I am working with photolysis by MP UV of chloramines and other chlorine by-products in swimming pool water. Since the system is polychromatic I find it difficult to find a representative method to measure the intensity of the UV-light. I managed to get my paper published by linking the UV treatment to a full scale system there the electric energy used for the treatment could be measured but the general problem remains in my opinion unresolved.
So how can I measure the UV light intensity in a collimated beam with polychromatic UV-light in a way that is realistic for photochemistry?
Article Photolytic removal of DBPs by Medium pressure UV in swimming...
You can measure UV intensity with a radiometer. You can purchase them for certain (UV) wavelength ranges. Some cover the entire 200-400 nm range, others cover only the UVA, UVB or UVC fraction.
The radiometer should be on the same height as the surface of your collimated beam sample, and you should measure in a grid (say, each 0.5 cm left, right, up and down from the central point) to determine how homogenious the output of your collimated beam is. You'll have to correct for deviations. Other factors to account for are reflection and sensor sensitivity.
The downside of relating power consumption to degradation is that you don't take lamp aging, efficiency and sleeve fouling into account. A collimated beam is more straightforward; the irradiation that you measure is the irradiation that your sample will recieve (provided you allowed your lamp sufficient time to reach stable output).
Ferrioxalate actinometry is also commonly used for collimated beam tests. Rather than measuring the intensity (i.e. energy output in W/cm2), it measures photon quantities that are produced (einstein photons / min). This can be helpful as well, if the photon quantity is more critical for your process than photon energy, i.e. when most photon wavelengths send out by your lamp would be sufficient to break down your chloramines.
Hi Henrik,
You can an Ocean Optics radiometer; the instrument measures the complete spectral from which you can obtain the intensity of the spectral region you are interested in.
I would very much appreciate if you could email a soft copy of your publication (Photolytic removal of DBPs by medium pressure UV in swimming pool water.)
It depends on your budget and knowledge of the calculations involved for some approaches. If you use actinometry, you will need to also have a good quality spectrum of the lamp emission. The calculations for that approach are in several papers, but here's one I wrote a few years back that specifically used ferrioxalate:
Environ. Sci. Technol., 2003, 37 (9), pp 1933–1940
DOI: 10.1021/es025814p
If you aren't too savvy about those calculations etc., and if you have ~ $15k (I think) then I would recommend the Ocean Optics radiometer mentioned by Safarzadeh-Amiri Ali. Easy to operate and gives you the "spectral irradiance" (i.e., the intesity at each wavelength in the spectrum).
I strongly recommend to read the Recommendations from IUPAC we wrote about Actinometry. You can download the document free from the IUPAC page at: http://pac.iupac.org/publications/pac/pdf/2004/pdf/7612x2105.pdf
As pointed out above, actinometry should be done with monocromatic light, because you want to match absorbances at a particular wavelength between sample (your water) and reference (e.g., ferrioxalate). Otherwise you have to integrate over wavelengths with a good knowledge of the lamp spectrum. Please get acquainted with the problems by reading the recommendations and then you will have an idea of what you need.
For the definitions employed, such as spectral irradiance, please have a look at the Glossary of Terms used in Photochemistry (also freely downloadable at:
http://pac.iupac.org/publications/pac/pdf/2007/pdf/7903x0293.pdf)
See refs: Environ. Sci. Technol. 2007, 41, 7486-7490
Environ. Sci. Technol. 2008, 42, 3248–3253
Journal of Photochemistry and Photobiology A: Chemistry 193 (2008) 50–55
I'll add one more here that may help, by Jim Bolton and Karl Linden:
Bolton, J. and Linden, K. (2003). ”Standardization of Methods for Fluence (UV Dose) Determination in Bench-Scale UV Experiments.” J. Environ. Eng., 129(3), 209–215.
doi: 10.1061/(ASCE)0733-9372(2003)129:3(209)
I have used ferrioxalate actinometery with the spectral irradance from the lamp manufacturer. You can estimate the relative intensity of the lamp for each nm or 10 nm range and use the appropriate quantum yields for each wavelength. This will give you a fair estimate of the intensity in the reactor for different wavelengths.
True, I was using new lamps and was working with the lamp manufacturer, so I had pretty detailed data on the lamps. A spectroradiometer was not available, so I had to be creative. Also, it is hard tot get a true intensity for a more complicated reactor geometery using a spectroradiometer, in my experience.
Hello, we use a spectroradiometer (OceanOptics HR2000) previously calibrated equipped with a 600 micron fiber. We use the option of photon flux measurement by numerical integration in the range of desired wavelengths. The device allows you to use other measures such as W/cm2, etc.
I think you can first calibrate your spectralradiometer by 254nm collimated beam apparatus, then using it to calculate the MP lamp's power withing range for example 200 to 300nm. Here the Keitz formular should be used.
Ferrioxalate actonimetry, as mentioned by mr. Bolton in this thread, to calculate the average photon intensity. This solution is (of course) light-sensitive, so prevent light exposure (other than your MP UV lamp) as much as possible during the measurement.
When I was doing collimated beam experiments as an Msc student, we used ferrioxalate actinometry a few times as a reference, and used a spectroradiometer on a day-to-day basis. If I recall correctly, it was the same model / filter as described by mr. Diaz. Keep in mind, though, that the UV intensity varies, depending on the location where it is measured. To correct for this, we measured the UV intensity at about 10 locations on the x-axis, and 10 on the y-axis after the MP UV lamp was warmed up. The outermost measurements are the edge of the petridish, and the center where the axis cross, is also the center of the petridish, and the center under the collimated beam tube.
I don't recall specifically what assumptions we used to translate these measurements to an irradiation distribution across the entire petri-dish surface. Of course, this can also be determined experimentally by measuring the complete grid (which hopefully is only required once).
Have you tried doing actinometry for your lamp? I recommend the book
https://photochemistry.jp/acv/ENGLISH/Chemact.pdf
- Badges
- Science topic
- Similar topics
- Mathematics