SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Estimate of reflected irradiance at various locations within a sample room
  7. Discussion and Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

Ultraviolet germicidal irradiation can be used to prevent airborne transmission of infectious diseases. A common application is to irradiate upper-room areas, by passing air from the lower room into the irradiated zone. Well-designed systems do not expose people directly; however, some UV radiation may be reflected off ceiling tiles and wall paints into the lower room. Lower room exposure should be limited to the American Conference of Governmental Industrial Hygienists threshold limit value of 6 mJ·cm−2 of 254 nm radiation per day. To limit the lower room exposure, the reflectance of upper-room surfaces must not be high. The reflective properties of wall paints have been studied, but less is known about the UV reflectance of ceiling tiles. Using a double monochromator spectroradiometer and an integrating sphere reflectance attachment, the UV spectral reflectance of 37 ceiling tiles was measured from 200 to 400 nm. The reflectances varied from 0.020 to 0.822 in this range, and from 0.035 to 0.459 at 254 nm, the main wavelength emitted by upper room low-pressure mercury germicidal lamps. These data were then used to estimate an 8 h exposure based on several simplified workplace scenarios. The implications for workplace safety are then discussed.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Estimate of reflected irradiance at various locations within a sample room
  7. Discussion and Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

Safety limits for exposure to ultraviolet (UV) radiation (200–400 nm) have been established by the American Conference of Governmental Industrial Hygienists (ACGIH; 1) and adopted by the American National Standards Institute (2). The ACGIH threshold limit value (TLV) for UV exposure is a cumulative, effective (spectrally weighted) dose of 3 mJ·cm−2 in an 8 h period. Common sources of UV radiation include arcs, gas and vapor discharges, fluorescent and incandescent sources and solar radiation. People may be directly exposed to these sources or to their reflections, e.g. off room surfaces or a snow-covered ground in the case of solar radiation. Occupational safety professionals should ensure that worker exposures do not exceed the TLV during an 8 h work day.

A UV radiation source of modern interest is ultraviolet germicidal irradiation (UVGI) lamps. Interest in this technology is being revived due to concerns over multidrug resistant (MDR) and extensively drug-resistant (XDR) tuberculosis, pandemic influenza, bioterrorism and other infectious threats (3,4). UVGI is used for disinfection purposes and can be used to disinfect air, surfaces and water. Low-pressure mercury discharge lamps are commonly used in UVGI applications and emit radiation in the UV-C range (UV-C: 180–280 nm), primarily at 254 nm. This radiation damages the DNA of microbes and inactivates their ability to replicate (5). Engineered to prevent airborne infections such as tuberculosis, a common application of UVGI is to irradiate an entire cross-section of the upper room, by passing lower room air through the irradiated zone (6–8). Using well designed, baffled lamps and installing them properly restricts high levels of radiation to a plane not accessible to occupants (9); however, some of this radiation can be reflected into the lower room and must be limited to safe levels.

The TLV for 254 nm radiation is an unweighted dose of 6 mJ·cm−2. This relatively low limit reflects the fact 254 nm radiation can readily cause photokeratitis (10) to the eyes (“welder’s flash”). The minimal dose required to cause erythema (11) to the skin (“sunburn”) varies with each person, although 21 mJ·cm−2 has been accepted in some circles as a minimum erythema dose for persons with skin Type-II (e.g. fair-skinned Caucasians; 12). Photokeratitis and erythema from UV-C radiation can be quite painful, but the effects are temporary and usually resolve within 24–48 h. A radiation of 254 nm, like other types of UV radiation, has been identified as a possible contributor to the risk for non melanoma skin cancer (NMSC). However, the International Commission on Illumination (Commission International de l’Eclairage or CIE) does not consider UV-C exposures in the workplace to present the recognized risk for skin cancer. The CIE has developed a standardized action spectrum for NMSC, which has a maximum relative effectiveness of 1.0 at 299 nm; 254 nm radiation is far less effective, with a relative effectiveness value of 0.0119 (13). The greater penetration depth of longer wavelength solar UV radiation is another reason why chronic exposure to outdoor sunlight is the main source of skin cancer risk (14,15).

Computer models have been developed to predict the in-room distribution of radiation from UVGI systems (16,17). To estimate the radiation reflected into the lower room and restrict it to safe levels, the reflectance of materials in the upper room, e.g. ceiling tiles and wall paints, must be known. The reflectance of wall paints and other building materials have been addressed in previous studies; however, there is little if any data available with regarding to ceiling tile reflectance (18–20). The data that does exist on UV reflectance of materials often does not extend to or specifically apply to 254 nm, and reflectance data provided by manufacturers most often applies only to visible radiation. This study is an effort at closing some of the current knowledge gap and is intended to assist in furthering the science of UVGI.

The reflectance of a material, ρ, is defined as the ratio of the reflected radiant flux to the incident flux in the given conditions (for incident radiation of given spectral composition, polarization and geometrical distribution; 21). Reflectance may be specular, reflection in accordance with the laws of geometric optics, without diffusion (i.e. mirror like). Reflectance may also be diffuse, diffusion by reflection in which, on the macroscopic scale, there is no specular reflection. Most real-world materials will exhibit partly specular and diffuse reflection, or mixed reflection (i.e. have a reflectance comprised of both a diffuse and specular component).

In this study, measurements were made of the diffuse UV spectral reflectance of 37 ceiling tiles with and without the specular component included, and the specular component was inferred from the difference in these measurements. We then estimated the irradiances reflected into a sample room by the ceiling tiles and wall paints, and compared them to the ACGIH exposure limits (1). The full range of ultraviolet wavelengths from 200 to 400 nm was measured, because new designs of upper room UVGI fixtures may include sources based on light emitting diodes or other lamp technologies. This could lead to a wide range of possible emission wavelengths for UVGI lamps in future years.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Estimate of reflected irradiance at various locations within a sample room
  7. Discussion and Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

Instrumentation and experimental setup for reflectance measurement.  The spectral reflectance was measured using the following equipment from Optronic Laboratories (Orlando, FL): a model 750-C double monochromator spectroradiometer (5 nm entrance slit and 3 mm exit aperture), with a model 750-C spectroradiometer controller, a model 740–70 integrating sphere reflectance measurement attachment, and a model 750-HSD-310 photomultiplier tube (PMT) detector head operated at 650 V. The light source used was a model 750-20D/UV automated UV-visible dual source attachment containing a 40 W deuterium arc lamp and a 150 W quartz tungsten halogen (QTH) lamp, with a model 45D deuterium lamp precision current source operated at 500 mA and a model 83A programmable direct current source operated at 5.0 A.

This system measures the reflectance factor R of a material. The reflectance factor of a material is defined as the ratio of the radiant flux reflected in the directions delimited by a given cone with apex at a point on the surface under test to that reflected in the same directions by a perfect reflecting diffuser (Lambertian) identically irradiated (22). The reflectance factor measurements reported in this study approach the true reflectance ρ of the material for the specified geometry and are referred to simply as reflectance. Reflectance was measured in 5 nm intervals from 200 to 400 nm and at the single wavelength of 254 nm. The geometric conditions of the reflectance measurement were conical–hemispherical, and the measurements were performed using the comparison method theory, to compensate for possible changes in the integrating sphere’s efficiency. The incident beam irradiated the samples at 10° from normal with an irradiation cone of ±2.8° and a 6 mm beam diameter at the sample. The reflected radiation was measured over the entire hemisphere. The reflectance measurement attachment included a PTFE specular light trap plug, which was inserted into an integrating sphere aperture and used to include the specular component in the reflectance measurements. A specular light trap with a ±6° exclusion cone was inserted into the same aperture and used to exclude the specular component in other reflectance measurements. For the majority of measurements, the deuterium lamp was used from 200 to 305 nm, and the QTH lamp was used from 310 to 400 nm. The sample and sample aperture were covered with black foil to prevent light leakage.

The spectral reflectance including the specular component of each distinct model and item number ceiling tile was measured a total of 10 times. The specific area of ceiling tile being irradiated was changed before each new measurement to account for macroscopic, non homogeneous surface characteristics. The 10 measurements were then averaged at each wavelength, and the standard deviation (SD) of the reflectance was calculated. The spectral reflectance excluding the specular component of each model and item number ceiling tile was measured twice. These measurements were performed to give an estimate of the diffuse and specular components of the reflectance, and the SD was not calculated.

The ceiling tiles were provided by Armstrong (Lancaster, PA), CertainTeed (Valley Forge, PA) and USG Interiors (Chicago, IL). Table 1 lists the manufacturer, model, item number and ID letter (for identification in proceeding Tables and Figures) of the ceiling tiles used in this study. All tiles were white unless specifically noted in Table 1. More detailed tile characteristics may be found in the manufacturers’ literature.

Table 1.   Ceiling tile information*,†.
Mfr.Item nameItem no.MaterialsNotable surface featuresID
  1. *All tiles are white unless noted in parentheses next to the model, or in the model name itself; ID used to identify specific tiles in Figures, Tables, and text.

ARMSTRONGARMATUFF860Wet-formed high density mineral fiberLatex paintR
CERAMAGURAD605Ceramic and mineral fiber compositeScrubbable plastic finishKK
CIRRUS574Wet-formed mineral fiberLatex paintHH
CIRRUS (camel)589Wet-formed mineral fiberLatex paintS
CIRRUS (platinum)589Wet-formed mineral fiberLatex paintX
CLEAN ROOM MYLAR1721Wet-formed mineral fiberSoil-resistant polyester filmG
CLEAN ROOM VL870Wet-formed mineral fiberVinyl-faced membraneK
CORTEGA769Wet-formed mineral fiberLatex paintV
CROSSGATE2625Wet-formed mineral fiberLatex paintAA
DUNE1772Wet-formed mineral fiberLatex paintU
ENDURA639Wet-formed high density mineral fiberVinyl latex paintGG
FINE FISSURED SCHOOL ZONE1714Wet-formed mineral fiberLatex paintFF
FINE FISSURED (camel)1729Wet-formed mineral fiberLatex paintT
FINE FISSURED (haze)1729Wet-formed mineral fiberLatex paintDD
FINE FISSURED (tech black)1729Wet-formed mineral fiberLatex paintD
FINE FISSURED (white)1729Wet-formed mineral fiberLatex paintZ
FINE FISSURED OPEN PLAN1754Wet-formed mineral fiberLatex paintII
GEORGIAN1753Wet-formed mineral fiberLatex paintCC
GRAPHIS8005Wet-formed mineral fiberLatex paintJ
LATITUDES588Wet-formed mineral fiberLatex paintW
LEDGES8011Wet-formed mineral fiberLatex paintP
MESA680Wet-formed mineral fiberLatex paintE
OPTIMA3151Fiberglass with acoustically transparent membraneAcoustically transparent membrane and latex paintB
PAINTED NUBBY3101FiberglassLatex paintEE
PEBBLE2989FiberglassLatex paintN
RANDOM FISSURED2910FiberglassScrubbable vinyl film facingF
SANSERA573Embossed wet-formed mineral fiberLatex paintY
SHASTA2906FiberglassScrubbable vinyl film facingH
STRATUS531Wet-formed mineral fiberLatex paintJJ
VL871Wet-formed mineral fiberVinyl-faced membraneL
CERTAIN-TEEDECOPHON GEDINA EFiberglassSound-resistant coatingBB
FINE FISSURED HIGH NRC454Wet-felted mineral fiberLatex paintM
SYMPHONY F1340,2,4FiberglassLaminateC
THEATRE BLACK F1910,2FiberglassLaminateA
VINYLROCK1140,2-CRF-1GypsumScrubbable vinyl film facingI
USGASTRO CLIMAPLUSMineral fiberO
BRIO CLIMAPLUSMineral fiberQ

Uncertainty in reflectance measurements.  The SD in the reflectances including the specular component, averaged over all wavelengths for all ceiling tiles, was ±0.020. Nearly, all reflectances including the specular component had a SD less than ±0.040 at all wavelengths. Four samples had SDs that was noticeably higher, and these ranged from ±0.049 to ±0.099. The typical accuracy and precision reported by the manufacturer were ±1% and ±0.25%, respectively.

Geometry of in-room exposure to UV radiation reflections.  The following equations were used to approximate the lower room irradiances from upper-room UVGI reflections, and were derived from radiometric principles (23). The ceiling tiles and wall paints act as the source and the exposed room occupant is the receiver. Figure 1 diagrams the general exposure geometry used in the derivation.

image

Figure 1.  General exposure geometry for a source (upper-room surface) and receiver (person). Adapted with permission from McCluney (23).

Download figure to PowerPoint

From the definition of radiance, the flux received from a source is given by:

  • image(1)

where Φ is the flux of the source in W, L is the radiance of the source in W·cm−2·sr−1, dAr is an infinitesimal element of the receiver area, dAs is an infinitesimal element of the source area, θ is the angle made by the direction of emerging flux with respect to the normal to the surface of the source, Ψ is the angle made by the direction of incident flux with respect to the normal to the surface of the receiver, and R is the distance between points on dAr and dAs.

The irradiance incident upon the receiver, or flux per unit area, is found by dividing both sides of Eq. (1) by dAr, resulting in the simplified equation of:

  • image(2)

where Er is the irradiance incident upon a point on dAr in W·cm−2.

Treating the source as a Lambertian surface, it can be derived from radiometric principles that the irradiance incident upon the surface from the UVGI lamp, E0, is diffusely reflected into the surrounding hemisphere with the following radiance:

  • image(3)

where ρ is the reflectance of the source.

Substituting Eq. (3) into Eq. (2), and approximating the integral as a sum of N finite elements, Er is given by:

  • image(4)

If the entire source area has the same reflectance, ρ can be placed outside the summation. Similarly, if all finite areas are equal, ΔAs can be placed outside the summation.

The ACGIH exposure limits for UV radiation apply only to sources within ±40° of the normal to the receiver area (1), and thus Eq. (4) applies only to Ψi ≤ 40°. Because Eq. (4) assumes finite elements, there was a possibility that part of a finite source area, ΔAs,i, could be located within Ψi = 40°, but be excluded from the calculation of Er, an error that may become problematic with large ΔAs,i. Therefore, a “worst-case,” conservative Ψ was calculated, so that if any part of ΔAs,i was within Ψi = 40°, the entire finite source area and associated E0,i would be included in Eq. (4).

Exposure approximation example.  Our estimated exposure is an approximation and is shown for illustrative purposes only; this should not be considered as a substitute for actual radiometric measurements. Our example calculated the reflected radiation from upper-room surfaces, originally emitted by a wall-mounted UVGI fixture in a 3.66 × 4.57 m2 (12 × 15 ft2) small office with a 2.44 m (8 ft) ceiling. We assumed that the UVGI fixture was located on the center of one wall at a lamp height of 2.13 m (7 ft). Our example also assumed that the UVGI fixture typically emits the UV beam at an upward angle of ca 3–5°; this is often the case, although some fixtures exist that are designed to irradiate in a flat plane. It should be noted that this is not an ideal installation. In an ideal installation, the room ceiling would likely be higher, and the UVGI fixture would be mounted and shimmed so that the UV beam does not strike the ceiling. The beam would only reach the opposite wall, thereby increasing the germicidal capability and reducing the exposure to room occupants. Only reflections off the ceiling tiles and the wall opposite the UVGI fixture were considered for this calculation. The values used for the ceiling tile reflectances were based on those measured in this study and ranged from 0.05 to 0.50. For the reflectances of wall paint, we used 0.10 and 0.35, which were reported for water-based paints by the IES (19) and a variety of sources whose data were collected by Kowalski (21).

The exposures were calculated for several orientations: standing vs sitting, facing the UVGI fixture vs facing the opposite wall, and at several distances away from the fixture. The irradiance incident upon a horizontal plane (simulating the top of the head) and the irradiance incident upon a vertical plane parallel to the fixture wall (simulating the eyes and face) were calculated. The maximum standing and sitting heights were based on the 99th percentile anthropometric data for males and females wearing shoes with erect posture (24,25). Figure 2 diagrams the 2-D exposure geometry for a single finite source element, i, incident upon a horizontal receiving surface. The standing height (i.e. full stature) of 99th percentile males is 6 feet 4.5 inches (1.94 m) for males, and the standing height of 99th percentile females is 5 feet 11 inches (1.80 m). Figure 2 was adapted from Vasconez et al. (26). It is important to note that the human eye actually gazes slightly downward at 15° below the horizontal during normal tasks (27). This would direct the eyes away from the irradiance reflections, and decrease the risk still further. Hence, our example illustrates a higher risk than would be expected in a real situation.

image

Figure 2.  Model room exposure from a single source element with area, As,i, incident upon a horizontal receiver (top of head) at a source-receiver distance Ri. Total exposure found by summing contributions from all source elements. Adapted with permission from Vasconez et al. (26).

Download figure to PowerPoint

Our irradiances were based on measurements of the emissions from a model LIND24-EVO Hygeaire upper-room UVGI fixture (Atlantic Ultraviolet, Hauppauge, NY) installed at the Laser/Optical Radiation Program Laboratories. The UVGI source was a G24T5L lamp, operating at 25 lamp watts and 8.5 W UV output. Portions of the vertical and horizontal cross-sections of the irradiance distribution as measured in that laboratory are shown in Fig. 3a,b. The laboratory where the UVGI lamp was installed did not have the same room dimensions as our 12′ × 15′ example room, nor was the UVGI lamp installed at a height of 7 feet. The Figures illustrate where the walls and ceiling would be relative to the UVGI lamp for our example. The irradiance was measured with a model X1-1 Optometer with a model UV3718 UV-C detector (Gigahertz-Optik, Newburyport, MA). Spatial constraints and the detector’s angular response prevented us from performing irradiance measurements in the same location and geometry as the ceiling reflections (i.e. facing directly downward). Instead, measurements were taken pointing the detector toward the lamp, and the measured irradiance at each location i was multiplied by the tangent of 4° to calculate E0,i that would be incident upon the ceiling. Irradiances that would be incident upon the wall opposite the fixture were measured at the center of 0.418 m2 (1.5 ft2) rectangular grid-like sections at 4.57 m (15 ft) from the fixture. The total reflected irradiance incident upon the receiver was determined in each case by calculating Eq. (4) from each source (i.e. the ceiling tiles and wall paints), and summing them together.

image

Figure 3.  Vertical (a) and horizontal (b) irradiance distributions for the model room used to estimate the in-room exposure to UV reflections from ceiling tiles and wall paints. (b) Illustrates the irradiance distribution located 1 foot (30.5 cm) above the level of the UVGI lamp. The numbers in the contour lines are the irradiances in μW·cm−2.

Download figure to PowerPoint

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Estimate of reflected irradiance at various locations within a sample room
  7. Discussion and Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

Ceiling tile reflectance measurements

Figure 4a,b plot the spectral reflectance of the ceiling tiles including the specular component from 200 to 400 nm. For illustrative purposes, Fig. 4a plots the reflectance of tiles with a reflectance less than 0.100 at 254 nm, and Fig. 4b plots those with a reflectance >0.100 at 254 nm. Table 2 tabulates the 254 nm reflectance of the ceiling tiles including and excluding the specular component. As shown in the Table, the 254 nm reflectance varied from a low of 0.035, to a maximum of 0.459. It should also be noted that the measured specular reflectance only applies to the geometry employed by this system, i.e. specular reflection at 10°. The spectral reflectance excluding the specular component was not plotted, as these values were very similar to those including the specular component. Because the reflectance of the ceiling tiles did not exhibit an appreciable specular component, this justified the assumption of a Lambertian source in Eq. (3). The minimum and maximum specular reflectances were 0.022 (measured in the UV-C range) and 0.822 (measured in the UV-A range; UV-A: 315–400 nm). The reflectances measured across the full ultraviolet range (from 200 to 400 nm) are also available in a supplemental electronic file.

image

Figure 4.  Spectral reflectance including the specular component of ceiling tiles with a 254 nm reflectance (a) less than 10% and (b) greater than 10%. Plots are base-10 logarithmic. See Table 1 for identification of the tiles.

Download figure to PowerPoint

Table 2.   254 nm Reflectance values of ceiling tiles.
ID ρ, Incl specular ρ, Excl specularID ρ, Incl specular ρ, Excl specularID ρ, Incl specular ρ, Excl specular
A0.035 ± 0.0010.034N0.096 ± 0.0030.086Z0.260 ± 0.0480.288
B0.043 ± 0.0020.041O0.106 ± 0.0060.102AA0.260 ± 0.0350.243
C0.048 ± 0.0010.047P0.107 ± 0.0240.076BB0.273 ± 0.0070.274
D0.051 ± 0.0010.050Q0.122 ± 0.0070.131CC0.275 ± 0.0160.280
E0.052 ± 0.0020.047R0.151 ± 0.0080.150DD0.276 ± 0.0130.283
F0.055 ± 0.0060.054S0.152 ± 0.0090.164EE0.276 ± 0.0260.290
G0.060 ± 0.0020.059T0.165 ± 0.0210.169FF0.282 ± 0.0290.306
H0.064 ± 0.0040.062U0.237 ± 0.0110.227GG0.334 ± 0.0240.333
I0.070 ± 0.0010.065V0.240 ± 0.0330.270HH0.353 ± 0.0310.363
J0.071 ± 0.0050.071W0.240 ± 0.0630.220II0.362 ± 0.0280.364
K0.071 ± 0.0010.069X0.244 ± 0.0230.249JJ0.423 ± 0.1030.410
L0.074 ± 0.0010.071Y0.245 ± 0.0700.277KK0.459 ± 0.0150.438
M0.076 ± 0.0030.067      

Estimate of reflected irradiance at various locations within a sample room

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Estimate of reflected irradiance at various locations within a sample room
  7. Discussion and Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

Tables 3a,b, 4a,b and 5a,b list the reflected irradiances calculated for the various conditions and reflectance values. Bolded values would lead to a daily dose above the TLV, if one were to assume a very conservative constant 8 h exposure. The movements of room occupants play a very significant role in the exposure of room occupants to upper-room UVGI (28). Pertinent results from our study are summarized below:

Table 3.   Reflected 254 nm irradiance (μW·cm−2) incident upon a horizontal plane (standing/sitting). Potentially hazardous irradiances shown in bold.
ρ Ceiling tilesρ Wall paintDistance from back of UVGI fixture
0.46 m (1.5 ft)1.37 m (4.5 ft)2.29 m (7.5 ft)3.20 m (10.5 ft)4.11 m (13.5 ft)
(a) Male, 99th percentile
0.050.100.00/0.000.00/0.000.02/0.010.06/0.010.01/0.06
0.100.00/0.000.01/0.000.04/0.020.12/0.030.06/0.07
0.250.00/0.000.02/0.010.10/0.04 0.30/0.070.11/0.08
0.500.00/0.000.03/0.020.20/0.08 0.61/0.14 0.23/0.11
0.050.350.00/0.000.00/0.000.02/0.010.06/0.010.06/0.19
0.100.00/0.000.01/0.000.04/0.020.12/0.030.12/0.20
0.250.00/0.000.02/0.010.10/0.04 0.30/0.07 0.30/0.24
0.500.00/0.000.03/0.020.20/0.08 0.61/0.14 0.60/0.30
(b) Female, 99th percentile
0.050.100.00/0.000.00/0.000.02/0.010.05/0.010.05/0.07
0.100.00/0.000.01/0.000.04/0.020.10/0.030.10/0.08
0.250.00/0.000.02/0.010.09/0.04 0.25/0.07 0.24/0.11
0.500.00/0.000.03/0.020.18/0.08 0.49/0.13 0.48/0.17
0.050.350.00/0.000.00/0.000.02/0.010.05/0.01 0.31/0.20
0.100.00/0.000.01/0.000.04/0.020.10/0.03 0.36/0.21
0.250.00/0.000.02/0.010.09/0.04 0.25/0.07 0.50/0.25
0.500.00/0.000.03/0.020.18/0.08 0.49/0.13 0.74/0.30
Table 4.   Reflected 254 nm irradiance (μW·cm−2) incident upon a vertical plane facing wall opposite fixture (standing/sitting). Potentially hazardous irradiances shown in bold.
ρ Ceiling tilesρ Wall paintDistance from back of UVGI fixture
0.46 m (1.5 ft)1.37 m (4.5 ft)2.29 m (7.5 ft)3.20 m (10.5 ft)4.11 m (13.5 ft)
  1. *If 15° downward gaze is included, these irradiances are reduced to 0.30–0.40 μW·cm−2, and the other irradiances in bold are all reduced below 0.20 μW·cm−2. If 15° downward gaze is included, all of the irradiances in bold are reduced below 0.20 μW cm−2.

(a) Male, 99th percentile
0.050.100.03/0.030.05/0.040.07/0.040.12/0.01 0.23/0.00
0.100.04/0.040.05/0.040.08/0.040.12/0.01 0.24/0.00
0.250.04/0.040.05/0.040.09/0.050.13/0.01 0.27/0.00
0.500.06/0.060.05/0.040.12/0.060.15/0.01 0.32/0.00
0.050.350.12/0.100.18/0.14 0.23/0.14 0.40*/0.03 0.78*/0.00
0.100.12/0.100.18/0.14 0.23/0.14 0.40*/0.03 0.79*/0.00
0.250.12/0.100.18/0.14 0.23/0.14 0.41*/0.03 0.82*/0.01
0.500.12/0.100.18/0.14 0.24/0.14 0.43*/0.03 0.87*/0.01
(b) Female, 99th percentile
0.050.100.03/0.030.05/0.040.06/0.040.11/0.000.08/0.00
0.100.03/0.030.05/0.040.06/0.040.11/0.000.09/0.00
0.250.03/0.030.05/0.040.07/0.040.12/0.000.12/0.01
0.500.03/0.030.05/0.040.07/0.040.13/0.000.17/0.01
0.050.350.12/0.100.18/0.14 0.22/0.13 0.36/0.01 0.26/0.00
0.100.12/0.100.18/0.14 0.22/0.13 0.36/0.01 0.27/0.00
0.250.12/0.100.18/0.14 0.22/0.13 0.37/0.01 0.30/0.01
0.500.12/0.100.18/0.14 0.23/0.13 0.39/0.01 0.35/0.01
Table 5.   Reflected 254 nm irradiance (μW·cm−2) incident upon a vertical plane facing wall with fixture (standing/sitting). As shown, none of the calculated irradiances were potentially hazardous.
ρ Ceiling tilesρ Wall paintDistance from back of UVGI fixture
0.46 m (1.5 ft)1.37 m (4.5 ft)2.29 m (7.5 ft)3.20 m (10.5 ft)4.11 m (13.5 ft)
(a) Male, 99th percentile
0.050.100.00/0.000.00/0.000.00/0.000.00/0.000.00/0.00
0.100.00/0.010.01/0.010.01/0.000.01/0.000.00/0.00
0.250.01/0.010.01/0.020.02/0.010.02/0.000.00/0.00
0.500.01/0.030.03/0.030.05/0.020.04/0.000.00/0.00
0.050.350.00/0.000.00/0.000.00/0.000.00/0.000.00/0.00
0.100.00/0.010.01/0.010.01/0.040.01/0.000.00/0.00
0.250.01/0.010.01/0.020.02/0.010.02/0.000.00/0.00
0.500.01/0.030.03/0.030.05/0.020.04/0.000.00/0.00
(b) Female, 99th percentile
0.050.100.00/0.000.00/0.000.00/0.000.00/0.000.00/0.00
0.100.00/0.010.01/0.010.01/0.000.01/0.000.00/0.00
0.250.01/0.010.02/0.010.02/0.010.02/0.000.00/0.00
0.500.02/0.030.03/0.030.05/0.020.04/0.000.00/0.00
0.050.350.00/0.000.00/0.000.00/0.000.00/0.000.00/0.00
0.100.00/0.010.01/0.010.01/0.040.01/0.000.00/0.00
0.250.01/0.010.01/0.020.02/0.010.02/0.000.00/0.00
0.500.01/0.030.03/0.030.05/0.020.04/0.000.00/0.00
  • 1
    Sitting persons: As shown in Table 3a,b, the worst-case irradiances for sitting conditions were those incident on a horizontal plane, and these were calculated for wall paint reflectance = 0.35, and ceiling tile reflectance = 0.10–0.50. These irradiances ranged from 0.23 to 0.30 μW·cm−2. At worst, this irradiance could only lead to an ocular dose in excess of the TLV if the exposed person looked up at the ceiling enough times for the exposure duration to exceed 5 h in a given work day (1), and would not pose an erythema risk for skin Type-II persons even after 12 h of continued exposure (5). This would not appear to pose a realistic risk for overexposures in workplace settings, although it seems possible that a hospital patient reclining in a bed for long periods could look at the ceiling long enough to receive an 8 h dose in excess of the TLV. This would pose some risk to the eyes, and possible skin risk for people with a sensitive skin type, or who may be taking photosensitizing antibiotics. This is qualitatively consistent with the results of a personal dosimeter study by First et al. (27), which included four hospital subjects who were either in bed or in a wheelchair. This study included some caveats that the personal dosimeters’ exposure geometry cannot not perfectly replicate the exposure to the eye, and this could lead to overestimates in the measured doses. The maximum eye-level irradiances for those four subjects ranged from 0.27–0.79 μW·cm−2, and the highest recorded daily dose for those four subjects was 1.25 mJ·cm−2. This illustrated some reason for concern, although the daily doses in that study were still less than the 6 mJ·cm−2 TLV (1).
  • 2
    Standing persons, irradiances incident upon horizontal plane: For standing persons, some irradiances were potentially hazardous, but would not pose a hazard under realistic conditions. As shown in Table 3a,b, the potentially hazardous irradiances were determined for a variety of ceiling tile and wall paint reflectances, for persons facing the opposite wall from the UVGI fixture. The irradiances ranged from 0.23 to 0.74 μW·cm−2. At worst, 0.74 μW·cm−2 would lead to an ocular dose in excess of the TLV after more than 2 h of exposure (1). This would not be likely unless the person looked repeatedly at the ceiling many times during the work day. This irradiance could only pose an erythema risk for skin Type-II persons after almost 8 h of exposure (5), and workers would be unlikely to stand in one location long enough during normal work activities for this to occur.
  • 3
    Standing persons, irradiances incident upon vertical plane: As shown in Tables 4a,b and 5a,b, the worst-case calculated irradiances from reflections ranged from 0.22 to 0.87 μW·cm−2, for persons facing the opposite wall from the UVGI fixture. Whereas the tendency of room occupants to move during normal work activities would render low irradiances unlikely to pose a realistic UV hazard (e.g. persons standing near the wall and receiving 0.40 μW·cm−2 would require over 4 h of ocular exposure in a work day to exceed the TLV [1]), the highest irradiances would be a cause for some concern. For example, persons standing near the wall opposite the UVGI fixture would receive 0.87 μW·cm−2 reflected irradiance to the eyes and would require just under 2 h of ocular exposure to exceed the TLV (1). Whereas it seems unlikely that workers would stand for a sufficient amount of time at this location to accumulate a reflected dose that exceeds the TLV, this is the type of issue that would need to be addressed for safety reasons. In this example, the reflections could only pose an erythema risk for skin Type-II persons who stood near the wall for more than 6.5 h in a work day (5).
  • 4
    Effect of downward gaze on reducing ocular hazard: We noted in the Materials and Methods that, during normal tasks, the human eye tends to be directed at 15° below the horizontal (27). This would further reduce ocular exposure to UV radiation from in-room reflections. As shown in Table 4a,b, the worst-case calculated irradiances would then be decreased to 0.30–0.40 μW·cm−2, for 99th percentile males, standing in close proximity to the wall, with ρ = 0.35.
  • 5
    Skin cancer risk for this example: These data are consistent with the CIE’s position that exposure to outdoor sunlight is a much greater source of risk for NMSC than exposure to UVGI in the workplace. The worst case calculated irradiance in the above examples was 0.87 μW·cm−2. If we multiply this irradiance by the CIE NMSC action spectrum’s value for 254 nm, which is 0.0119, we obtain an NMSC-effective irradiance of 10.4 nW·cm−2. This effective irradiance is more than 50 times less than the NMSC-weighted solar spectral irradiances that are typical for sunlight near the horizon, let alone for when the sun is higher in the sky (29).

Discussion and Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Estimate of reflected irradiance at various locations within a sample room
  7. Discussion and Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

As mentioned in the Introduction, a properly designed and installed upper-room UVGI system should not directly expose the room occupants to UV emissions. Occupant exposure is then dependent on the amount of radiation reflected off upper-room surfaces like ceiling tiles and wall paints into the lower room. In this study, the measured 254 nm reflectances from ceiling tiles ranged from 0.035 to 0.459. These data may be used in computer aided design programs to help calculate the irradiance in a wide range of room dimensions and locations (16,17). It should also be noted that UV reflectance is not a main quality of interest for ceiling tile manufacturers. Materials and manufacturing processes may change in the future to improve qualities, such as fire resistance, appearance, etc. It is possible that the tiles may be provided under the same description in the future, but the UV reflectances could be different.

Using our measurements, and the reflectance data for water-based paints obtained in other studies (18–20), we estimated the irradiance incident on horizontal and vertical surfaces for standing and sitting men and women whose heights were in the 99th percentile (24,25). The calculated irradiances could theoretically pose a hazard for persons who were sitting or reclining for long periods (e.g. hospital patients), especially if the person had a sensitive skin type or was taking photosensitizing antibiotics. However, none of the calculated irradiances appeared to pose a realistic hazard for normal workplace conditions, especially if a slight downward gaze is included to simulate the performance of normal tasks. This is qualitatively consistent with the results of a study of in-room exposures from upper room UVGI to 16 subjects at hospital, classroom and homeless shelter rooms (27). In that study, although some of the irradiances measured in that study would raise concerns for UV hazard, none of the measured 8-h doses exceeded the ACGIH UV hazard limit (1).

Our estimates are not a substitute for actual radiometric measurements, as wall surfaces may be composed of materials with significantly higher reflectance. For example, the reflectance of white plaster ranges from 0.4 to 0.6 (22). Such materials would need to be painted over with a paint or coating that reduces the 254 nm reflectance prior to the installation of a UVGI system. This study provides examples of the influence that the reflectance of ceiling tiles and paints can have on improving the safety of a UVGI installation. The IES has recommended that all upper-room surfaces that are likely to be irradiated have reflectances “as low as 5%” (20). This study is not only consistent with that statement but also suggests that higher reflectances than 5% could also be acceptable. Even if materials with relatively low UV reflectances are used, maintenance workers would still be required to follow standing operating procedures that include deactivating the UVGI source if it is necessary to perform any tasks in the UVGI source’s emission plane. If the UVGI source cannot be deactivated, then personal protective equipment like face shields and clothes with tightly woven fabrics that cover the skin would be required (30).

Design and operation guidelines for the application of upper-room UVGI have been developed in other studies (31,32), although no application standard yet exists. These data on ceiling tiles may now be included as part of this discussion, if such a standard is developed. Design guidelines for upper-room UVGI with low-pressure mercury lamps should include not only a maximum reflectance of 0.10 for all upper-room surfaces but should also require that the facility pass an acceptance survey by a UV radiation safety officer, performed with a UV-C meter with diffusing optic with a calibration that includes the 254 nm wavelength. Any areas where more than 0.2 μW·cm−2 irradiance is measured for a horizontal or vertical surface should only be considered unsafe after a realistic assessment of the exposure duration for room occupants within those areas, as well as their movements during that period (9,30,31). It is very important to note that our sample calculations were based entirely on reflected UVGI. It is also not uncommon (in sufficiently large rooms) for there to be a direct line-of-sight to the UVGI lamp. The presence of a line-of-sight would depend on the spacing and depth of the louvers, the angle of the emitted beam, and the relative height between the exposed person and the fixture. In our example of a 12′ × 15′ room, this was not an issue. However, if the room dimensions were doubled in size, then a direct line-of-sight to the UVGI lamp would be possible for persons standing in the more distant areas of the room. First and Nardell et al. noted such risks in a discussion of UVGI lamps installed in a variety of settings (31), and suggested the use of baffles or sconces to block direct line-of-sight where necessary. This again shows the need to require that each facility with UVGI lamps pass an acceptance survey by a UV radiation safety officer, which has been noted in other studies (4,31).

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Estimate of reflected irradiance at various locations within a sample room
  7. Discussion and Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

Acknowledgements— This project was supported in part by an appointment to the Internship/Research Participation Program for the US Army Public Health Command (USAPHC) administered by the Oak Ridge Institute for Science and Education through an agreement between the U.S. Department of Energy and the USAPHC. The views expressed in this article are the views of the authors and do not reflect the official policy or position of the Department of the Army, the Department of Defense, or the U.S. government. Use of trademarked name does not imply endorsement by the U.S. Army, but is intended only to assist in identification of a specific product.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Estimate of reflected irradiance at various locations within a sample room
  7. Discussion and Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information
  • 1
    American Conference of Governmental Industrial Hygienists (2011) American Conference of Governmental Industrial Hygienists (ACGIH): TLVs and BEIs. ACGIH, Cincinnati, OH.
  • 2
    American National Standards Institute/Illuminating Engineering Society of North America (ANSI/IESNA) (2005) Recommended Practice for Photobiological Safety for Lamps and Lamp Systems—General Requirements (ANSI/IESNA RP-27.1-05). IESNA, New York.
  • 3
    Reed, N. G. (2010) The history of ultraviolet germicidal irradiation for air disinfection. Public Health Rep. 125, 1527.
  • 4
    CIE (2003) Ultraviolet Air Disinfection (CIE 155:2003). CIE, Vienna.
  • 5
    Hockberger, P. E. (2002) A history of ultraviolet photobiology for humans, animals and microorganisms. Photochem. Photobiol. 76, 561579.
  • 6
    Riley, R. L. and E. A. Nardell (1989) Clearing the air: The theory and application of ultraviolet air disinfection. Am. Rev. Respir. Dis.. 139, 12861294.
  • 7
    Escombe, A. R., A. J. Moore, R. H. Gilman, M. Navincopa, E. Ticona, B. Mitchell, C. Noakes, C. Martinez, P. Sheen, R. Ramirez, W. Quino, A. Gonzalez, J. S. Friedland and C. A. Evans (2009) Upper-room ultraviolet light and negative air ionization to prevent tuberculosis transmission. PLoS Med. 6(3), 111.
  • 8
    Centers for Disease Control and Prevention (CDC) (2009) Environmental Control for Tuberculosis: Basic Upper-Room Ultraviolet Germicidal Irradiation Guidelines for Healthcare Settings. (DHHS National Institute of Occupational Safety and Health (NIOSH) Publication no. 2009-105). CDC NIOSH, Atlanta.
  • 9
    Nardell, E. A., S. J. Bucher, P. W. Brickner, C. Wang, R. L. Vincent, K. Becan-McBride, M. A. James and M. M. Wright (2008) Safety of upper-room ultraviolet germicidal air disinfection for room occupants: Results from the tuberculosis ultraviolet shelter study. Public Health Rep. 123, 5260.
  • 10
    Zuclich, J. A. (1998) The cornea-ultraviolet action spectrum for photokeratitis. In Measurements of Optical Radiation Hazards (International Commission on Non-Ionizing Radiation Protection (ICNIRP) 6/98, Commission International de l’Eclairage (CIE) x016-1998) (Edited by S. DiDomenico, R. Matthes, P. Murray, R. Phillips, D. Sliney and S. Wengraitis), pp. 143158, Märkl-Druck, Munich.
  • 11
    CIE/International Standards Organization (ISO) (1998) Erythema Reference Action Spectrum and Standard Erythema Dose (CIE S 007/E-1998/ISO 17166:1999(E)). [Standard] CIE, Vienna.
  • 12
    Parrish, J. A., K. F. Jaenicke and R. R. Anderson (1982) Erythema and melanogenesis action spectra of normal human skin. Photochem. Photobiol. 36, 187191.
  • 13
    CIE/ISO (2006) Photocarcinogenesis Action Spectrum (Non-Melanoma Skin Cancers) (CIE S019/E:2006/ISO 28077:2006(E)). [Standard] CIE, Vienna.
  • 14
    Bruls, W. A., H. Slaper, J. C. van der Leun and L. Berrens (1984) Transmission of human epidermis and stratum corneum as a function of thickness in ultraviolet and visible wavelengths. Photochem. Photobiol. 40, 485494.
  • 15
    CIE (2010) UV-C Photocarcinogenesis Risks from Germicidal Lamps. (CIE 187:2010). CIE, Vienna.
  • 16
    Vincent, R. L. (2010) Workshop—Introduction and Demonstration: Computer Aided Design (CAD) Program for the Design of Complex Upper Room UVGI Systems, Available at: http://www.ghdonline.org/uploads/2010-08-03_1500_Vincent_CAD_Workshop.pdf. Accessed on 31 January 2012.
  • 17
    Rudnick, S. N. (2001) Predicting the ultraviolet radiation distribution in a room with multilouvered germicidal fixtures. AIHA J. 62, 434445.
  • 18
    Illuminating Engineering Society of North America (IESNA) (2000) Nonvisual Effects of Radiant Energy-Effects on Microorganisms-Germicidal (Bactericidal) Ultraviolet Irradiance. In Lighting Handbook. (Edited by M. S. Rea), pp. 153155. IESNA, New York.
  • 19
    American Welding Society (1976) Ultraviolet Reflectance of Paint. Battelle Co., Columbus.
  • 20
    Kowalski, W. (2009) Ultraviolet material reflectivities (UVC/UVB range). In Ultraviolet Germicidal Irradiation: UVGI for Air and Surface Disinfection, pp. 491. Springer, New York.
  • 21
    CIE (1998) Practical Methods or the Measurement of Reflectance and Transmittance (CIE 130-1998). CIE, Vienna.
  • 22
    CIE (1979) Absolute Methods for Reflection Measurement (CIE 44-1979). CIE, Vienna.
  • 23
    McCluney, R. (1994) Introduction to Radiometry and Photometry. Artech House, Norwood.
  • 24
    Eastman Kodak Company (1986) Ergonomic Design for People at Work. Van Nostrand Reinhold, New York.
  • 25
    Auburn Engineers Inc. (1997) Ergonomics Design Guidelines. Auburn Engineers, Auburn.
  • 26
    Vasconez, S., P. Boyce and J. Brons (2003) Controlling Tuberculosis Transmission with Ultraviolet Irradiation. Lighting Research Center of Rensselaer Polytechnic Institute, Troy.
  • 27
    Kraiss, K. F. and J. Moraal (1976) Introduction to Human Engineering. Verlag TUV Rheinland GmbH, Köln.
  • 28
    First, M. W., R. A. Weker, S. Yasui and E. A. Nardell (2005) Monitoring human exposures to upper-room germicidal ultraviolet irradiation. J. Occup. Environ. Hyg. 2, 285292.
  • 29
    CIE (2003) Spectral Weighting of Solar Ultraviolet Radiation (CIE 151:2003). CIE, Vienna.
  • 30
    International Commission on Non-Ionizing Radiation Protection (ICNIRP) (2007). In Protecting Workers from Ultraviolet Radiation (Edited by P. Vecchia, M. Hietanen, B. E. Stuck, E. van Deventer and S. Niu), pp. 3536. ICNIRP, Munich, Germany, ICNIRP Document 14/2007, Oberschliessheim, 2007.
  • 31
    First, M. W., E. A. Nardell, W. Chaisson and R. Riley (1999) Guidelines for the application of upper room ultraviolet germicidal irradiation for preventing the transmission of airborne contagion—Part I: Basic principles. ASHRAE Trans., 105, 869876.
  • 32
    First, M. W., E. A. Nardell, W. Chaisson and R. Riley (1999) Guidelines for the application of upper room ultraviolet germicidal irradiation for preventing the transmission of airborne contagion—Part II: Design and operational guidance. ASHRAE Trans., 105, 877887.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Estimate of reflected irradiance at various locations within a sample room
  7. Discussion and Conclusions
  8. Acknowledgments
  9. References
  10. Supporting Information

Table S1. Total spectral reflectance of test samples (specular + diffuse).

FilenameFormatSizeDescription
php1193_sm_Reflectance-data.xlsx26KSupporting info item
php1193_sm_TableS1.docx37KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.