Document ypmXkYgddDk1YOxR8ZJLqzdY4

Laboratory air quality and room ventilation rates: An update This report extends previously published empirical testing of laboratory air quality at varying room air change per hour (ACH) rates under controlled releases of organic solvents. Additional testing was performed after modifying the ventilation controls to enable reliable operations at room ventilation rates as low as 2 ACH, and with additional solvents for some test conditions. The accumulation, peak concentration, and clearance of airborne contaminants continued to be inversely proportional to the overall room ACH rate, demonstrating that low laboratory ventilation rates require exceptionally long times to clear. By Robert C. Klein, Cathleen King, Anthony Kosior INTRODUCTION Previous work1 investigated the rela tionship between laboratory air quality and room ventilation under controlled releases of organic solvent during simulated routine bench-top work as weli as small spills across the range of 6-16 room air changes per hour (ACH). The accumulation, peak con centration, and clearance of airborne contaminants were found to be inver sely proportional to the overall room ACH rate and significantly influenced by the distribution pattern of supply (make-up) air. In addition, breathing zone measurements were nearly always significantly higher, earlier, Robert C Klein is affiliated with the Office of Environmental Health & Safety, Yale University, 135 College Street, New Haven, CT 06510, USA (Tel: 205 73 7 2131; fax: 203 785 7588; e-mail: rob.klein@yale.edu). Cathleen King is affiliated with the Office of Environmental Health & Safety, Yale University, 135 College Street, New Haven, CT 06510, USA. Anthony Kosior is affiliated with the Facilities Operations, Yale University, 2 Whitney-Grove, New Haven, CT 06510, USA. and more variable than those from the combined exhaust duct for general room and chemical fume hood exhaust across all ACH rates tested. This was not surprising since the exhaust duct represents the one location through which all room air ultimately passes, and therefore has undergone the most complete mixing and homogenization prior to sampling. Of the ACH rates previously evalu ated, the greatest relative improve ments in chemical concentration and clearance time from the room, regard less of release mode, occurred between about 6 and 8 ACH, with diminishing returns for rates greater than 12 ACH. This suggested that ACH rates above 12 are generally unnecessary while those below about 8 warrant careful consideration, with even lower levels suitable only for low-hazard opera tions or during periods of inactive non-occupancy. A key finding was that the re-engineering of supply air diffu sers to provide optimal location, num ber, and style is an effective means to functionally increase the efficiency of laboratory ventilation and potentially allow for designs at lower ACH rates. This current report updates previous investigations by extending the labora tory room ACH rate down to 2, some thing that was not possible earlier due to limitations of the ventilation control system. In addition, the effects of vapor pressure were evaluated by using addi tional solvents during some of the tests. As in the original study, these tests were designed to demonstrate differ ences in ventilation efficiencies in a typical laboratory room layout at vary ing air change rates, and should not be interpreted as a study of personal expo sure levels resulting from chemical usage or spills. MATERIALS AND METHODS Laboratory indoor air quality was assessed through measurements of volatile organic compounds released under two different chemical release modes, as described previously.1 Tests were performed in the same laboratory room as previous work, under nonoccupied conditions with the doors closed. The room was an unoccupied modern laboratory enclosed by walls, non-opening windows, continuous suspended tile ceiling, and doors, located in a building constructed in the early 1990s specifically for biome dical research. The original building design called for airflows of 14 to 15 ACH in laboratories and their support spaces. The room contains approxi mately 700 ft2 of floor area, with a floor-to-ceiling height of 9.5 ft. The room was operated with 100% exhaust (single pass) ventilation. Clean supply make-up air was delivered through overhead diffusers, and room air exhausted through a 5 foot wide che mical fume hood and a single overhead exhaust grill near the hallway door. Both the fume hood and the single overhead room exhaust grill were con nected to the same dedicated com bined room exhaust duct. As part of the original study, the number and type of supply air diffusers were modified to evaluate opportunities to improve 1871-5532/$36.00 dor: 10.1016/j .jchas.2010.10.002 Division of Chemical Health and Safety of the American Chemical Society Elsevier Inc. All rights reserved. 23 the most efficient arrangement using the three flush face radial flow style supply air diffusers. Reliable lower room ACH rates were accomplished by installing a second airflow damper in-line (manual set) in the supply air duct to provide a wider turn-down ratio than the previous existing damper alone could provide; no additional modifications were required on the exhaust side. Figure 1. Airborne concentrations generated from smaii spills (50 mL) of diethyl ether at floor level under varying ACH rates, as measured by photoionization detection from the combined room exhaust duct centerline. RESULTS AND CONCLUSIONS Chemical concentrations again varied predictably and significantly by ACH, mode of chemical release, and chemi cal, with patterns generally similar to those previously reported. As discrete events with a small limited volume of liquid, the spills showed a rapid increase in chemical concentration fol lowing spillage, a short duration peak after evaporation to dryness, and then a prolonged decay to background as 0 20 40 60 80 100 120 140 160 180 chemical vapor was diluted and Time (minutes) Figure 2. Airborne concentrations generated from small spiiis (SO mL) of acetone at floor level under varying ACH rates, as measured by photoionization detection from the combined room exhaust duct centerline. removed from the space (Figures 1 3). Chemical concentrations were sig nificantly higher at lower ACH rates and decreased as ACH rates were raised. We included two additional chemicals in this testing to evaluate the effect of vapor pressure on clear mixing and contaminant clearance Tests were performed while operat ance during small spills at various air from room air. ing the ventilation system identically as change rates. Given its high vapor Chemical concentrations were mea previous, but with additional runs at 2 pressure, diethyl ether had the fastest sured and datalogged at 1 minute and 4 ACH, expanding the overall test evaporation and shortest clearance intervals using photoionization detec range from 2 to 14 ACH. Of the origi times, while spills of acetone and etha tion (PID) instruments equipped with nal three make-up air conditions eval nol showed progressively slower and 10.6 eV lamps (Model ppbRAE, RAE uated, only the final arrangement was broader evaporation peaks, which Systems, Sunnyvale, CA). For this used during this additional testing, i.e., additional testing, measurements were only collected from the centerline of the combined room exhaust duct through which both the chemical fume hood and the single overhead general exhaust air travel. Measurements were again made under two different release modes: (1) small spills at floor level, and (2) continuous releases at bench level representing worst-case routine open air work. The bench-top work was performed exclusively with diethyl ether, while the small spills were gen erated with diethyl ether as well as with acetone and ethanol (95%) for comparisons across a moderate range of vapor pressures. Figure 3. Airborne concentrations generated from smalt spiiis (50 mL) of ethanol (95%) at floor level under varying ACH rates, as measured by photoionization detection from the combined room exhaust duct centerline. 24 journal of Chemical Health & Safety, March/April 2011 range of measurements to lower room ACH rates and other chemicals. Whether plotted by clearance time or peak concentration, measurements again showed an inverse power func tion with ACH. It can be concluded that compounds with even lower vapor pressures than ethanol will take propor tionately longer times to clear. Since all of these tests were performed with the optimized supply air distribution pat tern only, we also conclude that these Figure 4, Clearance time to background levels after small spills (50 mL) of diethyl ether, acetone, and ethanoi (95%), by ACH rate, as measured by photoionization detection from the combined room exhaust duct centerline. results would have shown significantly longer clearance times had the original supply air conditions been used. These results also support the recom mendations of ANSI/AIHA2 and others3,4 that local exhaust ventilation is critical for source control, and effec tively argue against establishing any sin gle ACH rate for all laboratories. This is reflective of the significant differences between laboratory spaces, their mechanical systems, and their hazar dous operations and materials, as well as the recognition that laboratories in real-life are used by humans and there fore dynamic and imperfect. These issues strongly support a risk-based Figure 5. Chemical concentrations generated from open benchiop releases of diethyl ether under varying ACH rates, as measured by photoionization detection approach to laboratory ventilation, with higher ventilation rates for more from the combined room exhaust duct centerline. Containers were left open for hazardous operations. While specific approximately one hour to achieve stable concentrations, and then recapped to observe differences in the speed of chemical removal from room air. The graph shows the last 20 minutes of open container evaporation, recapping of all contain ers at "0" time, and then the first 120 minutes of subsequent cfearance time. Although not shown on the graph due to scaling, clearance time to background took nearly 250 minutes at the 2 ACH rate. attention must be given to HVAC sys tem application and layout under all ventilation rates, this work confirms earlier conclusions that levels much below about 6 ACH should only be considered for laboratories using small quantities of non- or low-hazard reagents or for inactive, non-occupied became more pronounced at the low mode lengthened dramatically as the periods, regardless of the control sys est ACH rates (Figure 4). ACH rate was lowered (Figure 6). tem. In addition to creating potentially Continuous bench-top releases were These additional tests reinforce our hazardous conditions for occupants, again used to exaggerate open air previousobservations1 by extendingthe excessively low ventilation rates in most bench-top laboratory operations that emit contaminants into room air from multiple locations. During these tests, chemical concentrations showed gen erally stable steady-state plateaus, fol lowed by clearance to background levels once the containers were recapped and closed (Figure 5). The steady-state plateaus prior to recap ping as well as the speed of clearance after recapping again varied predicta bly by ACH rate. Tire highest steady- Air changes per hour state concentrations occurred at the lowest ACH rates and, like spills, clear Figure 6. Clearance time to background levels after open bench-top releases of diethyl ether, by ACH rate, as measured by photoionizaiion detection from the ance times under the bench-top release combined room exhaust duct centerline. Journal of Chemical Health & Safety, March/Aprii 2011 25 regions of the U.S. will create spaces that are difficult to heat or cool without the installation of local supplemental conditioning systems. ACKNOWLEDGEMENTS We again thank colleagues from Yale University's Facilities Operations and Physical Plant staff, including Larry Busillo for controls modifications and calibration, and Ed Lipsett and Dave Spalding for engineering sup port. references 1. Klein, R. C.; King, C.; Kosior, A. La boratory air quality and room ventila tion rates. /. Chem. Health Saf. 2009, 16(5), 36-42. 2. American National Standards Institute/ American Industrial Hygiene Associa tion (ANSI/AIHA). American National Standard for Laboratory Ventilation. ANSI/AIHA Z9.5-2003. AIHA: Fairfax, VA 3. DiBerardinis, L.; Greenley, P.; Labosky, M. Laboratory air changes: What is all the hot air about? /. Chem. Health Saf. 2009, 26(5), 7-13. 4. Memarzadeh, F. Effect of reducing ventilation rate on indoor air quality and energy cost in laboratories. /. Chem. Health Saf. 2009, 26(5), 20-26. 26 Journal of Chemical Health & Safety, March/April 2011