TECHNICAL FEATURE FIGURE 2 (a) Airflow patterns in an office space showing large recirculation zones in the occupied zone (b) and mean age of air (s) showing large values at the locations of large recirculation zone in the second cubicle. Mean Age of Air(s) (a) (b) >1800.0 1620.0 1440.0 1260.0 1080.0 900.0 720.0 540.0 360.0 180.0 0.0 infected individual is facing the two other individuals located in a cubicle away from the door. The office space is designed for a supply airflow flow rate of 135 cfm (63.7 L/s) or 3 air changes per hour (ACH), which corresponds to the total cooling load of 3000 Btu/h (879 W) for a 20 degree difference between the supply and return air temperature. Two CFD models of the office space were developed: one without UVGI and one with two UV-C fixtures for identical office spaces (Figure 1). These fixtures are placed on the opposite walls along the long side of the room. The two UV-C fixtures are assumed to generate equal UV-C output. The UV-C fixtures are equipped with the extended baffles that direct the UV-C radiation perpendicular to the fixture. The k-ω turbulence model was employed to compute the turbulent viscosity of the air. A computational mesh of about 0.75 million hexahedral cells was created by placing fine mesh near the strategic locations. The infectious aerosol release from an infected person is simulated as a continuous tracer gas emission. The ventilation performance of each configuration is evaluated using the Spread Index as described below. The probability of infection evaluated using the Wells-Riley correlation is also described. The mean age of air is computed by solving the passive scalar transport equation.6 The UV-C distribution in the range of 254 nm was computed using the discrete ordinates (DO) radiation model. The reflectivity of UV-C radiation from walls is assumed to be 0.05. The UV-C dose (D, J/m2) is computed by multiplying the UV-C intensity (W/m2) with 50 ASHRAE JOURNAL ashrae.o rg J U N E 2023 the mean age of air (s). The UV-C effectiveness (η) is computed using the following equations:1,3 η = 1-S S = e-kD (1a) (1b) Where S is a survival fraction - a ratio of the microbial population with and without UV-C and k is a species dependent inactivation rate constant (m2/J). The species is assumed to be the SARS-CoV-2 virus with a k value of 0.08528 m2/J, which is based on the average of rate constant data available at the time this study was performed.2,7 The transport of virus is computed as a scalar variable. This value was used in all simulations. As the virus is transported through various locations in the space, it experiences a cumulative inactivation as it passes through the zones of spatially distributed dose. As shown in Table 1, a total of four analyses were performed for the UV-C output varying from 0.35 to 1.5 W that corresponds to volumetric UV-C intensity of 4.6 to 19.6 (mW/m3). Spread Index (SI)TC Ventilation performance for each case was analyzed using the Spread Index, which is the ratio of the volume of the space occupied by the infection risk level above a certain value to the total volume of the space.8 Ideally, the ventilation systems should minimize the spread of contaminants and reduce the probability of infection everywhere in the space. Assuming the target concentration is a safe exposurehttp://ashrae.org