Intercomparison of Instruments Used for Safety and Performance Measurements of Ultraviolet Germicidal Irradiation Lamps

Abstract

Public health concerns such as multidrug-resistant tuberculosis, bioterrorism, pandemic influenza, and sick building syndrome have brought about increased interest in the use of ultraviolet germicidal irradiation (UVGI) to prevent the spread of airborne infection. UVGI lamps require that radiometric measurements be performed to ensure their safe and effective use. This study evaluates 10 detectors that measure the ultraviolet radiation hazard of low-pressure mercury UVGI lamps, including a polychromator spectroradiometer, narrowband detectors designed to measure the ultraviolet radiation in a short range of wavelengths, and broadband detectors with a varying spectral response designed to follow the UV hazard action spectrum. The angular responses, spectral responses, and linearity of the detectors were measured and compared. The agreement between the measured angular responses and the ideal cosine responses varied widely among the detectors, and in general, the detectors with diffusing optics agreed significantly better with the ideal cosine response. The spectral responses at 254 nm also varied widely among the detectors, and, in general, the narrowband detectors agreed more closely with the 254 nm irradiances measured under the same conditions by a double monochromator spectroradiometer. All detectors displayed good linearity. The angular and spectral response data were then used to develop correction factors for the effective irradiance measurements of two UVGI sources, each measured at 10, 20, and 30 cm. The measured effective irradiances were compared with those measured by a double monochromator spectroradiometer with an integrating sphere input optic. Prior to correction, the effective irradiances measured by the detectors varied widely, ranging from 0.29 to 2.7 times those measured by the spectroradiometer. The application of cosine and spectral response correction factors significantly improved the agreement for the effective irradiances measured by all of the detectors, typically to within 10–20%. Awareness of these detector characteristics can play a key role in ensuring the accuracy of health hazard measurements of UVGI lamps.