On the power-law model for predicting the VOC emissions from office workstation systems

This paper examines the validity of using the power-law model (i.e., E = a t^-b, where, E is the emission factor in ug/h per workstation; t is time in h, and a, b are empirical coefficients determined from chamber test data) to predict the emission rate at 14^th day based on 7 days (or less) of emission test data. Theoretical analysis and experimental results from full-scale chamber tests of typical workstation systems are used for the investigation. Experimental results show that the power-law model represented within the experimental error the emission factors of the tested products over the 72h to 336 h period for compounds that had significant decay in emission factor, but not for compounds whose change of emission factor over time was within the experimental uncertainty. Predictions of the emission factors at 14^th day based on the 72^nd and 168^th data points using the power-law model ranged from 5 to 55% (with majority within 30%) of the measured values except for cases with very slow or negative decay rates (b<0.20) or low chamber concentrations that approached the lower limit of quantification (LOQ). Specific recommendations are made on the use of the power-law model to simplify the testing procedure and predict the emissions at 14^th day based on 7 day testing.