UV LED used to purify water

[NewBie]

The SET devices used in the module, which were developed in collaboration with Asif Khan’s group at the University of South Carolina, emit at 280 nm, and the 4 cm3 aluminum treatment chamber is designed to maximize light reflection at this wavelength.

In tests using sterile tap water contaminated with E coli at a concentration of 10 000 microbes/ml, the UV LEDs destroyed at least 95.5% of the bacteria.

http://compoundsemiconductor.net/articles/news/9/3/18?alert=1

 

[KevinL]

Very cool. UV systems purifying water are not unknown (www.uvaquastar.com, Philips UV Sterilamp tube - yes, they even had a CPF special some time back /ubbthreads/images/graemlins/wink.gif) but this is the first time I've heard of UV LEDs doing the job.

The UV Aquastar module is fairly large and is housed in a 1L Lexan bottle, but has a much better kill ratio, however going forward maybe they might consider integrating UV LEDs into a smaller half-liter bottle which is smaller and easier to carry.

 

[Grubbster]

[ QUOTE ]
NewBie said:
In tests using sterile tap water contaminated with E coli at a concentration of 10 000 microbes/ml, the UV LEDs destroyed at least 95.5% of the bacteria.

[/ QUOTE ]
I'll stick with the filtration devices. Don't think I would want to drink 4.5% E. coli. /ubbthreads/images/graemlins/ooo.gif

 

[Cornkid]

I second that Grubbster.. Nice technology though..

-tom

 

[KevinL]

It's just the beginning, really. There is a specific dosage you need to achieve the 99.99% kill ratio that the UV Aquastar does, they are driving the tube at around 6W for a one liter, non-flow-through system.

Flow-through is challenging because the water is continually moving and thus spends less time getting irradiated. However, it's a good first step to seeing UV LEDs doing the job, eventually as lifespans increase and power outputs increase (or system design improves), they will get there.

 

[NewBie]

They are still working on it this is just one of the early prototypes.

Too bad regular faucet filters don't filter out all the bacteria either.

Too bad ceramic filters don't work that long and get plugged up.

DARPA is funding alot of this, as UV not only kills most bioweapon bacteria, but also breaks down many chemical agents.

Imagine a quick tent or array you set up that a contaminated soldier in his NBC suit can walk through, and/or drive contaminated vehicles through, for initial decon.

 

[chimo]

Filtering at municipal water production facilities is usually still required with to get the water to a sufficient clarity to allow UV radiation to fully penetrate the water flow. UV is especially usefull in rendering Cryptosporidium and giardia inactive. Typical chlorine treatment is not very effective with these two nasties.

It's good to see the UV LED progress in this form of water treatment. It should make for an energy efficient and, more importantly, a reliable form of water treatment.

Paul

 

[beezaur]

It is used most heavily in treated wastewater discharge streams, the stuff that comes out of sewage treatment plants.

I do most of my design work in drinking water treatment, and seldom do anything with UV treatment, but it has been around for quite a while. LEDs are new, however.

I know that some of the difficulties with the technology have been bulb life and getting enough power in the right wavelengths. Hopefully LEDs will help both of those.

It sounds like a few of you might have experience with this. I'll try to give a brief rundown for those from other fields.

My understanding of the biology is that the radiation does not predominantly kill the microbes, but mostly renders them unable to reproduce. They are thus unable to cause disease. This is why it is called "inactivation" and not killing.

Whenever you treat water, you get results based upon how thoroughly you do the process. You can filter at a certain rate and get a certain removal, or chlorinate at a certain concentration for a certain time and get a certain inactivation, or whatever.

Treatment results go on a convoluted "log" scale -- 1-log inactivation is where 90% are inactivated. 3-log removal is where 99.9% are removed.

You can design pretty much any treatment process to have any effectiveness, but it might not be practical. Getting 7-log inactivation of a certain organism with a certain chemical might make the water taste bad, or it might be too expensive.

Some nice things about UV light:
- adds no bad tatses (theoretically)
- does not create carcinigens as chlorine can

The bad thing with drinking water is that the light does not stay with the water as it flows through miles of pipe to your tap. Water can get re-contaminated by colonies that have grown in the pipes.

This can also be used for things like fruit juice or point-of-use disinfection.

Scott

 

[2dim]

[ QUOTE ]
Too bad regular faucet filters don't filter out all the bacteria either.

Too bad ceramic filters don't work that long and get plugged up.

[/ QUOTE ]

Portable reverse osmosis faucet and under-the-counter units, that include charcoal, are highly effective for personal use. Some also have ozone producing cartridges, but all require water pressure to operate. Bear in mind that not only bacteria/viruses must be eliminated...our municipal water supply here in Toronto originates from a lake that's become CHEMICAL SOUP!

 

[beezaur]

Reverse osmosis (RO) removes pretty much everything, even a lot of ions. In fact, the problem with RO is that water really likes to have stuff dissolved in it. Water from RO units is corrosive, and leaches stuff from your pipes. A lot of time syou have to redissolve things like calcite to make it taste right and "stabilize." But RO will clean up your chemical soup. Or Seawater.

Anyway, I am really intrigued by this LED thing. At low rates, you should be able to use battery or solar power. There are a lot of off-grid applications where that would be nice: hiking, remote locations, power failures. Pretty cool!

Scott

 

[KevinL]

[ QUOTE ]
beezaur said:
Anyway, I am really intrigued by this LED thing. At low rates, you should be able to use battery or solar power. There are a lot of off-grid applications where that would be nice: hiking, remote locations, power failures. Pretty cool!


[/ QUOTE ]

Scott, there is already a unit that does that, although it is based on the Philips Sterilamp UV tube, not a LED. They're talking 3-4 log inactivation because the UV tube is powered at 6W and you zaaaaap the thing for a minute by simply pressing a button. Treats a liter of water, and the bottle is included.

It looks every bit as cool in person as it does in the photo when it's lit in a darkened area. Powered by two lithium 123s from our favorite CPF supporting dealer (Batterystation.com).

 

[beezaur]

KevinL, that's pretty slick! Thanks for finding that.

Apparently that unit gets ~70 cycles (70 liters) per battery change. I don't know how much energy from that lamp is dumped into 240-280 nm (apparently low pressure mercury arc gives a line at 253.7 nm) compared to LED. I would presume that an LED would put a lot more of its total radiated energy into that range, possibly allowing for shorter times ane thus more treated volume per battery change.

One thing that hampers water disinfection with UV is turbidity in the water. Bugs can be shielded by floating masses, and cloudiness prevents the full dose of light from penetrating the water.

Now for some fine print issues with backcountry water.

I looked up efficacy against crypto cysts. UV is "effective" against Cryptosporidium parvum but. . .

"When exposure time was less than 150 min, the reduction in infection incidence was 36% and less. Infectivity was eliminated totally only upon exposure to UV light for 150 and 180 min." [Faust and Aly, Chemistry of Water Treatment, 2nd Ed., p. 530].

That's a lot of batteries to prevent your various "beaver fevers."

Given that crypto cysts are ~4-12 micrometers, and because of the turbidity issue, I would recommend prefiltration with a ceramic filter that can do at least 4 microns. For turbid (cloudy) water, you would probably want to precede that with coarser filtration, to extend your main filter's service life.

I don't mean to knock this device at all. I think it is probably more effective than tablets given real world conditions. I'm thinking I might want to pick one of those up. I was needing a long shelf life viral disinfection method for my disaster bag.

Scott

 

[KevinL]

You're welcome. /ubbthreads/images/graemlins/smile.gif Mentioned it in my second post, actually. It's a good unit, simple to use, and I put one in my BOB although you're right, prefiltering is a must. I talk more about that in the BOB thread.

It's good to hear an independent opinion from someone who actually works with treatment systems, because I'm a layman where it comes to water purification and I'm basically trusting the manufacturer to make sure it works because I don't have any means of independently verifying how well it kills the nasties.

Spec sheet: http://uvaquastar.com/info_pages.php/pages_id/31

UV-C Tube: Philips TUV4T5 Germicidal 4W UV-C (254 nm) Sterilamp®, hot-cathode, low-voltage. Rated useful life (to 85% efficiency) = 6,000 hours. Expected life as used in AquaStar ™ (driven at 6W, producing 0.8W UV-C at 20 ° C) = 2,000 hours.

254nm - close enough to 253.7. They say they get 0.8W of it. From what I know of LEDs vs tubes, LEDs tend to be able to get specific wavelengths and concentrate the energy in those bands. Colored Luxeons are binned by wavelength with just 5nm separation between the bins (for InGaN).

More interestingly:
>99.99% effective against protozoa (including Giardia, Entamoebic Dysentery, and Cryptosporidium)

I think it has something to do with the dosage, power levels, and so on. Maybe you want to ask them about it?

 

[beezaur]

I missed your post above -- that's what I get for reading too fast.

"99.99% effective against protozoa (including Giardia, Entamoebic Dysentery, and Cryptosporidium)"

I am not sure what that is about. It could be that they are referring to free-swining protozoans, which are pretty easy to inactivate, or it could be that they have some studies where cysts really were inactivated in 1 minute.

A strange thing about water treatment research is that data can be all over the place. You might have 5-log inactivation in one study, but 1-log inactivation in another. The chemistry of natural waters is extremely complicated. Some humus compounds' UV transparency is dependent upon pH, for exmple. It is very difficult to reproduce results in natural waters with unpredictable concentrations of who-knows-what.

From an engineering standpoint, you have to forget about maybes and just go for the all-inclusive, "bounding" cases. I have had things go wrong with jobs where manufacturer's claims were relied upon. They tend to pretty optimistic. It is one thing to get a product to work brand new in the lab, another to get it to work in use. You have to leave a lot of room for Mr. Murphy in water treatment.

Scott

 

[KevinL]

Understood. I deal with different systems at work, and we overspec everything to deal with the worst case scenarios, which has saved us many a time. Think I'll just nuke the water a few times if I'm really unsure /ubbthreads/images/graemlins/grin.gif though in a bad situation we may not have that luxury.

Sorry for the thread hijack, but what do you think of mixed oxidant systems? These are the ones that take salt, zap it to produce chlorine and mixed oxidants, then you pour the mixture into the water. Too much can make it taste bad, but how good are they at killing the nasties?

There is a little handheld unit that also runs on two CR123s.

 

[turbodog]

[ QUOTE ]
KevinL said:
Understood. I deal with different systems at work, and we overspec everything to deal with the worst case scenarios, which has saved us many a time. Think I'll just nuke the water a few times if I'm really unsure /ubbthreads/images/graemlins/grin.gif though in a bad situation we may not have that luxury.

Sorry for the thread hijack, but what do you think of mixed oxidant systems? These are the ones that take salt, zap it to produce chlorine and mixed oxidants, then you pour the mixture into the water. Too much can make it taste bad, but how good are they at killing the nasties?

There is a little handheld unit that also runs on two CR123s.

[/ QUOTE ]

Msr's unit is the only portable thing out right now, I think. It still does not get crypto though.

Note: the 4 hour wait time can be cut to about 30 minutes of prefiltered.

You don't have to get 100% kill anyway. And you're not going to find super-contaminated water unless you're drinking straight from an outhouse.

 

[enLIGHTenment]

[ QUOTE ]
beezaur said:
I'm thinking I might want to pick one of those up. I was needing a long shelf life viral disinfection method for my disaster bag.


[/ QUOTE ]

Wouldn't a non-electric distiller be a better choice for a disaster kit? The shelf life of the equipment is essentially indefinite and the only consumable is fuel for a heat source.

In a disaster, firewood may well be easier to find than batteries.

 

[beezaur]

"Wouldn't a non-electric distiller be a better choice for a disaster kit? . . ."

I keep a couple of poly bottles in my bag, so the UV unit would be a little extra weight but essentially zero extra space. Even with one set of batteries (70 liters of water) that gives me the time to set up another means. I have a glass distillation setup for my lab water, but it is pretty slow and requires cooling water. I think I filtration followed by bleach (for viruses) instead.

And I am a Red Cross volunteer. As soon as I get my home squared away in a disaster, I have to be portable.

KevinL,

To be honest I have no experience with on-site generated hypochlorite, i.e., zapped salt. My understanding is that it performs the same as prepared sodium hypochlorite (bleach), which is what my clients prefer since they are used to it (rural systems here). I think the zapped stuff would be more effective than bleach because it would be fresher, but probably not for any other technical reason. Hypochlorite is a good oxidant no matter how you get it. I can see people switching to that system once they become more comfortable with it.

Another benefit of chemical oxidants is that your equipment stays disinfected. You don't get "yuk-bottle" if you add a drop here and there /ubbthreads/images/graemlins/wink.gif

Scott

 

[KevinL]

The thread about the MSR MIOX water purifier is here.
Runs on two CR123s, bring your own salt and containers. The device zaps the salt and water, then you pour the concentrate into the water you're purifying.

Most of us are not too concerned about battery availability, I have a Surefire SC1 (6 cells) in the BOB, and lights I can scrounge 'em from if the need arises. There is also a 2-cell SF light in the BOB, loaded and ready to go.

The logical question would be "which system is better" but there are probably lots and lots of factors and reasons /ubbthreads/images/graemlins/thinking.gif /ubbthreads/images/graemlins/wink.gif

Good point about the chemicals, I recall reading in a paper about UV systems that the primary advantage of MIOX was residual effect - UV has none, so the water has to be consumed within 24-36 hours or re-treated.

 

[Sub_Umbra]

"Wouldn't a non-electric distiller be a better choice for a disaster kit? . . ."

A couple years ago I looked into building a small, emergency pasteurizer for the semi-automated production of drinking water. (It was just a lark as we have a 2.75 gal gravity powered Katadyn filter which we use all the time and does not reguire electricity or water pressure.)

You can save a huge amount of energy by Pasteurizing instead of boiling, which makes it very suitable for emergency planning. I wanted to make a stove burner sized coil out of copper. It would be somewhat automatic as it used a radiator thermostat that opened at about 150F IIRC. That was the hardest part to find. I found lots of cool info in the web.

beezaur,

Are there any inherent weaknesses to using a Pasteurization proccess in place of boiling to make emergency drinking water? I mean, besides the obvious one, of not getting the process right? IIRC it only takes a third of the energy needed to boil the water.