Comparison of UV-A Transmission of Lamp Coating Materials

Background

Current GMP practices in food and pharmaceutical processing plants demand that production areas are glass free, or that any glass if broken, cannot find its way into the products. This poses problems in the use of EFK's (electronic fly killers) and sterilising plant which use ultra violet emitting lamps. This is because the UV radiation is notoriously harmful to polymers that would be able to protect the lamps. Any polymers used would not only have to resistant to the UV but also be transparent to it over the whole life of the lamp.

Various polymers have been tried over the years, and many found to be totally useless as they mostly age with the dose of UV radiation that they receive discolouring, and worse becoming brittle. The most successful have been found to be the fluoropolymers containing high amounts of the -CF₂- and -CF₃ in the molecule chain, and very best appear to be the completely fluorinated polymers.

Manufacturers data on the UV transmission is variable or non-existent, and it has been found not possible to compare one material with another from their data. Tests commissioned by Adtech using a UV spectrophotometer to measure the transmission of flat polymer films showed up a unforeseen problem. It was found that at UV frequencies the light was scattered by the polymer films, which were clear to visible light. The parallel beam of light used in the spectrophotometer was scattered passing through the film and gave a value significantly less than the total transmittance. The answer to this was to test the films or sleeves insitu, and measure the loss in output of the UV lamp with the various coatings.

Testing

The following test method was developed. A standard cylindrical UV lamp was mounted in a holder, and a calibrated UV radiometer mounted with its sensing head 25mm from the surface of the lamp. As the output of UV lamps varies significantly over time it was important to use one lamp for all tests, and not turn off the lamp between testing. Therefore coatings were removed (cut) from lamps and reapplied to the standard lamp by making a join on the lamp opposite to the sensor head. The various coatings could be quickly applied and removed this way, yet the lamp can remain on and the sensor and lamp do not move from their relative positions

Readings were taken as 1 minute total dose of UV, made alternatively on the various samples. This was then repeated 6 times including the lamp uncovered as control..

Technical Details: Lamp: Sylvania F15 T8 BL 15 watt 25mm diameter.

Sensor : EIT UVIRAD UV integrating radiometer Model UR365CH1

Spectral Response 320-390 nm (UV-A)

Calibrated 27 Jan 1999

Results of Tests

Sample2 & 7 Adtech FEP 0.25mm

Sample3 Adtech FEP 0.5mm

Sample4 & 8 Competitive coating 1 0.22mm

Sample5 Adtech development product XP 05 0.3mm

Sample6 Adtech development product XM 04 0.2mm

Results are in millijoules/cm²/minute

Test1 Test2 Test3 Test4 Test5 Test6 Total %Loss of UV-A

Batch 1

No coating 258 257 249 255 255 249 1523 0

Sample2 245 243 243 243 246 237 1378 4.33

Sample3 221 243 226 229 232 227 1378 9.52

Sample4 207 211 205 213 207 207 1250 17.92

Sample5 225 224 220 218 220 218 1325 13.00

Sample6 249 244 244 242 240 239 1458 4.26

Batch 2

No coating 252 253 255 252 253 253 1518 0

Sample7 245 244 245 247 245 242 1468 3.30

Sample8 211 210 214 218 211 214 1278 15.81

Conclusions

The test method is very consistent, easy to do and gives the actual pratical losses shown by lamp coatings as in use. There is a very big difference between the coatings tested. A loss of 15% or more is unacceptable in EFK's as the efficiency of the machine attracting and destroying flying insects is proportional to the UV-A output.

Note- The test apparatus used here is portable and can be demonstrated by request.

© Adtech Polymer Engineering Ltd. 1999

UVtest1.lwp 11/02/99