1、502 2009 ASHRAEThis paper is based on findings resulting from ASHRAE Research Project RP-1083.ABSTRACTThis research focuses on measuring the off-gassing of disinfection by-products (DBPs) from chlorine-based pool water treatment, filtration and chemistry of an indoor test swimming pool, to provide t
2、he Mechanical System Designer with specification guidelines for improved air quality. Aqueous derived effluent from the mechanical dehumidification (MDH) system is analyzed for chloroform (a trihalomethane or THM), chloramines and pH using commercially available products. Pool vapor content and comp
3、osition are determined by gas chromatography using standard NIOSH and OSHA analytical methods and includes preliminary nitrogen trichloride (trichloramine or TCA) measurement using published meth-ods. Based on observation, this research confirms and high-lights that nitrogen trichloride is the vapor
4、 (gas) that causes the most irritation and pool air containing TCA is essentially toxic over long term exposure. Tests conducted with top level ventilation and re-circulation rates confirmed the need to understand Henrys Law and the physical properties of TCA. When used in conjunction with deck (gro
5、und) level ventilation, low exhaust air movement at ASHRAEs current ventilation rate (0.5 cfm/ft2) can be successful in maintaining this heavier than air TCA gas at low levels within the indoor space.INTRODUCTIONChlorine (hypochlorite ion) is the predominant disinfec-tant of choice to keep indoor sw
6、imming pool waters within health management guidelines, building codes and occupant safety. This research focuses on the interaction and relation-ship between the indoor pool water chemistry, physical prop-erties and mechanics of the indoor facility and the ventilation pattern in order to provide sp
7、ecification guidelines for improved air quality. Present ASHRAE standards suggest ventilation air flow of 0.48 cfm/ft2of enclosed indoor pool and deck area plus 0.06 cfm/ft2of spectator area (ANSI/ASHRAE 2007).The amount of chlorine halogen injected into the pool water for disinfection is directly d
8、ependent on the greatly varied occupancy load, which determines the amount of biological soils brought into the pool. Chlorine reacts with the nitrogen functional group on urea, proteins or ammonia and produces chloramines, the most offensive of which is nitrogen trichloride (also known as trichlora
9、mine or TCA). TCA is very volatile, has limited solubility in water and is very easily off-gassed from aqueous solution (EPA 1999) (Kumar, et al. 1987) (Judd and Black 2000) (Lvesque, et al. 2006) (Scott 2003)(World Health Organization 2006) (Holzwarth, et.al. 1984).Additionally, the use of chlorine
10、 for the disinfection of swimming pool water can chlorinate various organic materials and produce trihalomethanes (THMs). One of the principal THMs coming from a pool surface is chloroform (Aggazzotti 1998) (Clifford 1992). Chloroform has a higher degree of water solubility and is not eliminated as
11、easily as TCA.Current research suggests that excessive exposures to TCA during early childhood can chronically damage lungs in children, causing airway changes that seem to predispose them to develop asthma and recurrent bronchitis (Bernard, et al. 2005)(Bernard, et al. 2007). Research has also prop
12、osed that a cause-effect relationship has been established between the presence of chloramines in the air of the indoor swimming Chemical Off-Gassing from Indoor Swimming PoolsRichard C. Cavestri, PhD Donna Seeger-ClevengerFellow ASHRAERichard C. Cavestri is president and director and Donna Seeger-C
13、levenger is a research associate and technical writer at Imagination Resources, Inc., Dublin, OH.LO-09-047 (RP-1083) 2009, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions 2009, vol. 115, part 2. For personal use only.
14、 Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAEs prior written permission.ASHRAE Transactions 503pools and the increased occurrence of asthma, irritant eye, nasal and throat symptoms in swimming instructors and life-guards (Mass
15、in 1998) (Lvesque 2006) (Jacobs, et al. 2007).To maintain healthy conditions for the occupants of recre-ational water, any ventilation and dehumidification design consideration must account for methods to limit TCA levels below the World Health Organization (WHO) recommended guideline of 0.5 mg/M3(W
16、orld Health Organization 2006).SPECIFIC ANALYTICAL PROTOCOLChloroform will off-gas from a swimming pool surface, but the salient compound of interest is nitrogen trichloride (TCA). TCA is a liquid at saturated pressure conditions but has a high vapor pressure, is heavier than air and is only partial
17、ly miscible in water. Light and temperature reduce its concentra-tion in water within several hours (Holzwarth, et al. 1984)(INRS 2006)(INRS 2005). Although problematic because of this quick elimination rate, TCA content in pool water can be determined by measuring a vacuum sample using the Institut
18、 National de Recherche et de Scurit (INRS) spec-ified procedure (INRS 2005).TCA also has a relatively high Henrys constant of 435 (Stottmeister and Voigt 2006), which means it can readily transfer itself to fresh air. As occupant levels increase in a conventional, athletic, leisure or waterpark appl
19、ication, the increased surface area of the bathers skin, water layer films and any kind of water/air agitation device (i.e. slides, water-falls) increase the amount of TCA released into the pool atmo-sphere. The Henrys constant of 435 plays an important role in the judgment of the mechanical system
20、designer as to how much the air volume level should be increased and how much pool surface air movement should be increased with concur-rent increase of complete air extraction. The ASHRAE suggested rate of 0.5 cfm/ft2was found to be adequate in the indoor test pool; yet, as the complication of air
21、movement increased due to bather load, deck level total air exhaust of 0.5 1.0 cfm/ft2was needed to keep the pool area at WHO guide-line levels of 0.5 mg/M3. Air changes per unit hour consisting of large volumes of either outside or re-circulated air do not reduce the increased release of TCA from t
22、he pool surface but instead only concentrate the material above the pool.Analysis of Indoor Pool WaterNormal analysis of the pool water was performed using a commercial wet chemistry colorimetric kit that addresses the five basic rules of properly managed pool chemistry: free available chlorine, com
23、bined available chlorine, pH, alkalinity and hardness. This type of pool kit appears sufficient to accu-rately and reproducibly manage pH, alkalinity and hardness, but it is unable to provide the levels of detection needed for management and control of chloramines.A spectrometric kit available from
24、Hach (Method 10172) more accurately detects chloramine levels between 0 10.0 mg/L Cl2 but is designed for chloramines detection and not specifically TCA. Recent research by INRS (INRS 2006) proposed a laboratory method of measuring the amount of TCA in water by extraction and transfer into the gaseo
25、us phase for analysis in the conventional airborne chloramines measurement methods (INRS 2005).A separate spectrometric method from Hach (Method 10132) was used to determine the total amount of THM from 0 200 ppb as chloroform in the pool water. It must be noted that chloroform is formed over a peri
26、od of many weeks and when dirt is introduced into the pool water. When only test urine is introduced in the absence of dirt, the amount of chlo-roform produced is minimal to none.Analysis of Indoor Pool AtmosphereThe air sampling method for THMs is detailed in the well known NIOSH-1003 and OSHA-1001
27、 techniques. Briefly, these techniques use specially prepared commercially avail-able adsorbent (ORBO) tubes wherein at least 50 L of air are drawn through a carbon matrix using a mini vacuum pump specifically designed for this application. Desorption of the tube with carbon disulfide followed by GC
28、 analysis with flame ion detector using capillary columns provides the concentra-tion in the ambient atmosphere.Synthetic urine was created following published analysis of the urea and protein content of human urine (NASA CR-1802 1971). Initially, to effectively introduce measurable levels of TCA, t
29、he amount of urine was based on a five-person bather load admitted every two hours into the 12 ft diameter test pool. Testing showed this was excessive and the amount was recalculated to a one-person bather load.Solid phase samplers for collection of the TCA air samples were constructed from literat
30、ure and a report published by NIOSH, essentially following the widely used method published by Hry (NIOSH 2006)(Hry 1995)(Thick-ett 2002)(INRS 2006)(BGIA-Arbeitsmappe 36). In order to have timed sampling rates, a manifold was constructed using timers, pneumatic valves, rotometers and vacuum pumps.TC
31、A air samples were drawn through PVC sampling tubing placed in the return air and exhaust air, as well as a pickup tube located five to six inches above the test pool surface. Air was drawn through the sampler at a rate of 2 L/min for four hours. Pool air containing TCA is first drawn through a sulf
32、amic acid treated silica gel pre-filter and then finally through the quartz or reagent glass fiber disks coated with sodium carbonate, diarsenic acid and glycerin. The filters are then desorbed with de-ionized water and analyzed for chloride ion by ion chromatography.DISCUSSIONInside our mechanical
33、engineering test facility, a sepa-rate, environmentally controlled room (16 ft 16 ft 12 ft) was constructed to house the test pool (5 ft deep 12 ft diam-eter), which totals 256 ft2inclusive of the deck area. The mechanical dehumidifier (MDH) was a one-ton dehumidifi-cation system with a fresh air du
34、ct attached to the intake duct before re-circulation. The pool atmosphere air flow was 504 ASHRAE Transactionscontrolled by ASME flow nozzle stations at rates of 0.7, 0.5, and 0.3 cfm/ft2. The original test pool design had an in-ceiling exhaust port, return air port 7 ft above the pool surface and a
35、ir re-circulation rate of 540 cfm. Room temperature was thermostated at 82F (72.89C) and the pool water was controlled via a propane heater to maintain constant test conditions.The original work involved filling a 16 ft 16 ft 7 ft volume (1792 ft3, 50 M3) with TCA gas at a constant rate with ventila
36、tion schemes of 0.5, 1 and 1.5 times the recommended ASHRAE rate of 0.5 cfm outside air per square foot of pool deck area. TCA gas was generated by injecting synthetic urine at a rate equivalent to that of five persons introducing sweat and urine into the water at 250 mL/hr per person. At these expe
37、rimental rates, no decrease in TCA content was found in the pool water, MDH condensate and exhaust air for each air level over a one week period regardless of the pool chemistry, as shown in Table 1. The TCA dose was cut to the equivalent of one person. Based on the subsequent observed (measured) TC
38、A air sampling, there was no drop in TCA levels over a period of several weeks.It was ascertained that since the air (vapor) extraction rate of pool air containing TCA vapor and any internal pool vapor contaminant is unknown and must be measured, it is also crit-ical to know the physical chemical pr
39、operties of TCA vapor (Table 2). Although pure TCA vapor density has not been fully measured because of its explosive properties, it is estimated based on the chlorine content to have a vapor pressure very close to trichloroethylene, chloroform and perchloroethylene (PERC).Known quantities of PERC v
40、apor with initial target samples of 46 mg/M3increasing to 100110 mg/M3were introduced into the test pool area with re-circulated air flow over the water surface of 12 ft/min. Pool air sampling was instituted 15 minutes after introduction and then repeated every four hours. The sampling points were f
41、ive to six inches above the pool level surface and at the return air into the MDH. Using gas chromatography, there was very little or no signal except at the 100 mg/M3mark.Although the air flow in the test room was approximately .303 air changes per minute (18.2 air changes per hour), it was possibl
42、e the low re-circulated air flow had a negative response. Therefore, a 500 cfm fan was introduced to drive air across the surface of the pool. At the 100 mg/M3mark, there was a very weak PERC signal detected in the return air, indicating down-ward air velocity is necessary to re-entrain vapor for re
43、-circu-Table 1. TCA Sampling Results5 Person Injection (750 ml)/2 hrs within 15 min 80F Pool, 82F AirConditionPool MDH ExhaustSoluble* Insoluble* H2O (ppm) Soluble* Insoluble*Condensate H2O (5 gal)Soluble* Insoluble*Baseline 510 750800 1100 1020 800 3540 510 500190 cfm 1015 300550 1150 1020 750800 1
44、042 510 650700128 cfm 7.510 550800 1150 1020 800850 4050 510 67580090 cfm Not determined due to low air flowPool = 6 above pool surface; MDH = Re-circulating air; Exhaust = Exhaust airSampling 2/day at 2.1 cfh (1.0 M3)*Average of 14 samples reported as chloride ion (mg/M3)Pool Setpoint Conditions: p
45、H = 6.97.0; Sanitizer = 3 ppm ORPTable 2. Test Vapor PropertiesTrichloramine Trichloroethylene Perchloroethylene Chloroform R-22NCl3C2HCl3C2Cl4CHCl3CHClF2Liquid density (g/mL)1.653 explosiveas a liquid1.462 1.623 1.499 1.206Henrys law constant (H 2025C) 435 550 851 170 .03Vapor density (g/L) N/A 4.5
46、 5.8 4.1 3.0Vapor pressure ( 20C)20 kPa(150 mmHg)7.7 kPa(58 mmHg)1.7 kPa(13 mmHg)21.2 kPa(159mmHg)938.4kPa(7038.5 mmHg)H2O solubility (g/100 mL) Nil 0.11 Nil 0.795 SlightBoiling point (C)71 86.7 121 62 -40.8Sources: “Applying Henrys Law to Groundwater Treatment”, Pollution Engineering 1996; www.osha
47、.gov; webbook.nist.gov; “The Fate of Chlorine and Chloramines in Cooling Towers”, Water Res. 1984; and ASHRAE Transactions 505lation. Because of its higher vapor pressure and lower boiling point than PERC, the test was repeated with chloroform vapor. The results showed the top exhaust duct was not r
48、esponsible for elimination of the chlorinated vapors (Figures 1 6). Since the top exhaust vent was shown to not be a factor, the exhaust scheme was changed to include perimeter deck level exhaust ports (Figure 7).Because R-22 is a gas at ambient conditions, it can be indicative of TCA off-gassing fr
49、om an indoor pool surface; therefore, additional testing was processed using R-22. Samples were taken from the pool surface (1 inches to 2 inches above surface), return air and exhaust ducts. The content of R-22 found in the exhaust air is indicative of the extraction rate. The resulting salient analytical vapor content showed deck level exhaustion had the highest amount of R-22 vapor and top level exhaust had the least amount (Figures 8 11). Very noteworthy was the fact that the re-circulated t
