public.

Light hits a lot more surface than just you are spraying and wiping.

In things like grocery stores being cleaned every night; or even with modern proximity sensors being able to turn on a 5 min burst of uv-c without anybody in the room. Uv-c (while yes requiring precautions) should work great.

I happened to have bought a uvc wand in November. It has a button and tilt sensor so it's hard to accidentally scan over your eyes. It's definitely reassuring to "wipe" the phone and door handles when I come back in the house at close proximity for a few seconds.

It's sad that even the BBC allows people who don't know what they are talking about to spread propaganda, not even knowing it. It kills everything the size of a virus because of its size, not some magical attributes the virus may or may not have. The other comments show three recent scientific publications in top-tier journals showing the narrow wavelength (222nm) that doesn't harm human skin or eyes.

>It kills everything the size of a virus because of its size, not some magical attributes the virus may or may not have.

It wouldn't cause a tiny iron sphere to disintegrate. It's not true that everything the size of a coronavirus can be "messed up" by UV light.

Time for led bulbs that also contain uvc light, perhaps with a battery so that you can clean a room after you turn of the light.

I’ve always wondered. Why can’t we put uv lamps in Hvac air handlers? Especially on airplanes and cruise ships.

I guess before there was no compelling reason to have the extra cost but now it might be time to rethink it.

We can. Half a lifetime ago, I replaced the HVAC in my house and one of the add-ons, which I bought, was a UV disinfecting lamp.

I have no idea if it was effective or snake oil but it was on offer. Hospitals and medical labs have (demonstrably effective) UV lights in their HVAC systems.

UVC lamps produce ozone which is not healthy to breathe.

I think some HVAC systems already do

Is it the seasonal changes in UV that make flu and similar viruses less transmissible during the summer months? Maybe it has nothing to do with temperature as some have suggested.

It's been argued that both temperature and humidity make viruses less transmissible. Not sure about the UV intensity.

Question for biologists, doctors etc:

AFAIK we don’t understand why “flu season” ends.

Is it not likely that increased UV in spring/summer may tip the balance? Has this explanation been proved/disproved?

Zinc works better than UV light, and has fewer side effects.

So article says: "No" and then goes on to describe how it kills virus. WTF

Crappy headlines are a major form of fake news.

(I wrote to a major paper once to complain about this, and the author of the article replied: Well, I don't write the headlines. Hooray.)

I wrote to the Denver Post once, and the editor replied "We don't write the articles." It was a local story, but written by the paper's owning entity, for publication in all of its owned papers, including the Post where the incident happened. I guess sort of like an in-house news wire "service."

They do obviously write some of the articles that appear in their paper.

Reply on news.ycombinator.com