September 21, 2016
After having just written an article on the Air Oasis machine, I thought that now might be a good time to share some of the work I’ve done related to UV light and air purification. As noted in the Air Oasis article, there is certainly some degree of mostly unknown toxic byproducts produced by the PhotoCatalytic Oxidation (PCO) technology. While I believe the Air Oasis holds real promise in helping to knock the super small biotoxins that make those with Chronic Inflammatory Response Syndrome (CIRS) sick, I have reservations when it comes to continuous use, or use by those with MCS.
As it turns out, about 9 months ago I spent a good chunk of time researching the benefits of UV light. This research started out with the development of a small, bright UV probe that I built. The probe could be inserted deep into a nose just like a MARCoNS swab – past the turbinates to the nasophyranx. Now before you go thinking I’m completely crazy shining a bright light deep in my nose, take a few minutes to review the RhinoLight. The RhinoLight is a medical device used to improve nasal immunology in the treatment of everything from allergies, to sinusitis, to nasal polyps. Eventually, I hope to write about all my experimentation related to alternative MARCoNS treatment.
Getting back to the topic at hand, as my UV light research expanded outward, I realized that there was real potential in UV air purification devices like those made by companies such as Sanitaire and Sanuvox. This became really clear when I uncovered well-controlled government backed studies that looked at the killing ability of UV-C light systems in relation to bacterium, mold, and viruses. It was clear that with enough intensity, UV-C light could clean up the air.
What excited me even more was the fact that this could be accomplished with relatively safe, ozone free, UV-C light. In contrast to the Air Oasis technology, my Light Tube would only use natural, high intensity UV-C light – no breaking apart of water molecules required. Except for having to exercise a modicum of care not to exposes ones personage to the light, there seemed to be much less risk than other air purification technologies.
The “icing on the cake” was when I realized that I could actually build my own unit. My do-it-yourself (DIY) “Light Tube” would have much greater capacity than commercially available units for about one-third the cost. In fact, as I learned about how to calculate the killing ability of my homemade UV Light Tube, it became clear that some of the units on the market didn’t have a “snowball’s chance in hell” of neutralizing even the weakest of toxins. For the remainder of this article, I’m going to talk a bit about light, look at controlled studies, explain how to calculate the killing power of a UV-C devices, determine the killing power of my Light Tube and other units, and finish with details on how you can build your own powerful UV Light Tube.
UV Light Basics
Light consists of waves of electromagnetic energy. There are numerous uses of electromagnetic energy that depends upon the length of the wave – the distance between the peaks and valleys. The uses range from gamma rays used in medicine to radio waves. When it comes to visible light, humans can see a narrow band of light energy in range of about 380-750 nano-meters. A nano meter is one one-billionth of a meter.
For the purposes of this discussion, we’re interested in what’s called ultraviolet (UV) light. Most UV light is just outside the range visible by humans. The forms of UV light that people are most familiar with are UV-A, UV-B, and UV-C. UV light is produced naturally by the sun. When it comes to creating air or water purification systems using UV light, all the studies I’ve seen use the shorter UV-C light. More specifically, ultraviolet germicidal irradiation (UVGI) lamps are almost always UV-C lamps that produce 254nm long wavelengths.
UV-A (315nm – 400nm): Black lights and tanning lamps – ages skin and harmful to eyes
UV-B (280nm – 315nm): Helps produce vitamin D – causes sunburn and skin cancer
UV-C (100nm – 280nm): Air and water purification – damages DNA in cells (blocked by the ozone layer)
What got me interested in UV light was studies I read about how UV-C light could kill mold, bacteria, and viruses. In addition, if UVV light at 185nm is added in, as is the case with some Sanuvox units, ozone is produced. Ozone can neutralize volatile organic compounds (VOCs). The Light Tube I made only uses UV-C light, but I’m just saying that it’s not uncommon for UV-C light and perhaps a little UVV (also called VUV) light to be used in various commercial UV light purification units.
Before we leave this introduction into UV light, it’s helpful to understand UV light units of measure. The power or intensity of UV lamps is typically expressed in uW/cm^2 (MicroWatts per Square Centimeter). When it comes to energy, or killing capacity, units of uW-sec/cm^2 are used (MicroWatt-Seconds or MicroJoules per Square Centimeter). In other words, the total energy or killing capacity is a product of the power of the lamp multiplied by the length of exposure. You can increase the killing energy by either increasing the intensity, or increasing the exposure time, or both. As we’ll find out, when it comes to killing pathogens in an airstream rushing past UV-C light, the power of the lamp(s) has to be quite high as the time it takes to pass the lamp(s) is typically very short (fractions of one second).
UV Light Studies
Studies that initially peaked my interest came from Sanuvox. For example, a double blind study done by McGill University and published in the The Lancet Medical Journal showed that UV-C light used in the ventilation systems of offices “reduces over all sickness by 20%, reduces respiratory symptoms by 40% and has resulted in a 99% reduction of microbial and endotoxin concentrations on irradiated surfaces within the ventilation system”. In this study, UV-C light was shone upon the air-conditioning coils inside the ductwork.
As I mentioned in the DuctWork – Mold – Health article, A/C coils are very often moldy. Certainly, shining continuous, DNA damaging, UV-C light on the A/C coils is going to thwart mold and bacterial growth. In so doing, folks in these building are going to feel better. However, while the results are significant, this study really doesn’t say much about the ability to use UV-C light to “nuke” pathogens zipping by a lamp suspended in an air stream. Incidentally, make very sure if you decide to install a UV light above the A/C coil in your home that the coil and ductwork is mold free. Otherwise, the light is going to kill the mold resulting in toxic fragments being distributed through out your house.
When it comes to knocking out pathogens in an airstream with UV light, we need to look to Defining The Effectiveness of UV Lamps Installed in Circulating Air Ductwork study. This was a well-controlled study funded by the U.S. Department of Energy (DOE), the Air-Conditioning & Refrigeration Institute (ARI), and others. In this study, specific pathogens were introduced into the airstream of a full sized test rig that controlled for temperature, humidity, and air flow. The kill rate of the pathogens that passed by various UV light configurations was then measured.
The study measured the kill rate on seven pathogens, Serratia marcescens, Staphylococcus epidermidis, Pseudomonas fluorescens, Bacillus subtilis, Aspergillus versicolor, Penicillium chrysogenum, and Cladosporium sphaerospermum. The first four are bacteria and the last three are fungi. In the study, it was found that 1,341 uW-sec/cm^2 of energy killed 99%of the bacteria. The maximum energy applied of 7,509 uW-sec/cm^2 defeated 16% of Cladosporium, 10-17% of Penicillium, and 66-88% of Aspergillus. The mold spores were much harder to kill.
Later on, the same testing procedure was used by EPA’s Homeland Security Research Program (HSRP) to determine the efficacy of specific commercially available units. For example, in Biological Inactivation Efficiency by HVAC In-Duct Ultraviolet Light Systems – Sanuvox, the Sanuvox UV Bio-Wall 50 was tested. The results from testing numerous manufacture’s systems can be found in Biological Inactivation Efficiency of HVAC In-Duct Ultraviolet Light Devices.
In the HSRP testing, they used a smaller and different set of organisms, Bacillus Atrophaeus (mold), Serratia Marcescens (bacteria), and MS2 Bacteriophage (virus). Unlike the mold spores used in the DOE testing, all of these pathogens are fairly easy to kill. In particular, if you look at Industry Accepted Kill Charts, it can be seen that 90% of B. Atrophaeus spores (formerly B. Subtilis) can be killed with 11,600 uW-sec/cm^2 of energy. Similarly, 90% of S. Marcescens bacteria can be neutralized with 2,420 uW-sec/cm^2, and 90% of the Bacteriophage MS2 can be killed with 2,600 uW-sec/cm^2. In comparison, the kill charts state that it takes between 44,000 to 132-000 uW-sec/cm^2 to defeat Aspergillus mold spores depending on the particular strain.
|UV-C Kill Chart||UV energy in µW-sec/cm2 needed for kill factor|
|Bacillus anthracis – Anthrax||4,520||8,700|
|Bacillus anthracis spores – Anthrax spores||24,320||46,200|
|Bacillus magaterium sp. (spores)||2,730||5,200|
|Bacillus magaterium sp. (veg.)||1,300||2,500|
|Bacillus subtilis spores||11,600||22,000|
|Leptospiracanicola – Jaundice||3,150||6,000|
|Salmonela paratyphi – Enteric fever||3,200||6,100|
|Salmonella typhosa – Typhoid fever||2,150||4,100|
|Shigella dyseteriae – Dysentery||2,200||4,200|
|Shigella flexneri – Dysentery||1,700||3,400|
|Vibrio comma – Cholera||3,375||6,500|
|Mucor racemosus A||17,000||35,200|
|Mucor racemosus B||17,000||35,200|
|Bacteriopfage – E. Coli||2,600||6,600|
|Poliovirus – Poliomyelitis||3,150||6,600|
|Common yeast cake||6,000||13,200|
Given the weaker pathogens used, commercial units that produced around 20,000 uW-sec/cm^2 of energy like the Sanuvox UV Bio-Wall 50 and the Steril-Aire model SE1VO killed over 90% of the B. Atrophaeus mold and 99% of the S. Marcescens bacteria and Bacteriophage MS2 viruses. In other words, it fairly easy to deliver enough killing energy with UV-C light to knock out bacteria and viruses. When it comes to the hardier mold spores that are found in water damaged buildings (WDB) like Aspergillus and Penicillium, much higher amounts of energy are required to achieve 90% and above kill rates.
Of additional interest in the Circulating Air Ductwork study were discussions on humidity, VOCs, and the importance of cleaning. The study noted that when the humidity was above 70% or below 40% that the kill rate drops off for some organisms. Furthermore, no detectable amount of ozone was produced by the UV-C lamps. Given that it is ozone that neutralizes VOCs, none of the VOCs measured in the study were reduced. And finally, lamps and reflectors should be cleaned on a regular basis to ensure maximum kill rates. These facts help to define the limits of any UV-C light purification system.
So that’s all well and good. With enough energy, UV-C light can knock out most bacteria and viruses along with a goodly number of mold spores. However, when it comes to CIRS, what we really care about are the 99.8% of biotoxins smaller than the 0.3 micron particle size that HEPA filters are able to capture. Not surprisingly, this information is nearly impossible to come by. I would think that cell wall components like endotoxin, beta glucans, and mannans should be easily knocked out.
However, when it comes to mycotoxins, in Reduction of Feed-Contaminating Mycotoxins by Ultraviolet Irradiation it was found that a power of 0.1 milliWatts/cm^2, or 100 uW/cm^2, applied for 60 minutes was required to neutralize 100% of the mycotoxins. This equates to (100×3600)= 360,000 uW-sec/cm^2 of energy to neutralize all of the mycotoxins. Looking at kill rate tables, we can guestimate that a kill rate of 90% would require roughly half that amount or 180, 000 uW-sec/cm^2. When it comes to on-the-fly applications, this is a massive dose!
Having said this, unlike spores that have some ability to repair themselves from UV damage, I doubt that this is the case for mycotoxins. If this is correct, then with enough passes through a less powerful machine, it may be that the mycotoxins will eventually be destroyed. This hypothesis finds some support in Effect of Fluorescent and UV Light on Mycotoxin Production Under Different Relative Humidities in Wheat Grains.
UV Light Energy Calculations
Now that we understand that UV-C light can knock out pathogens along with having a sense for the amount of energy required, we’re at a place where we can look at the killing capacity of various home UV-C units. For this discussion, I’ll look at the Sanitaire RS140, the Sanuvox R4000GX, and my Light Tube. In order to do this, we need to cover the basics on how the amount of UV energy a system can impart is calculated.
Basically, the total energy imparted is determined by multiplying the power of the UV lamp(s) measured in uW-sec/cm^2 by the number of seconds of exposure. Energy equals Power multiplied by Time. Happily, manufacturers typically list the power output of their lamps in uW-sec/cm^2. In addition, two additional adjustment factors for the distance from the lamp and the type of reflective surface around the lamp are also multiplied into the equation.
UV Energy = UV Power * Time * Intensity Factor * Reflectance Factor
The intensity factor is looked up in a chart like this one. The power specification listed by manufacturers is the lamp output at one meter. In contrast, it’s best to have all the air moving past a UV-C lamp as close as possible to the lamp where the intensity factor is at a maximum. Ideally, this is within 2” of the lamp where the intensity factor is 32.3. Note: Other sources like Aquarium and Pond list intensity factors double the values shown here. I’ve used the more conservative values in all of the calculations that follow.
The reflectance factor (irradiance) is determined based upon the type of material surrounding the lamp. In publications like Fresh-Aire UV UVGI System Design Guide, it can be seen that aluminum has the highest reflectance factor – very little UV power gets absorbed by aluminum. As a result, the UV light bounces back-and-forth within aluminum housings thereby amplifying the overall power of the lamp. It’s generally accepted that the reflectance factor for galvanized steel is 1.57, for aluminum foil it’s 2, and for mirror polished aluminum it’s 3. To maximize the killing power, I used mirror-polished aluminum for my Light Tube.
Sanitaire RS140 (NutriPure 3B-SC)
OK, were now at a place where we can calculate the energy a given UV light device is able to impart on the pathogens in the air rushing past the lamp(s). Let’s begin with the Sanitaire RS140. Looking at the RS140 specifications, we see two 25 watts UV-C lamps that are 21.875” long housed inside a 7.75”x7.75”x34.5” box. The unit uses a 105 cubic-feet-per-minute (CFM) fan. We don’t know what the box is made of.
Technically, in order to determine the average intensity factor, a person would have to develop a complex formula that sums the intensity of UV light at each of the individual points within the rectangular housing. This is because the intensity is decreasing the further from the lamp and the light from the two lamps is overlapping. However, my reasonable estimate for this application is to double the average intensity factor of a single lamp placed inside a box with a cross-section of 3.875”x7.75” – a box half the width of the original housing. The average intensity for a single lamp in this narrower box may be estimated using one-third of the intensity factor values at 2”, 3” and 4”. In other words, I calculate the intensity factor for the box to be roughly (2/3)*(32.3+22.8+18.6)= 67.5.
We don’t know what type of lamps they’re using. We don’t know if they are powerful GHO-Lamps (also called Low Pressure High Output lamps/Amalgam lamps) or standard quartz GPH-Lamps. There are lots of different types of UV-C lamps. For the sake of being really generous, let’s assume they used a powerful UV lamp in a mirror-polished box. Using LightTech LightSources specs, the power of the strongest 25 watt lamp is roughly 80 uW/cm^2.
Before we can calculate the energy output, the last piece of information we need is the amount of time it takes air to travel from one end of the lamps to the other. We know that the fan moves 105 cubic-feet-per-minute (CFM) and that there are (12x12x12)= 1,728 cubic inches in a cubic foot. This equates to a fan capacity of (105*1728)/60= 3,024 cubic-inches-per-second. Furthermore, the cross sectional area of the housing is (7.75*7.75) = 60.1 square inches. Therefore, the speed of the air moving through the box is (3,024/60.1)= 50.3 inches-per-second. We know the lamps are 21.875” long so the length of time the air is exposed to the lamps is (21.875/50.3)= 0.43 seconds.
We’re now ready to calculate the killing power of the Sanitaire RS140. The absolute maximum possible Energy output of the Sanitaire RS140 UV Energy = 80 * 0.43 * 67.5 * 3 ≅ 7,000 uW-sec/cm^2. From the studies above, this is enough to knock out 99% of the bacteria and viruses along with the weaker mold spores. You can buy a used Sanitaire RS140 for $800 on ebay. The latest version with similar specs is the Sanitaire NutriPure 3B-SC. Note: In all likelihood, the actual energy output is probably one-third to one-half this amount.
The stated power of the Sanuvox P900GX portable UV air purifier is 8,540 uW/cm^2. This value incorporates the intensity and reflectance factors. Although the unit uses a variable speed fan of unknown capacity, even at a super long exposure time of one second, the unit only imparts (1*8,540)= 8,540 uW-sec/cm^2 of energy. At faster fan speeds, the unit would impart even less energy to pathogens zipping by in the air. On par with the Sanitaire RS140, the P900GX sells for around $800 new and $150 on ebay. Note: In all likelihood, the actual energy output is probably one-third to one-half this amount.
My Light Tube
I knew I could build a more powerful unit for less. I made the 5” diameter Light Tube out of mirror-polished aluminum sheet that I purchased on ebay for about $30. I took the 2’x4’ sheet to a local HVAC shop and they formed it up into 5” diameter tubes that I used pop-rivets to hold together. To prevent against direct UV exposure, I fitted each end of the aluminum tube with a standard, adjustable, sheet metal elbow available from any big-box building supply store.
At the end of one elbow, I installed a 5” sheet metal end cap that I cut a hole in for the fan and drilled four small holes to mount it. I then fitted a Silverstone Tek 92mm FQ91 44.2 CFM Fan for $15. The fan was driven by a FAVOLCANO AC to DC 12V 1A Transformer for $8. I also purchased a 4pin T5 Socket With Wire, a 4 pin PWM Extension Cable, and Silicone Fan Mounts for a total of $22.
Inside the duct, I fitted two copper mounts I soldered up to hold the GPHVA843T6L high-power 127 watt amalgam UV lamp rated at 400 uW/cm^2 and costing $16 plus $80 shipping. I powered the lamp with a PA8-1800-150 ballast that cost $55 plus $50 shipping. I could only find overseas sources for the powerful lamp and ballast. In total, all the parts cost about $270. If it wasn’t for the pricey overseas shipping, the cost would have been about half. Note: An alternative to soldering up lamp brackets is to use Milford Pipe Hangers.
To calculate the intensity factor, I divided up the cross sectional area of the 5” diameter tube into four concentric rings with area and intensity shown. Knowing that the total area of all four rings is 18.85 in^2, the average intensity is:
Intensity Factor = [(7.07*29.2)+(5.49*41.5)+(3.93*59.8)+(2.36*83.5)]/18.85 = 46.2.
Additionally, I confirmed that I achieved the desired 40 CFM air flow rate, using a Handheld Windspeed Meter. At 40 CFM, the air is rushing past the lamp at 4.89 feet-per-second (FPS). The lamps are 33.2” or (33.2/12)= 2.77 feet long. This works out to a UV exposure time of (2.77/4.89)= 0.566 seconds. Looking at the equation below, the energy output of my Light Tube is more than three times the Sanitaire and Sanuvox units.
Light Tube UV Energy = 400 * 0.566 * 46.2 * 3 ≅ 31,350 uW-sec/cm^2
Real World Testing
I was pretty excited after I got all the kinks worked out. The first version I built was way undersized and I ended up running around in circles until I finally figure out how to wire the ballast – each pair of prongs closest together on the lamp goes to one of the ballast terminals. I was convinced that at roughly 31,000 micro-Joules that the Light Tube would knock out a lot of pathogens. Unfortunately, the crude testing I did was inconclusive.
The first round of tests I did involved successively running the Light Tube with and without the lamp on while inside a moldy barn. Each time, we’d hold a petri dish filled with Malt Extract agar at the outgoing end of the tube. Within a few days, mold started to grow in both types of dishes. There wasn’t any appreciable difference between the dishes.
I then revised the test. My thinking was that the amount of turbulence at the end of the tube was pulling in mold spores hanging around the end of the tube that hadn’t been treated by the UV light. This revised method involved sterilizing a new, plastic garbage can that had two large holes drill in it. The hole in the lid received the Light Tube. The hole near the bottom allowed me to place petri dishes within.
Once again, we ran the Light Tube with and without the lamp on. We also placed a small, portable fan within the garbage can to ensure good mixing. After 15 minutes of run time, the Light Tube was turned off, a petri dish was place into the bottom of the can, and the can was sealed off. After 15 minutes of exposure, the petri dish was removed. Once again, there was no appreciable difference between the dishes.
So I’m not sure what the testing says. Perhaps I’m not appreciating how easy it is to cross-contaminate the petri dishes with inadvertent spores that waft into the garbage can as I’m placing the dish. Maybe I’m not giving enough time for the remaining viable spores in the garbage can to float down and land on the dish. Maybe the fact that I didn’t seal the seams of the Light Tube is resulting in spores being drawn into the tube at the downstream end away from the UV light. I don’t know. It’s frustrating.
What I do feel confident in is that the data says UV light can clean up the air and that my Light Tube has real killing power. Also, unlike the Air Oasis that I wrote about, the Light Tube does not produce ozone along with unwanted by-products. I’m not saying there aren’t any unintended chemicals being produced, just that the levels should be significantly less. If people show enough interest, perhaps I’ll try another round of testing.
For what it’s worth, I gave my Light Tube to my Mom. Our house is clean and I’m doing well. On the other hand, she’s terribly ill with CIRS. You can “knock her over with a feather”. So much so that she sent back the expensive Austin HEPA and carbon filter we sent to her because “it kicks up too much dust”. She likes the Light Tube and runs it often.
Besides the fact that it takes a lot of energy to knock out some mold spores and mycotoxins, you should be aware that UV lamps contain a small amount of mercury. Also, using a Stetzerizer meter, I found that UV lamps produce a lot of “dirty electricity”. I installed a Stetzerizer Filter in the same outlet as the Light Tube and it cleaned up the readings nicely.
Furthermore, none of the units discussed is meant to cover large areas. I don’t expect the Light Tube to clean up much more than a bedroom. Using small CFM fans also means you’ll need to create a certain amount of mixing using another fan in the room. We know this based upon the work of Gregg Weatherman who’s shown that without forced mixing with fans, only a small fraction of the air ever gets pulled into blowers – not to mention the tiny fan on UV devices.
Finally, the amalgam UV lamp selected does not produce ozone but I did notice a slight odor during the first few days of use. Anyone who builds there own unit with MCS should plan on running the device in a garage for several days to a week. I estimated the cost of running the 127 watt lamp 24 hours a day at about 37 cents daily.
After it has cooled down, make sure to clean inside the tube and lamp regularly with a clean cloth dampened with water or alcohol. The oil on people’s hand is enough to cause the lamp to fail prematurely so if you inadvertently touch the lamp, make sure to clean it with alcohol. Also, the bulb life for continuous use is about one year. The lamp will still continue to work but its output drops over time.
Want Your Own Light Tube?
So that’s pretty much all I want to get into when it comes to using UV-C to clean up air. I did research the in-duct UV-C systems too. Just like the room models, it’s entirely possible to build a much more powerful unit for a fraction of the cost. In our home, I installed a powerful 5-foot long UV-C lamp inside the round ductwork on our Heat Recovery Ventilator (HRV) that we use to bring fresh air into the house. I wired the lamp to come on when the HRV turns on. Although it’s hard on the lamps to turn them on and off, the fact that the HRV has long on and off times means the net impact on the lamp will be to lengthen its 2-year life.
For those that are interested, I talked to an old carpenter/cabinet maker friend of mine. He’s detail oriented like me. We both grew up building stuff since we were kids. He said that if there was enough interest, he’d be willing to build Light Tubes for people. That’s good because I’m definitely not interested. I haven’t checked into pricing when buying in larger quantities but I’m guessing the total price to be around $300-$350 for an assembled unit including shipping.
Alternatively, my friend could put together a kit of all the parts needed to build your own Light Tube for around $200. The kit would include everything from the tube and ballast to the lamp holders and wiring harness. I could create some detailed construction videos. Everything would be included except for the galvanized ductwork that can be purchased at any big-box building supply store. Really, anyone that can use a drill and a screwdriver can build a tube. Note: The kit would use heavy aluminum tape to line the inside of the galvanized tube instead of polished aluminum. From the picture, you can see these surfaces are similar.
I’m not going to knock myself any further until I get a sense for whether there’s any interest. If there is, I think it’s a win-win for my friend and purchasers. I don’t plan on making any money on the deal unless we start talking about serious cash. This is basically me trying to help out where I can.
I can say that the light tube I’m envisioning for purchasers would be a 6” diameter tube with two of the high-power amalgam UV lamps – not just one. I roughly estimate the energy output of this upgraded configuration to be around 90,000 uW-sec/cm^2. Now that’s some serious killing power. As James Kirk would say in Star Trek, “Scotty, I need more power!” Well Captain Kirk, I just added in second warp drive by using two lamps. How’s that for more power!
Please be aware that you’d be buying a working prototype. At best, a cover will be fitted over the bulk of the electronics. What you’re buying will look a lot like the pictures. The Light Tube will not be UL listed or anything like that so if you’ve got kids or just can’t keep yourself from sticking your hand into hot shiny gadgets, then you should definitely steer away.
We didn’t get enough interested people to merit offering light tubes or kits. This is a bit surprising to me because I know MCS folks can’t tolerate much and something like my light tube is really the only option besides a good HEPA and carbon filter. Nonetheless, for those that are really interested, all the specs are in the article for making a very powerful two-lamp light tube.