National Academies Press: OpenBook

Indoor Pollutants (1981)

Chapter: VI. Monitoring and Modeling of Indoor Air Pollution

« Previous: V. Factors that Influence Exposure to Indoor Air Pollutants
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 259
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 260
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 261
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 262
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 263
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 264
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 265
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 266
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 267
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 268
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 269
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 270
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 271
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 272
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 273
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 274
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 275
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 276
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 277
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 278
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 279
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 280
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 281
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 282
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 283
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 284
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 285
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 286
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 287
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 288
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 289
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 290
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 291
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 292
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 293
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 294
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 295
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 296
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 297
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 298
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 299
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 300
Suggested Citation:"VI. Monitoring and Modeling of Indoor Air Pollution." National Research Council. 1981. Indoor Pollutants. Washington, DC: The National Academies Press. doi: 10.17226/1711.
×
Page 301

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

HI MONITORING ED MODELING OF I~R AIR POLLUTION This chapter discusses the research tools required for measuring or estimating indoor pollution and exposure to it. Techniques and instruments used for the measurement of outdoor pollution may be modified for the sampling of indoor environments. Several problems emerge with such modif ications, and those problems are discussed here, as well as instruments designed specif ically for the sampling of indoor air. Personal monitors are increasingly recognized as powerful scientific tools for determining individual and population exposure to a it pollutants . Although they are still in the early stages of application, it is clear the personal monitors can yield data that are useful in associating human activities with exposure to air pollution. The benefits and def iciencies of personal monitors are discussed in a separate section of this chapter. The extent of indoor air pollution can be estimated with numerical models; mass-balance equations are used to estimate concentrations of indoor pollutants as fractions of outdoor concentrations and to estimate infiltration rates, indoor source strengths, pollutant decay r ates, and mixing factors . Several models have been developed, but f ew have been validated against data obtained from measurements. In estimating the total exposure of humans to pollutants {exposure to pollutants encountered indoor and outdoors, in industrial sites and other workplaces, etc.), it is essential to know not only the pollutant concentration., but also individual patterns of mobility and use of t ime . The available information pertinent to the last two characteristics has been gathered roo~;~cly by social scientists and, although interesting, does not meet the information needs for assessing exposure to air pollution. The final section of this chapter discusses the idea of total exposure and what knowledge is needed to measure it. FIXED-STATION SAMPLING AND MONITORING There is an extensive data bare on outdoor air quality, and much of the knowledge gained from studies of outdoor air quality is applicable 2S9

260 to the characterization of nonindustrial indoor environments. However, the characteristics of indoor air quality in residentis1 and commercial buildings and at other indoor sites can be quite different from those of outdoor and hea~y-industrial environments. Thus, a n',mher of special problems arise: the quality of indoor air is affected by a broad spectrum of pollutants from both outdoor and indoor sources; measurements of indoor air concentrations may require sampling instruments considerably different from those used in the outdoor or industrial environment; and the air volume inside a building is finite, and the rate of air exchange (especially in residential units) may be very low, and therefore, when air samples are drawn from an occupied space by external samplers, the s~mplinq flow rate must be so low a. to have only a negligible effect on indoor air movement and on the air- exchange rate. Because of the effects of equipment heat and noise, an well as occupant inconvenience, sampling and monitoring equipment should (and usually can) be placed in remote locations outside the building being evaluated. Thus, it is common practice is to locate the instruments outside the building space and draw air-~ple streams to them. Se~plinq technigues fall into the following broad categories: - · Continuous sampling : Provides ~real-time. sampling: required to observe temporal fluctuations in concentration over abort periods. · Integrated or continuous asm-Dlinq: Provides an average sampling over a specified period; used when the mean concentration is either desirable or adequate for the purpose. · Grab or spot sampling: Provides single samples taken at specified intervals; typically consists of admitting an air sample into a previously evacuated vessel, drawing a sample into a deflated bag for later analysis, or drawing (by mechanical pump) a sample through a sample collector to extract a contaminant from the sir; suitable when .spot. samples are adequate for the measurement of a pollutant and knowledge of temporal concentration Variation over short periods is not important. Some instruments sample and measure pollutants directly, and others ample for later laboratory analysis. The direc~c-reading inetruments required for continuous monitoring use various types of physicocl~emical detectors that can measure the concentrations of pollutants in situ. Integrated or grab-sampling method. are used when there ts no suitable concentration sensor available, when the pollutants of interest are present at concentrations too low to permit use of direct-reading instruments, or when sampling sites are inaccessible to bulky instruments. Further information on Sampling and measurement methods for air pollutants is available elsewhere. ~ 53 l.' CONTINUOUS MONITORING ~- Cont~nuous monitoring is a technique for sampling and measuring the real-time concentration of pollutants. Indoor sir quality in subject

261 to both temporal and spatial variations, and data on these variations would be needed to determine the concentrations to which occupants are exposed or to model indoor air pollution. The choice of monitoring techniques must be consistent with the types of information desired and the resources and manpower available. Although continuous monitoring has numerous benef its, it also has a number of disadvantages. Two positive features of continuous monitoring are that peak short-term concentrations can be determined, in addition to average concentrations calculated over any period, and that concentration variation as a function of time can be correlated with source generation , inf iltration-ventilation , and other characteristics. The availability of continuous-monitoring instrumentation depends on many factors, including the chemical properties of the pollutant and the range of concentrations to be measured. Continuous monitors are commercially available for all the gaseous pollutants that are designated ~criteria. pollutants by the EPA--carbon monoxide, sulfur dioxide, nitrogen dioxide, ozone, and total nonmethane hydrocarbon. . The EPA has specif fed performance criteria for the instruments used to measure each of these pollutants, and all analyzers that meet these specifications in performance sts are designated ~EPA-approved. Continuous-monitoring system a, even with high~quality instrumentation, are not troub.' ~-free. For example, continuous monitors are expensive and rep ire frequent calibration and routine maintenance. In addition, they have their own power and ambient- temperature requirements and can create safety, heat, and noise problems if they are placed at the sampling points. For these reasons, monitoring systems are generally designed to have all equipment for continuous analysis and recording at a single remote site, often a mobile laboratory. Such a laboratory usually contains facilities for calibration and maintenance, and it may also provide electric power and suitable environmental conditions for the equipment. If sampling lines made of flexible fluorocarbon tubing or other nonreactive materials are used, air from several sampling points can be drawn into the laboratory for analysis. One set of continuous monitoring equipment can be shared by several sampling sites if the individual lines are sequentially sampled.' In this scheme, all instruments obtain air samples from a colon manifold, which, in turn, is supplied with air from one of four sampling sites (one of which is usually outdoors) or from a calibration system. ' ~ The length of the sampling interval for each site can be determined by the response times of the individual instruments , the actual transit time in the sampling line, and the details of temporal information required at each sampling site. Continuous monitoring requires highly trained field personnel, rigorous quality-control (calibration} procedures, and provisions for quality assurance ~ independent performance audits of routine monitoring and data-handling operation" ~ . Securing electric _ power and a suitable location for a mobile laboratory equipped with sampling lines and cables can require long-term planning and entail considerable expense. This type of f~xed-station monitoring is not suitable for large-scale s urveys, because of these time and cost considerations . For large-scale survey work, integrated sampling and grab-sampling techniques are generally snore appropriate.

262 INTEGRATED SAMPLING Integrated sampling, in which a known sampling rate is maintained over some period, is commonly used for pollutants that must be accumulated to permit analysis. The period of collection may be several minutes or several weeks or month.. Analysis may be performed at the collection mite or in a laboratory. The data resulting from analysis of integrated samples are expressed as an average concentration over the sampling period. A variety of particles, gases, and vapors are sampled by this technique. Particles can be collected on filter media for later gravimetric and chemical analyses. Size-selective particle samplers, such as various dichotomous air samplers58 and portable cyclone sailers, i. l.2 are used in indoor aerosol sampling when it is desirable to determine the concentration of fine particles (less than 2.5 ~ in aerodynamic diameter} or respirable particles (less than 3.S am in aerodynamic diameter ~ . The samples can be analyzed by beta gauge or gravimetric techniques capable of determining mass concentration; by x-ray fluorescence, neutron-activa~cion analysis, etc., to determine their elemental,composition; and by ~ variety of separation and analytic techniques to determine chemical Composition. Aerosol samplers must be placed directly at the sampling sites, to avoid the particle losses that occur when air i. drawn through sampling lines. me sophistication of particle samplers ranges from hand-held unit. that require manual operation to fully automated units that can be programed to operate unattended for several weeks. Gaseous substances can be collected by both passive (diffusion-controlled) and active {powered bulk air-flow) samplers. Soluble vapors, such as formaldehyde and ammonia, can be collected by liquid gas washers and bubblers. Air sampling with bubblers, as well as with other accumulating sample collectors {such as ad~orbers and condensation traps), requires that the total volume of the air sample be accurately known. This can be accomplished with dry- or wet-test meters, which measure sampled volume directly, or by measuring or controlling the sampling rate and tie e . Many techniques have been used to measure the concentrations of radon and radon daughters. Because of the low level of radioactivity usually found in buildings, integrated measurements are often necessary. Passive devices that use sensitive thermoluminescent dosimeter (TLD) chips, 23 passive fi", or track~tch techniques25 can record alpha decay over periods of weeks or Months to determine average radon concentrations. Radon-daughter concentrations can be determined by Massing a known volume of air through a filter paper . · . . . ~ · _ ~ ~ (typically for 10 min) and then measuring total alpha acezvley on one filter with an alpha-decay ratemeter. Integrated sampling techniques have several advantages: they are less expensive and require lest manpower than continuous monitors, they can be used to measure concentrations that are too low to be measured directly, samples can often be analyzed later at a more convenient time or place, and average concentrations over long periods are easily obtained. But they also have some disadvantages: short-term temporal

263 information is lost; samples must be taken frequently if temporal variability in concentration in to be assessed; transporting the sample to its point of analysis may require special handling, special environmental conditions, or rapid delivery to avoid deterioration; and quality control may be more difficult deco implement. GRAB SAMPLING In grab sampling, one sample is collected over a very short period. Grab samples have to be taken frequently, if temporal variability in pollutant concentrations is to be assessed. It is usually the least expensive technique for field sampling, unless very f requent samples are required. It may simply involve f illing a container with an air sample and transporting it to a laboratory for analysis, or it may involve extractive sampling, as in calorimetric detector tubes. It is moat useful when the laboratory equipment required for analysis is at a remote location, when a very large number of samples are required, or when manpower and equipment are limited. Sampling vessels commonly used include plastic bottles, glass tubes f illed with adsorbent, stainless-steel containers, and bags of aluminum polyester (Mylar), P~C film, and fluoroplastic film. '' Grab sampling has been used to estimate concentrations of radon, tracer gases, and organic compounds. Grab sampling can be used to measure radon concentrations by pumping a known volume of air through a filter into a Tedlar24 bag, which is impervious to radon. The time at which the sample is taken must be recorded, and, because of the decay properties of radon gas, analysis must be performed within a few days. If necessary, the sample can be concentrated with a cryogenic trap or transferred directly into a z inc sulf ide scintillation chamber s ~ for alpha-counting. The bags are inexpensive and can be mailed, with manual pumps, to field sites. Similarly, air samples can be qualitatively and quantitatively analyzed for organic compounds with gas-chromatographic techniques. Grab sampling, with its low cost and minimal manpower requirement, is suitable for large-scale survey work. However, a number of problems are associated with this technique. No information other than an ~ instantaneous. concentration can be obtained, and this value could be greatly affected by something as simple as the opening of a door or window. Sampled volumes are relatively small, and the laboratory measurement technique must be sensitive enough to determine ambient concentrations directly. Inward and outward diffusion of various gases has been observed for many material'; used in collection bags, and leaks in the containers and connectors are common. Par tickler attention must be given to degradation, adsorption, oontamina~ion, transformation, and the possible formation of artifact pollutants. The expeditious transport of grab samples with reference to time, temperature, sealing, and handling is important. Quality control is difficult to maintain, but must be established before this technique can be used with confidence.

264 MONITORING OF VENTILATION RATE Indoor air quality is directly affected by the rate at which outdoor air enters a building. Ventilation can be used to maintain low concentrations of indoor-generated pollutants . In turn, hen comfort conditions, such ss temperature and relative humidity, can be the determining factors in setting ventilation rates. Measurement of the infiltration-ventilation rate, the meteorologic factors that affect it (outdoor temperature, wind speed, and wind direction), and comfort factors (temperature and relative humidity} can be an integral part of fixed-location field monitoring. ventilation systems; vary considerably. Detached single residential units are ventilated primarily by infiltration--the uncontrolled leakage of air through cracks in the building envelope {around doors and windows, through walls and floor joints, etc.~--and by the controlled opening of windows and dove. Large buildings are usually ventilated by mechanical systems of varied complexity. So-called f resin air enters detached residential structures by infiltration; the term fair changes per hour. (ach) is routinely used for thin source of ventilation--~1 ache means that a volume of outdoor air equal to the volume of the interior building space ~leaks. inside each hour. That does not idly that the incoming air drives out or displaces the old air as it enters; rather, it is assumed that perfect mixing takes place. In practice, however, perfect mixing is impossible to achieve. Tberefore, an estimate of outdoor-air flow rate is based on the assumption of perfect mixing and homogeneity of indoor air to facilitate calculating infiltration rates. I3y far the most c ~ only used method of estimating air~exchange rates is the tracer-gas decay technique..' In this method, a tracer gas is released into the building space at one or more points, possibly a; - ~ - ~~ t~ ~¢ ~~ _~ "~= "~= ~^ arm. -~ this way' an attempt is made to produce a uniform concentration throughout the building ~pace. If homogeneity is maintained, the decay of the tracer gas is exponential, and the infiltration rates can be determined by sampling the air at several times. The air-exchange rate can be obtained from the slope of a semilogarithmic plot of tbe natural logarithms of the pollutant concentration versus time. In a similar method, the equilibrium-ooncentration method, ~ tracer gas is released at a constant rate into the building space. .' In the s~eady-steee condition with perfect mixing, the indoor concentration will reach a steady-atate value. From this and the injection rate, We infiltration rate can be calculated. With this technique, although it is simple to perform, it often takes many hours to reach a steady-state equilibr in - . More complex tracer~gas system can measure infiltration rates on a semicontinuous or continuous basis.~, Many gases have been used for tracer-gas measurements. Some of the properties that such a gas should have are easy measurement at low concentrations, minimal interference from other air constituents, chemical stability, nonreactivity, lack of. absorption by building contents, a density comparable with that of air, safety for hens, lack of explosiveness and flannosbility, absence of

265 other interior or exterior sources, low cost, and ready availability. Some gases commonly used are sulfur hexafluoride, nitrous oxide, and ethane.~° Mechanical ventilation systems vary considerably in design and complexity, and methods chosen to estimate ventilation rates must be suitable for the systems under consideration. The methods commonly used to estimate the ventilation rate for systems that use recirculation include pressure-measuring devices (such as inclined manometers and U-tubes), velocity meters (such as pilot tubes, hot-wire flow~eters, heated-thermistor flowmeters, and heated-thermocouple flow~eters), mechanical gas-flow indicators (such as rotating and deflecting-vane anemometers), tracer-gas techniques, and heat-balance techniques.S, Care must be taken to distinguish between the total rate at which air enters a particular zone and the rate at which outside air enters the zone. Temperature-measuring devices suitable for continuous monitoring include thermocouples, semiconductors, and thermistors. Typical indoor temperatures range f rom 15 to 40 °C. Thermocoupler present problems with low voltage outputs near 0°C and have nonlinear characteristics, but only when the cold junction is at 0°C. Semiconductor temperature sensors Chat use integrated circuits and have a voltage output linear with temperature are suitable for continuous recording. The most common temperature probe for measuring temperatures in this range is probably the thermistor, because of its high resistance ratio (which yields large voltage changes for small changes in temperature), its linearized output, and its wide operating range. Temperature gradients can be large in a building and even in an individual room. Probes should be placed where they will sense the temperature experienced by the occupants. Temperature probes should be calibrated against mercury thermometers that meet the specifications of the American Society for Testing and Materials (ASTM}. Relative humidity can be measured with sensors of the human hair. type, which expand and contract with changes in humidity, or with dewpoint-measuring devices. Commercially available dewpoint hygrometers, based on the principle that the vapor pressure of water is decreased by the pressure of an inorganic salt, are well suited to continuous monitoring. Relative humidity can be readily calculated from dry-bulb and dewpoint temperatures. Relative-humidity measuring devices can be calibrated with the aid of sling psychrometers. PERSONAL MONITORS Over the last 2 decades, a wide variety of miniaturized air samplers have become available that collect gaseous and particulate samples from the immediate vicinity of people, even as they conduct their normal activities. The initial devices used battery-powered samplers, defined as ~nonpassive.. Although widely used. there devices are often larger and heavier than desirable. Mare recently, a variety of diffusion- and permeation-con~rolled samplers have become available. These ~passive. devices are applicable solely to gas- and

266 vapor-~ling and are very seall and light. They all use sensitive chemical or physical analytic methods. Three recent workshops reviewed candidate technologies for personal sampling and monitoring of air-pollution exposure. A BrooNhaven group.' identified potential methods for gas- and particle- monitoring. An EPA feasibility study38 identified useful methods for monitoring sulfur dioxide, nitrogen dioxide, and ozone. Another EPA symposiums' explored the use of available technology for health~ef fects studies and other uses . Blood carboxyhe~oglobin {COBb) can be used as a measure of the actual dose of carbon Monoxide received by a person. Respirable-particle concen~crations are also of prime concern in health-effect" studies; come of the factors involved in obtaining reliable data have been evaluated. t.2 PERSONAL SAMPLING DEVICES Gas-SamPling The major techniques developed for sampling gaseous pollutants are passive (based on membrane permeation or diffusion through a geometrically defined air space) and nonpassive (in which air-pu~eping devices draw defined air volumes through devices of known collection e f f iciency) . Passive S~plere. Passive samplers use the kinetic energy of gas Molecules and the efficiency of the adsorbent collector to extract pollutant Molecules from the air at ~ known rate. The Ampler must be placed at the collection site, but has no requirement for ~ pump, flow regulator, or batteries. Such samplers therefore have major advantages with respect to weight, cost, and Maintenance. mere are two basic types: diffusion and permeation. Their use is liaised by the rate and amount of gaseous diffusion through a geometrically controlled air space or by transport through a permeable membrane that is specific for the pollutants being sampled. The choice and use of the diffusion-collector technique require knowledge of the coefficient of diffusion of the pollutant to be sampled in air under conditions similar to those normally encountered. Humidity effects have been encountered; ~ese are most probably caused by changing absorbent efficiency. One diffusion sampler has been developed.'''. for the measurement of abut nitrogen dioxide. It uses the principle of diffusion through the bore of an open tube that defines the rate of transport to the collector. The quantity of nitrogen dioxide diffused fras the open end of the tube to the collector surface (triethanol~ine) is calculable by Fick's fires law of diffusion, which may be expressed as: QNO2 D{A/L}Ct,Wh~re QNO2 ~ number of stoles of nitrogen dioxide

267 transferred during time t, A a cross-sectional area of tube (cm2), L ~ distance from open end to collector surface (cm), C ~ concentration difference between tube entrance and closed end (mol/cm3), and D a coeff icient of dif fusion of nitrogen dioxide in air (cm2/s) . The required nitrogen dioxide absorbed ~ 2 iS given as a time-weighted average concentration for the -sampling period. 2 Substituting typical values of the parameters--D ~ 0.154 cm Is, A ~ 0.71 cm2, and L = 7.1 cm--simplifies the expression to: Q~O2 = 2 ~ 3 (ppm-h) x 10-9 . In principle, this method is applicable to determination of any gaseous air pollutant for which an efficient, selective absorbent is available and for which an appropriate analytic chemical procedure may be devised. The size of the sampler can be varied, but attention must be given to scaling factors. ~ ~ ' ° Diffusion samplers have reportedly been used for water vapor and sulfur dioxide," nitric oxide,' t Is aniline,~3 benzene, 4 Is ammonia,'2 carbon monoxide, and NQx ' s An activated-carbon element has been used as the collector in a badge that has an open grid to define the geometry of the gaseous diffusion port. 32 This method of sampling requires use of gas-chromatographic (GC) analysis for measurement of the specif ic gaseous pollutant absorbed. A variety of organic compounds can be sampled and measured by this technique. A large variety of permeation samplers used for monitor systems are available commercially. All use membranes fabricated and calibrated to control the rate of permeation of the pollutant to the collector, which may be a solid medium, such as charcoal or Tenex GO for specif ic chemicals . Process ing of the collected sample var ies widely; chromatographic or calorimetric procedures are commonly applied for measurement. Transport of a gaseous pollutant across a membrane resembles the diffusion process. ~°7 However, permeation involves solution of the gaseous species in the membrane . Specif ic interaction between the gas and the polymer matrix introduces variables. As in a diffusion collector, the concentration of the gas approaches zero on the side of the membrane next to the collector, causing a gradient that results in flow from the ambient-air side. The permeability constant, P. of a membrane is def fined by the equation N = PA (C1 - C2) /S, where ~ ~ rate of transport across the membrane (mol/s ), p a permeably ity constant (cm2/s ), A ~ cross- sectional area (cm2 ), S ~ membrane thickness (cm), C1 ~ concentration of gas on ambient side of membrane ~ ~/m3}, and C2 = concentration of gas on collector side of membrane ~ ~g/m3) . As a function of time, Nt = PAClt/S. If W - Nt, the amount of gas that passes through the membrane in time t, then. because P. A, and s are constant for any device and gas, W = Clt/k, where k = s/PA, or C1 = Wk/t. Thus, the amount of a gaseous pollutant, W. trapped in the collector medium is proportional to its concentration, C1, in the air. The value of k must be determined by calibration.

268 Membranes used in permeation collectors are made from polymeric materials, such as dieethylailicones, silicone polycarbonate, silicones, cellulose acetate, TFI: Teflon, F!:P Teflon, Mylar, polyvinyl fluoride, Iolon, and Silastic. Thicknesses of 2.5-25 x 10 3 cm have been tested. " The permeation collector with activated absorbents is particularly useful for organic pollutants when GO analysts ts abetted. Co~ercia1 monitors are available from the all Corporation35 and Du pant that allow determination of Pore than 80 compounds by this method. Sulfur dioxide,63 .. i~ chlorine,37 ~~ vinyl chloride I'' "' nitrogen dioxide, ~ .. alkyl lead, I" and benzene ~ " have been determined. The utility of a spectrum of absorbents in these passive collectors teas been tested.. Nonpassive Svate~. A considerable variety of smopling system using pu - s to move the air have come into use over the last 2 decades, including impinger systems and solid adsorbere for gases and impingers, filters, and infractors for solid particles. Directly indicating devices using impregnated papers, chalks, and crayons and stain~detector tubes have also been used. 'these techniques have been reviewed in detail by Linch55 and Saltz~n.~ A recent development in personal monitors involves ~ pump that is- positioned next to the wearer's diaphragm by a light harness. The volume of air pumped by the motion of the thoracic cavity is recorded by an electronic package, which may be checked by ~ detached readout system. This system samples air for gases through costed diffusion-tube collectors or for particles through sasI1 filters at flow rates of 7S-SOO ml/min, depending on the wearer's breathing rate. m e complete apparatus weighs 590 9. In conjunction with spira~eter calibration, actual exposures to measured pollutants may be calculated. Particle-Samplinq Particulate samples are collected by using the same principles used for large-acale samplers: filtration, impaction, and liquid impingement. The separation of the respirable fraction of particles is of considerable importance for personal monitors, and collection of adequate numbers of samples for analysis is critical. All particle- sa~plers use some device for souring the sir sable and for separating the respirable~particle fraction. Collection is preceded by such ~ device as ~ cyclone presa~ler. me relatively low power available to drive the air- - plink pop usually limits particle collection by personal monitors to filtration, either in ~ single stage or in a second stage that follows a cyclone that collects the large. nonrespirable particles. For respizable-particle-~mpling. the air-flow rate must be precisely controlled. The theory of aerosol collection by filtration has been extensively reviewed by Dorman,2i Pich, " Fuchs, 28 Green and Lane,36 Liu and Lee,5' and Lippeann.~. Small-a cale impactors suitable for respirable-dust-sa~pling with a personal monitor bane been described by Marble and Willeke.~.

269 Biologic Monitoring Measurements made on the human body and its excretions constitute an alternative way of measuring exposure to environmental pollutants. They include measurements of blood, urine, feces, and hair. The methods that can be used to relate environmental pollutant exposure to human composition have been reviewed at length in other NBC reports (on carbon monoxide, nitrogen oxide., and various trace metals). USE OF PERSONAL MONITORS IN EXPOSURE STUDIES Exposures to air pollution usually vary with a person's mobility patterns and activities. Therefore, estimating the total exposure of a person from one or a few air-pollution measurements at stationary locations cannot properly characterize the variation in a population's or a person's actual exposures. To evaluate health effects, it ts necessary to know actual personal exposures and the distribution Of those exposures in a population. The need for direct measurement of personal exposure to pollution has been noted by several author. 67 Personal monitors for various pollutants are commercially available. The Brookhaven workshop identified four basic experimental designs: 1. Use of Individual Air Pollution Monitors for Direct . Determination of EX ~ sure. Each person in the study population would wear or carry an individual air pollution ~ . . . . monitor during the course of the study. The same individuals would also be subjected to continuous or periodic evaluation of health responses Individual exposure and response would thus be measured. . . . Because of economic constraints, only relatively small populations could be studied by this direct approach. 2. Use of Individual Air Pollution Monitors to Adiust Results from Fixed Stations. As previously indicated, there can be substantial variations between area level measurements and personal exposure measurements. By monitoring exposure of individuals with individual air pollution monitors in areas also monitored with fixed stations, one would obtain the distribution of individual exposures in relation to measurement. obtained at the fixed stations. If one or several relatively constant relations were found in various areas, fixed-station data would then be corrected for use in estimating population exposures. 3. Use of Representative Sampling to Determine Subgroup Exposure. A carefully selected sample of the study population would be asked to wear or carry individual air pollution monitors. The sample would be stratified, grouping those expected to have similar exposures (e.~., office

270 workers or street workers). The measured exposure of each subgroup in the sample could be used as representative of the entire group. 4. Use of Individual Air Pollution Monitors to Calibrate Personal Activity Models. Activity models have been developed that describe how and where people spend their time.... These models could prove to be useful in estimating population exposure. They have not been applied in air pollution epidemiology except in a very limited way, and they could be best calibrated or verified through experiments using individual air pollution monitors. In such an experiment, a carefully selected sample of the study population would be asked to wear individual air pollution monitors, and their measured exposure would be compared with the estimated exposure of the activity model. Gaseous Poll utants . Carbon Monoxide. Ott and Mage73 collected 425 integrated carbon monoxide samples over 21 ~ in November 1970 and January 1971 in downtown San Jose, California. Breathing-zone samples collected while the subjects walked typical pedestrian routes were compared with those measured at the fixed monitoring stations. Me mean pedestrian exposure we. 1.6 times the mean concentration measured by the fixed monitors, but individual measurements of exposure varied from those measured at the fixed stations by a factor of up to 10. Ott and Mage concluded that the fixed-station monitoring data provide a relatively . poor measure of the true exposure of members or the general public to air pollutants. ~ Wright _ al. ~ ~ ~ sampled exposures in Toronto with portable carbon monoxide monitors. They demonstrated a Insubstantial discrepancy between the carbon monoxide concentrations detected 5,~ the provincial network of f ixed-s ite sampling stations and the much higher concentrations commonly met by people living and working in a large metropolitan area such as Toronto. ~ Cortese and Spengler~. measured exposure of Boston commuters equipped with portable carbon monoxide instruments (Ecolyzers}. They reported that 1-h exposures exceeded fixed-monitor measurements by a factor of 1.3-2.1; 8-h mean exposures were considerably below the 8-h mean from the fixed monitor. Wallace I'. carried a carbon monoxide dosimeter during 30 commuting trips by bus to his office in Washington, D.C. Concentrations inside the vehicles were typically 2-4 ti'Des those continuously measured in the central city at the fixed monitoring station (Figure ~rI-l). There was no correlation between the ambient and personal in-vehicle measurements. Nitrogen Dioxide. In a personal-monitoring study of children in Ansonia, Connecticut,. nitrogen dioxide and sulfur dioxide were

271 35 30 25 20 Q g O 15 5 i _ l 1~- ~ . ~ I I ' t I I 1 ~ Carbon Monoxide Concentrstions in Vehicle A Arnb~nt CO Concentrations in City I ~ / \~ During Commuting Hours ~ I I 1 1 1 1 1 8 101214 16 18 20 22 4 6 8 1012 14 16 JULY AUGUST FIGURE VI-l Carbon monoxide concentrations in vehicle, compared withl~:bient concentrations in city. Reprinted from Wallace.

272 measured with bubblers. i' Twenty boys were equipped with suitcase sailers that they carried for one 24-h day. Exposure to nitrogen dioxide tended to be greeter in children exposed to smoking at home, but the dif ferences were not statistically signif icant. AS with sulfur dioxide, seean personal exposure values {61.3 ~ 7.2 ug/~3) were significantly lower than mean outdoor nitrogen dioxide concentrations (100.1 ~ 9.0 ~g/~3~. Palmes et al. has described a passive personal Idler for nitrogen dioxide suitable for occupational exposures. The Awe device has been used to estimate lark average indoor nitrogen dioxide concentrations in 109 dwellings with gas stoves and nine with electric stoves in metropolitan New York. Tt was found that the homes with gas stoves had Significantly higher nitrogen dioxide concentrations than those with electric stoves. Average values in the kitchens with gas stoves approached the U.S. primary a~oblent-air quality standard of 50 ppb (annual average). These dosimeter~ were used in a personal-monitoring study of five families with gas stoves and four families with electric stoves in Topeka, Ransas.2. In each family, the husband, wife, and one child wore the dosimetere for four 1-wk apples. Dosieeters were also placed outside, in the kitchen, and in the bedroom. No significant differences were found between the personal-acnitor and outdoor measurements for the families with electric stoves (Figure VI-23. For the families with gas stoves, personal exposures were significantly higher than outdoor values and correlated best with the fixed dosimeters in the bedroom. No significant differences in exposures were found between family members. Sulfur Dioxide. Exposure to sulfur dioxide has been estimated in several studies by the calculation of ~ tIme~weighted average exposure from the time spent and average concentration. in various places. i. 27 Sulfur dioxide personal monitors have not been extensively used in f ield studies . In the personal-aonitoring study of children in Ansonia, Connecticut, no significant difference was found among the sulfur dioxide persons1 exposure measurements of the boys.. The personal samplers had a mean of 5.5 ~ 0.07 ug/~3, which was significantly rarer than the outdoor mean of 12.0 ~ 2.2 ~g/~3. The reports of daily acelvities showed that the children were indoors between `~~::~ and 80%of theday. Passive personal n~onitore using the collection principle of gas permeation through polymer membranes have been shown to be sensitive to 24-h average concentrations of sulfur dioxide am to 0.01 ppe. Sampling at ambient concentrations with treated filters also appears feasible. '' Bowe~rer, no personal-~nitoring results with these techniques have been reported. Organic Substances. Several passive monitors have been developed that may have the sensitivity to measure mean personal exposure to organic substances for sampling periods of 1 d to 1 wk.' 32 33 35 Their use has yet to be demonstrated in personal-monitoring programs in nonindustrial environments.

273 Electric S - - 50 FIXED ~ 30 At O 10 lo 70 at o - ~ sa tar at z 30 o c' o z 10 PER=HAL F B 1 B: l ( ' I 1 2 3 WEEK t _~ _ 1 1 ~ ~ I I 3 4 1 2 3 4 WEEK G" Sto'. PERS - ~ Wlfe __ ~ _— _ Chi~ 4 1 2 3 4 WEEK Hu*~nd FIGURE VI-2 Week-lon8 nitrogen dioxlte concentrations, ng/m3, for an electric-cooking and a gas-cooking family in Topeka, Kaneas, May-June 1979. B. bedroo~; K, kieche,<` O. outdoor. Reprinted with permission from Dockery _ al.

. Particles 274 Respirable Particles. In the Ansonia study,. personal exposures to respirable particles were significantly higher among children who lived with one or more smokers. The mean personal exposure, 114.S + 9.0 ~/m3, was significantly higher than the mean outdoor concentration, 58.4 + 5.9 ~g/m3. The outdoor high-volume samplers, however, collected both respirable and nonrespirable particles. The principal conclusion of this study wan that ~ child's exposure ~load. of air pollutants, especially respirable particles, ts determined primarily by indoor exposures. Personal exposure to respirable particles and sulfates has been measured in two cities as part of the Harvard X-city study, ~ in which 37 people carried personal respirable-particle monitors with them during the day. Fixed-station monitors were run simultaneously in the main activity room of each home and at several locations outside. There were at least three complete sample days for each person. Mean personal exposures to respirable particles and sulfates for each city were determined on the basis of mean outdoor concentrations. In each city, there were significant difference" in result" between individuals, as determined by their activities. A linear increase in personal exposure to respi.rable particles with the number of Smokers in the home and workplace was found. Lead. Berlandi et al.S have reported personal lead-exposure measurements from 2 d of sampling in metropolitan Boston. Samples collected while subjects were driving into Boston had a time-weighted average of 4.5 ~/m the first day and 3~7 ~/m3 the second; indoor personal samples were all less than 1 ~/m3. Fuga' et al.26 estimated average air lead exposure of an office worker in Zagreb. Air lead was measured with a personal monitor inside and outside her home and her office and at other mites. Table VI-1 shows that the average concentrations were highest in association with outdoor activities--6.3 ~/m3--with only 1.7 ~/m3 or lees from indoor sampling. A time-weiqhted average exposure was then calculated _ on the basis of her activities each week (Table VI-1 ) . me average weekly exposure was estimated to be 1.1 ~/m3, compared with the average measured value of 0.72 ug/m3 outside the subject's home. Fugas27 extended this method of estimating exposure to lead in air to a middle-sized industrial town in Yugoslavia. Air lead was measured at various locations indoors and outdoors during the winter of 1972-1973. Estimated exposures were considerably higher than the average urban monitoring-station value of O.9 ug/m30 Lead concentrations up to 10 times that found in buildings may be found in vehicles or outdoors. The contribution of outdoor and vehicular exposure to mean personal exposures may be small. because of the comparatively short exposure times.

275 TABLE VI-1 Calculation of Time-Weighted Weekly Average Exposure of Office Worker in Zagreb to Airborne LeAda Average Lead Integrated Concentration, Duration of Exposure, Location or Activity ug/m Exposure, h/wk ug-h/m Workplace I. 2 42 50. 4 Outdoor activities 6.3 14 B8.2 Recreation O. 2 6 1.2 At home: Rest of day O. 7 22 15.4 Night O. 3 48 14. 4 Weekends O. 5 36 18. O Tot al Weightet-average weekly exposure: 1.1 1lg/m3 aData from Fuga8 et al. 28 168 ., 18~.6

276 Biolos ic Indicator ~ Direct measurement of individual dose by biologic means is possible for several pollutants. This method intrinsically coapenestes for different rates of uptake by different persons, as well as for differences in exposure . Carbon Monoxide. Actual carbon Monoxide dose received can be measured directly by measuring blood carboxyhe~aoglobin. Stewart et al." - ' measured Cold of blood donors in 26 American cities. They reported CON concentrations higher than those expected from fixed-station monitoring data. Goldauntz3. co~red C—b measurements for nonsmokers from 1969 through 1972 with carbon monoxide measurements from 37 EPA fixed-station epitome in 1973. Golds~unt2 argued that the measurements from some of these fixed stations may be inappropriately high, because of their siting. Morgan and Morris6' calculated CQHh concentrations that would be expected if the population were in equilibrium with the measured fixed-station carbon monoxide concentrations by the relation, ~ COBb - 0.16[CO] + 0.3, where [CO] is the carbon monoxide concentration {pp~3. Comparing these calculations with nonsmoker COMb measurements, they concluded that tithe average dose indicated by Comb levels exceeded that predicted from the fixed-station data by ~ factor of 2.. Stewart _ al., " in ~ similar analysis of data obtained in Chicago in 1970 and 1914, reported Comb concentrations close to, but consistently higher than, those predicted from fixed-station measurements. m ey noted that the use of fixed-station measurements to define population exposure may not reflect worat-exposure situations, indoor exposures to carbon monoxide fray cigarette smoke, or exposures from faulty heating systems. Kahn et al. 56 demonstrated that CORb concentrations among nonsmokers in the St. Louis population are strongly affected by occupational exposure and by exposure to smokers. Lead. There is little doubt of ~ correlation between the high exposures to lead in the air of industrial areas and indexes of lead absorption, such as blood lead, urinary lead, delta-aminolevulinic acid dehydrase {A~}, and delta-an~inole~rulinic acid (DALA). me relationship between these measures of lead dose and the lower concentrations of air lead characteristic of the co~unity~.e., less than 10 vg/m3--is not well established. A National Research Council committee has stated that Chore precise studies are needed of the relation between atmospher ic lead exposure in the urban environment and the concentration of lead in the blood, perhaps by the use of personal ~oonitors..'. In an attempt to characterize this relationship. Azar et al. compared average exposures to lead in air measured by personal monitoring with the biologic indexes of lead absorption. Two groups of 30 tax t drivers in two cities and three groups of 30 Du Pont employees in three cities carried personal particle monitors with the. for 2-4 wk. Their exposure to air lead was calculated as a tiee~weighted mean for their exposures at home and at work. Blood lead was determined

277 weekly, and urinary lead daily. Different relationships were found between average exposure to airborne lead and thc logarithm of the blood lead concentration in each city. The plots of the data for all five groups, however, had ~ similar elope, with different intercepts. The authors suggested that the different intercepts indicate that variables other than airborne lead, presumably ingested lead, are affecting blood lead content. No comparisons with fixed monitoring were made. MODELING OF INDOOR AIR QUAI'I]?Y The value of an indoor air pollution model is twofold. First, it provides a framework for interpreting experimental results and for planning new experiments. Specifically, a model is useful in relating indoor pollutant concentrations to various geometric, ventilation, source, and sink parameters. Modeling can be used to determine the accuracy and precision to which various quantities must be measured if the desired accuracy of prediction is to be achieved. It can also be used in sorting out trends in the experimental data. Second, and more important, a model provides a means to predict accurately come desired function of concentration {such as peak concentration or dosage) for placer and conditions other than those tested experimentally. In epidemiologic studies, it is important to consider the quality of the air to which subjects are actually exposed, in many cases, the air quality associated with the home, the mode of transportation, and the workplace should not be taken to be the same as that associated with the outside . Indoor-air~quality models are developed to aid in understanding and predicting indoor ai~-pollutant concentration. and dosages as functions of outdoor air-pollutant concentrations, indoor~outdoor air-exchange rates, and indoor air-pollutant sources and sinks. Air pollution indoors may be of outdoor or indoor origin. Outdoor pollutants may enter a structure through infiltration or ventilation. Pollutants of indoor origin may arise from point or diffuse sources. Regardless of their source, air pollutants may be transported and dispersed throughout various regions of the enclosure. Some pollutants may be removed by filters through which the makeup air or the recirculated air flows, by exf iltration or ventilation to the outdoors, and by chemical change. In the case of particles, surface removal and generation are often important. Given familiarity with the system to be described and with the purpose of developing and using an indoor air pollution model, the starting point in developing a model in usually a statement of the mass balance concerning the pollutant of interest. For example, consider a structure of volume fir, in which makeup air enter. from the outride and passes through a f Liter at a rate q0 . Part of the building air is recirculated through another filter at ~ rate ql, and air infiltrates the structure at a rate q2. Bach filter is characterized by a factor F - (Cinlet ~ Coutlet) /Cinlet In this example, the pollutant

278 concentration t8 ass~ to be uniform throughout the structure. The indoor and outdoor pollutant concentrations at tine t are C and CO, respectively. me rate at which ache pollutant ts added to the indoor air owing to internal sources is S. The rate at which the pollutant ts repoured from the air cuing to internal sinks is R. In this case, the appropriate starting equation is: TIC =qOCO(l—FO)+q~C(1—F~+qCO—(QO+q'~q2)C+ S — R. dt Input rate due to makeup, recircu- Output rate Source Sink lated, and infiltrated air rate rate (1) The decay rate i8 a function of C; however, in modeling indoor air quality, the sink rate is often considered constant. As indicated in Equation 1, solutions to mass-balance equations invariably Contain parameters that must be evaluated independently. Geometric parameters, such as volumes and surface areas, can be manured directly or obtained from blueprints. Accurate values of the ventilation parameters are usually Marc difficult to determine. Experimental techniques for obtaining net exchange rates between indoor and outdoor air have been reviewed by Georgii3' and Gilath.~i The use of sulfur hexafluoride, SF6, as ~ tracer for air-exchange-~ate studies appear" to be increasingS air maples can be collected by hand in the region of interest, 22 remotely, .. or with author ted instruments. '. Without forced ventilation, air exchange is due primarily to infiltrations in most forced-ventilation systems (which are balanced so that negative pressures are not created inside the building), the rate of infiltration is negligible, compared with the forced-`rentilation, rate. Although ventilation parameters are often hard to obesin, the most difficult to evaluate are usually those associated with the rate at which the pollutant is being released or being removed (i.e., the strengths of the sources and sinks). In view of the uncertainties associated with many of the parameter values and the difficulty of doing otherwise, coaparments have been widely used in modeling indoor air quality. Traditionally, ~ compartment is defined as a region within which spatial variations in pollutant concentrations can be neglected over the time scale of interest. At any given instant, the concentration of a pollutant might vary subetant$ally throughout the region of interest. Bowever, if dosages are similar throughout the region of interest over inter~rale that are shorter than the time during which receptors {people, plants, equipment, etc.} are emoted, and if the damage is primarily ~ function of the dose, then the region may be treated as a c~part~nt. Depending on ventilation conditions, ~ single room, a floor, or a whole building may be adequately approximated as ~ single c~par~nt. Bowever, when either sources or sinks are not uniformly distributed throughout the region of interest, and the rate of mixing throughout the region of interest is low, compared with the characteristic residence time, then the aingle-comparment model may not provide an

279 adequate description. For example, stratification in ~ room cannot be neglected when one is describing the movement of smoke and toxic gases associated with building fires however, even in such ~ case {where intense stratification ts to be expected}, only two compartments (coupled) were needed to obtain a satisfactory description. Sulfur hexafluoride tracer experiments. conducted with average- sized rooms (20 x 20 x 8 ft) in which one or more persons were moving and in which the air was being exchanged about 3 times per hour, have suggested that associated eddy diffusivities are around 103 cm2/s (D. D. Reible and F. B. Shair, personal communications; thus, about 5 min after an instantaneous point-source release, tracer Concentrations (although decreasing) were about equal throughout most of a room. The solution to the two-coPpartment model with constant coefficients is presented below, after a brief general discussion of multicompartment models. Examples of two-compartment and single- compartment models of indoor air quality are also discussed. MULTICOMPARTMENT MODELS - Most e-compartment models have been (or probably will be) described by n coupled first-order linear ordinary differential equations of the form: At ~ alX1 ~ a2X2 + a3X3 + "anXn + an ~ 1 dx2 blX2 ~ b2X1 + b3X3 + bnXn + bn + 1 dxn at .. .. In general, the terms al, bl, etc., represent the sum of f irat-order losses from the compartment due to exhaust strewing, filtration of any recirculating stress, and sources and sinks due to f irst~order chemical reactions. In most cases, the sources and sinks due to chemical reactions may be simulated 8$ pseudo first~rder, because of the low concentrations (parts per million, or less ~ of the pollutant. In cases where higher-order chemical reactions are important, the model equations will be nonlinear and generally will have to be solved numerically. In the case of particulate pollutants, the parameters al, bl, etc., will probably contain loss terms, owing to surface deposition. The coefficients a2 ... an' b2 ... ten' etc., represent the gain of pollutants in various compartments that may result from the intrusion of air from other compartments. me terms an ~ 1, An + ~ ~ etc., represent the sums of the zeroth~order source and sink terms associated with each of the compartments . As indicated by Equations 2, an e-compartment model will contain ntn ~ 1) parameters, whose values should be determined independently. Any temptation merely to fit the cats through blind adjustment of the (2)

280 values of the perimeters should be resist", if the Mel is to be of broad value. The aid of any model should be to explain {and predict} ss many data as possible with the Tallest possible number of ad j ustable parameters . ~ Because it is always poasiblc to define X~ as the concentration of the pollutant in the nub coapar~nt at any tiee minus =e initial concentration, the initial conditions for equations 2 may be taken as XltO) ~ X2~0) ~ Xn(°) '° T~COMPAR]MENT MOD131`S ~- In general, the equations that deacribe two~ par~nt ~els are of the form: dX1 + alX1 ~ a2X2 + a3 at and dX2 + blX2 ~ b2X1 + b3 at The initial conditions are: (4) X1 ~ X2 ~ 0 at ~ - 0. {5) When the coefficients of Equstione 3 and ~ are constant. fib 2b2 ( ,b~ + 2b2) t[(a,—by +4a b ],s) ~83 ( e ; e ) (6)

281 and x2 where and + a3b: l/a,b3 +a3b2\ 1 ye so ~ —b aide—a2D2 \a,b,—a2b2/ Ada—b,)2+4a2b2~J 3 \~(a~—b~2+4a b If/ x--(a~+b~/2~(a'—b')2+4a2b2~/2 §-- fan + b')/2—bias—by ~ 4a2b2 ~ ~s/2. In Equations ~ and 7, the first term on the right side represents the steady-state solutions that are reached after the transient terms '--- Woods et ad..__ _ ~ thermal and ventilation requirements for laboratory-animal cage environments lo.- Alan W~Ql°~\ -~ ~ ~ _~ ~~ decay. U;sc:u a cwo-comparemene model in their analysis of _ ~ ~ ~ — __ ~~ ..~_ ~ . 4~= rev AL Inks were one rem anu one animal cage. Mass or energy balances for each compartment were coupled by both free convection and forced circulation of room air through the cage. Their model. permit estimation of dry-bulb and dew-point temperatures and concentrations of gaseous particulate contaminants in cages, as well as in a laboratory room. Such models can be used to determine an acceptable means of safely reducing room ventilation rates with implications of reduced enerav COn,;umDtiOn Ann operational costs. Miller.. has used a two~compartment model in his description of the reentry of the exhausts from laboratory fume hoods. The two compartments were the building and the building wake (from which the makeup air is drawn). Isiller64 and Sasaki et al.92 have shown that the reentry of fume-hood exhaust is a much more perv;~si~re problem than is commonly recognized. Sulfur hexafluoride tracer experiments have shown that reentry of a portion of the fume-hood exhaust is usually the dominant factor in determining the concentrations of the pollutants to which all persons are exposed in the laboratory building. Indoor concentrations of fume-hood exhausts, normalized to the source ~ =~ _— —~ —~~~e ~~e— (8) (a)

282 strengths, range from about 1 to about 350 ppb per mole released per hour. Although the chemical nature of the fume~hood emission is of prime importance, persons typically complain often when they are in buildings whose normalized indoor concentrations are above 100 ppb per male released per hour from fume hop. SI=LE~COMPA"MEN!r MODELS Lidwell and LovelockS. were apparently among the first to compare concentrations of a pollutant with a mass-balance model. Their model involved the instantaneous introduction of a nonreactive pollutant into a room and considered the dilution resulting from a constant ventilation rate in which the input air was pollutant-free. They noted that, when the air in the room was not well mixed, the dilution by ventilation sir was not necessarily exponential; nor were the rates of dilution the same in all parts of the room. A portion of the inlet air stream often tends to bypass part calf the room. For instance, when both the inlet and exhaust duct. are on the ceiling, the lower half of the room {and especially the corners ~ is apparently bypassed and the air in it is diluted more slowly than expected. Brief. suggested the use of a mixing factor (a constant, usually rang ing in value between 1/3 and 1/10, that multiplies the ventilation rate) to account for dilution rates that are lower than would exist if the roar air were continually well mixed. Con~tancei5 also recommended the use of mixing factors. Drivas et al. 2 2 derived mixing factors ranging in value between 0.3 and 0.7, except when fans were used; with fans, the character istic time for mixing the air throughout the room was short, co~red with the characteristic residence time, and the mixing factors were close to unity. Milly'5 used a single-compar~nt model involving the instantaneous introduction of a nonreactive contaminant with a pollutant-free input air stream in his discussion of chemical attack of tanks and fortifications. Caller used a single-compartment model in his analysis of the protection afforded by building against biologic~warfare aerosol attack, he permitted the outside concentration to vary with time and took into account the surface removal of aerosol by Beans of ~ firet~order sink term' his results can also be used to describe doses associated with radioactive or chemical contaminants. Calder 12 also used a Ingest Mel to calculate dosages associated with the penetration of a forest canopy by aerosols. Milly and Thayer.. developed a technique for predicting indoor dosages of pollutants generated outdoors, on the basis of a single-co~art~nt model. Turkey presented a detailed analysis of the transient behavior of a single-co~rtment model involving ~ constant generation term, and a constant outside concentrations he then considered several special cases during hi'' analysts of the measurement of odorous vapors in test chambers . Bunt.. used ~ single-compartment model with a constant internal source and a first~order sink tens to interpret data regarding airborne

283 dust in post-office facilities. Bunt et a1.~. and Cote and Holcombe~' used a ningle-compartment model in their investigations of nonreactive gaseous pollutants indoors. Bridge and Corn' used a single- co~partment model to predict concentrations of carbon monoxide and particles associated with smoking of cigarettes and cigars; their results were in good agreement with measured values. Sabersky et al. ,° reported that a single-compartment model involving an outdoor concentration that varied sinusoidally in time and a f irst~order heterogeneous (surface] decomposition term gave qualitative agreement with data for indoor concentration. of ozone. Shair and Beitner' s started with a single-compartment model to develop a ~linear-dynan~ic. model by which the indoor concentrations of ozone can be related to those outside by means of a simple expression. Tests conducted with 24 forced-ventilation systems in 13 laboratory-off ice buildings Fielded values of k (the heterogeneous-loss constant) of 0.02-0.08 cn~ /cm2-s, with an average of 0.04 cm3/cm2-s. t. To save energy, the makeup-air flow rate In buildings has been reduced . In one case, the reduction in the makeup-air flow rate was sufficient to permit economical selective filtering of the makeup airstream (with activated charcoal) during the times when the outdoor a it quality was relatively poor (see Figure VI-3 ); use of the activated-charcoal filters only when needed and replacement of inexpensive pref liters every couple of months extended the life of the actisrated-charcoal f liters to about 3 yr . This system was designed with the aid of the ~linear-dynamic. model. ~ ~ Kusuda s ~ used a single-compartment model to examine the feasibility of intermittent operation of mechanical ventilation systems with an eye to conserving energy while maintaining acceptable indoor air quality. Moschandreas (personal communication ~ used a single-compartment model of air pollution in nonworkplace indoor environments in Baltimore, Washington , D.C., Pittsburgh , Chicago, and Denver . He monitored carbon monoxide, nitric oxide, nitrogen dioxide, sulfur dioxide, ozone, · methane, total hydrocarbons, and carbon dioxide continuously for per iods of approximately 14 d in each of f ive detached dwellings {townhouses}, six apartment units, two mobile homes, and one school. In addition, there was a 5~d period of monitoring in one hospital. The model discussed by Moschandreas et al. 72 73 explicitly included a chemical-decay term that was validated with the data base just mentioned. Numerical predictions of hourly carbon monoxide, nitric oxide, nitrogen dioxide, carbon dioxide, and nonmethane- hydrocarbon concentrations were found to be within 209 of the observed values 80% of the tine. The model did less well in predicting the indoor sulfur dioxide and ozone concentrations; this was attr ibuted to the chemical reactivity of the pollutants. A study was implemented to rank the sensitivity in magnitude changes in output of the model caused by perturbation of an input parameter . The rank ing of input parameters, in ascending order of sensitivity, is as follows: initial condition and volume of the structure, indoor source, and pollutant decay and air-infiltration rate of the structure. 72 Shair et al. used a single-compartment model to describe the moisture content of bathroom air during and after the use of a shower;

0.45 0.40 0.35 0.30 [o3] 025 PP~ 0.20 0.15 0.10 0.05 o 284 _~1 - 1.--- 1 RUN NO. 12 JULY 22, 1975 OUTDOOR CONCENTRATION Of OZONE lNl:)OOR CONCENTRATION OF OZOIME-WITHIN Ist, 2nd ond 3rd FLOORS OF SPALDiNG LABORATORY t ~ x' ~ ~ TlV fOR 0~ I ~ '4` ~ \ my' ~ / ~_~\- t -A ~ i Too _ ~~ - J ~ _ 1- i~SYSTEl~ I off I luX.S1StE~ ox 1:34} 2:00 2:30 3:00 3:30 PM TIME (PD.T.) Figure VI-3 Relative indoor and outdoor ozone concentrations. From F. H. Shair (personal c~nication).

285 they found good agreement with e~eri~ntal results by considering the solid surfaces in the bathroom to be a sink while the choir is on and a source shortly af ter the shower is turned of f . Repace and Lowrey.' have developed a one-compartment model that describes growth equilibrium and decay of tobacco~moke aerosol under different room mixing conditions. Bollowell et al. .2 have discussed the impact of radon on indoor air quality. Rusuda et al.S' used the available indoor-radon data to develop a single-compartment model with ~ firat~order radioactive~decay term and a constant generation rate: dC - qCO - ~VC ~ VS, at (101 in which V is ~roluree, CO is outdoor concentration, and C ts indoor concentration. In addition to terms previously defined, there is the radon~decay constant A ~ 1.258 x 10 4/min and the average source strength per unit volume of air, S. Setting the left-hand side of the above equation to zero and solving for the atr-exchange rate yields: qua ~ (S - iC)/(C - Co). (11) To facilitate determination of effective radon source strengths, future measurements of indoor-radon concentrations should be accompanied by corresponding outdoor measurements and air-exchange rates, as determined, for example, by a tracer~dilution technique. SUNDRY AND CONCLUSIONS The main purpose of an indoor air quality model is to show the relationships of indoor pollutant concentrations to those outside, to geometric and ventilation characteristics of a structure, and to internal sources and sinks. When the characteristic time for mixing throughout the region of interest is short, compared with the characteristic residence time, the region can be considered as a well-mixed ~compartment.. Even when that criterion is not met, the uncertainty in the values of the ventilation, source, and sink parameter. {with the difficulty of doing otherwise) usually does not justify the development of ~ more sophisticated model. Consequently, the starting point for essentially all indoor air pollution models has so far been a firat~order differential equation representing a pollutant mass balance in a compartment. In many cases, only the steady-~tate solution is needed, and the model reduces to an algebraic equation. In a few case., two or more caspertments (usually coupled by internal ventilation stream have been used to develop model of indoor air pollution. The main difference between various models arises from choosing different source or sink mechanzems. To obtain as much general information as possible, the researcher should include the following in the model: corresponding outside concentrations, appropriate geometric parameters, reasonably accurate

286 ventilation rates, and, if possible, sink and source strengtbs. Such information is required for any model based on a pollutant mass balance. ESTIMATION OF ~" ==SU" TO AIR POLL=ION Today's data on urban air quality come mostly from measurements at fixed monitoring stations. Such data probably show accurately the exposure of a hypothetical person who spends all his time at the station's intake probe. However, people are in constant notion in urban areas, moving front residential areas to places of work to commercial areas, etc. To determine individual human exposure to air pollution accurately, it is necessary to find some means to measure and correlate the movement of individuals in a population and the spatial variation in concentrations of pollutants, whether indoors or outdoors. One way to estimate better the total individual exposure to environmental pollutants is to equip a large number of persons with monitoring instruments and allow them to go about their daily activities in a normal manner. Bowever, no large-scale personal monitoring studies have been done or are in.progress, at least partly because the development of total-exposure monitor ing is still in an early stage. Although no large-scale national program to develop personal monitors has evolved, limited funds from federal agencies have resulted in the development of specific monitors, and priorate companies have developed some instruments that are portable, small, and reasonably priced. 6' '. Other approaches have used theoretical analyses and modern for estimating total exposure. Fugas27 made one of the first attempts to compute total exposure from experimental data; her approach we" intended only as an illustrative example. She obtained measurements of average concentrations of lead, manganese, and sulfur dioxide during the winter of 1972-1973 from official Monitoring stations in the city. The measurements were taken at the breathing zone in several streets during business hours, indoors close to the streets during business hours, and in the countryside. By estimating the time spent by inhabitants of the city in f ive locations--home, work, street 1, street 2, and the countryside--Fugas calculated the Weighted weekly exposure. (WWE) for each of these air pollutants {see Table w-27. An intermediate computation is the Integrated exposure, ~ which is the product of the average concentration and the time during which pollution occurs. To calculate the Was for sulfur dioxide, for example, we note that a person spent an average of 110 h/wk at home, where the average concentration was 89 ug/m3, for an integrated exposure of 9,790 -oh/ for the time spent at home. By adding all the integrated exposure components, Fuga' obtained the total of 16,896 ug-h/~3 for the week. The ~ to sulfur dioxide was then obtained by dividing by the number of hours in a week: 16, 896/168 ~ 101 /m3. 2 ~ ~ Duan23 has modified Fugas's approach by substituting the term Microenvironment types. for the ~locations. used to compute ~WE. In Duan's model, ~ person's integrated exposure over some period (for

287 TABLE: ~JI-2 Example by Fugas Illustrating Computation of Weighted Weekly Exposurea Type of Exposure Sulfur Duration of Dio~de Lead Manganese Exposure, h/wk 1— C Ct C -C-t . Home 110 89 9,790 2.S 275 0.04 4.4 Work 42 8 336 0.3 12.6 0.02 0.84 Street ~ 10 600 6,000 6.0 60 0.80 B.0 Street 2 4 180 720 3.5 14 0.12 0.48 Countryside 2 25 50 0.1 0.2 0.01 0.02 Total 168 - 16,896 - 361.8 ~ 13.74 Weighted -- -- 101 2.2 0.08 weekly exposure aData tram Fugas.27 Values for C (concentration) are expressed in Pg/m3, and values for Ct (integrated exposure) are expressed in ug-h/m .

288 example, a week) is computed as a weighted average of the exposures from various microenvironment types, weighted by the proportion of time spent in each microenvironment type: K jktij where Eij is the integrated exposure of the ith individual during the jth period, cijk is the average concentration in the kth microenvironment type during the jth period, and tick is the activity-pattern coefficient denoting the time the ith individual spent in the kth microenvironment type during the jth period. Duane suggested that a microenvironment type should be defined ~finely. enough to be homogeneous; that is, the Concentration coefficients should not vary appreciably over the individuals. However, the microenvironment types have to be somewhat ~coarse,. so that the analyst will not have too many types to deal with. Some types might be ~rush-hour highway commuting,. daytime urban office with sir- conditzoning, ~ and Weekend daytime outdoors in the park. ~ ~ ~ Moschandreas and Morse7. have suggested an analogous approach for computing air-pollution exposure and have applied it to real data. They introduced the application of Immobility patterns,. which are designed to capture tithe daily Decrements of individuals as they moire to and from work, from twine to points of amusements, adventure, business, and so on.~7' By examining the literature on activity patterns and Time budgets,. they arrived at the estimated time that all persons-- all races, ages, socioeconomic groups, workers, students, etc.--spend in various Environmental modes. {which are analogous to the Microenvironment types. of Duan23 and the ~locations. of Fugal27~. The population spends 72.8% of its time inside homes, but the figure is different for different population subgroups {workers, children, the elderly, etc.~. Because of the importance of considering the mobility patterns of population subgroups, Mbschandress and Morsel examined U.S. census data to define the percentage of each of six subgroups in the total population: housewives, office personnel, industrial workers, outdoor workers, elderly and infirm people, and students. However, students are not considered in the overall model, because few studies have been made of their mobility patterns. The model can be viewed as a three-dimensional drawing in which persons cove through time occupying different environmental models, thereby exposing themselves to the particular concentration that is associated with each environmental mode and period. On the bests of data on typical diurnal ozone concentrations for three environmental nodes (residential, office, and indoors) in the Boston area and estimates of the percentage of the population in each environmental He as ~ function of time, Moschandreas and Morse7. estimated that 21. of the population is exposed to ozone at 80 ppb or more. At the time, the federal NAAQS for ozone (l-in average) was 80 ppb, but it has since been raised. Among individual population subgroups, Moschandreas and Morse estimated that

289 outdoor workers are exposed to high ozone concentrations (over 80 ppb) for about 6 h and industrial workers for the 1-h period between 4:00 and ~ :00 p.~. If we take a .snapshot. of the estimated population exposures at some particular point in time, such as 3:00 p.m. {that is, 1500, see Figure VI-4), some 25% of the population (3.9% in outdoor activities and 23.39e in transit) is exposed to ozone at 80 ppb or more. Although these are excellent recent examples of ways to estimate population exposure to air pollution, they have several limitations. Health-related air quality standards usually do not have weekly averaging times. Thus, the value of the WEE calculated by Fugas27 cannot be compared directly with existing air quality standards. Duan's formulations was intended to be more flexible, allowing any averaging period (for example, a week or a month) to be used. The computation of exposure by Moschandreas and Worsen did not cause difficulties with averaging periods, because ozone has a 1-h NAAQS, and the population i s assumed to spend 1-h increments (or multiples of 1 h ), in each environmental ~ode. Each of the above approaches for calculating exposure gives the population's average exposure over some specified period, and a problem arises from the emphasis on the arithmetic mean. In any given period, come people will be involved in combinations of activities that can result in exposures much higher and much lower than the mean. In addition, the time spent in each activity varies from day to day and from person to person. Thus, the mean value for exposure concentration is not adequate to characterize the highest concentrations to which members of the population are exposed, and the var lance of exposures also must be considered. Ideally, there would be an effective technique for determining the entire frequency distribution of exposures of the population to air pollution. As discussed above, two approaches are used for estimating the frequency distribution of human air-pollution exposures: modeling, which relates the activities of persons as a function of time and the concentrations to which they are exposed; and field studies, which use personal monitors to cover a large enough population sample (or a stratified sample) to represent statistically the distribution of exposures. Although a large-scale field study of exposure has not been completed, efforts are under way to use computer simulation to model human exposures to air pollution. Ott " " has developed a computer-simulation model of human exposure to air pollution that includes the movements of individual people in a metropolitan area as a series of transitions from one .microenvironment. to another. Ott's computer program, Postulation of Human Air Pollution Exposures. (SHAPE), uses probability distributions of the time that people spend in each microenvironment--distributions derived from studier of human activity patterns. In each microenvironment, the concentration to which an individual is exposed is treated stochastically, with distributional models that are based on f told studies of air-pollutant concentrations reported in the research literature . In the simulation, the computer keeps track of the exposure received by each person as he or she moves forward in time and occupier successive microen~rironments.

290 8 - _ _ ._ C _ "D O o-O Me, ~ 0—~5 ~ _ =0 c0 c0 4_ ~ ~ ~ ~ ~ NlYiiNlllYij~ L.~--~\ , ~ ~x ~ ~ · · I I I · · I _ is a o _ qdd uol4~'u.:Uoo- ~4 a; o 0 0 o ," V V ~ 3 C) o o o :- o SAC o o ~ so O D X ~ - o Cat ~ en o v 0. C: 0 O ~ o So o V o o 8 so 3 ^ o . 0 , to ~ ·- X— O 1 "C: O o - lr~ S a ~o 0 o 0 v 0 o ~: s ~

291 ott77 definer an ~exposure. of person i to concentration c statistically as the joint occurrence of two independent events: person i is present in microenvironment j, and the concentration Ctj} c occurs in microenvironment j . C (; ~ denotes the probability distribution of the concentrations associated with the microenvironment and is based on f ield monitoring data . Then, the integrated exposure Hi of person i is computed as the sum of the products of the concentrations encountered in each microenvironment and the time spent there: £ = _ . ~ Cjtij, j = 1 where c. is the concentration associated with microenvironment j, ti is {he time spent by person i in microenvironment j, and J is the total number of microenvironments occupied by person i in some period of interest. Conceptually, this model can be represented by a three-dimensional array (see Figure VI- that is similar to a three-dimensional space developed by Moschandreas and Morse. 70 However, the computer simulation follows one person at a time through his or her daily activities, and the resulting distribution of exposures is obtained by considering those of all persons in the simulation. The times ti spent in the microenvironments are variable and do not need lo be integer multiples of 1 h. REFERENCES 2. 3 1. Amass, C. E. Passive membrane-limited dos~meters using specific ion electrode analysis, pp. 437-460. In D. T. Mage, and L. Wallace, Eds. Proceedings of the Symposium on the Development and Usage of Personal Monitors for Exposure and Health Effect Studies . U. S . Environmental Protection Agency {Environmental Monitoring and Support Laboratory, and Health Effects Research Laboratory) Report No. EPA-600/9-79-032. Washington, D.C.: U.S. Government Pr inting Of f ice, 1979 . American Conference of Governmental Industrial Hygienists. Air Sampling Instruments for Evaluation of Atmospheric Contaminants. 5th ed. Cincinnati: American Conference of Governmental Industrial Hygienists, 1978. Azar, A., R. D. Snee, and A. ~abibi. Relationship of community level" of air lead and indices of lead absorption, pp. 581-594. In D. Barth, A. Berlin, R. Engel, P. Recht, and J. Smeets, EdS. Proceedings. International Symposium. Environmental Health Aspects of Lead, Amsterdam, October 2-6, 1972. Bamberger, R. L., G. G. Esposito, B. W. Jacobs, G. E. Podolak, and J. F. Mazur. A new personal sampler for organic vapors. Am. Ind. Hyg . Assoc. J . 39: 701-708, 1978 . 5. Berlandi, F. J., G. R. Dulude, R. M. Griffin, and E. R. Sink. Electrochemical air lead analysis for personal environmental

292 ~\Ro~ 1 2 9 8 6 w at, 5 2 '1 ~ \ . ~ ~ MY W.' i i \\- \ !n , \ \ ~ Hi\ ~.~ \\ hi\ , 1.~ FIGURE VI-5 Graphic representation of person-environment-time Stray in computer simulation model of exposure to air pollution suggested by Ott.

293 surveys, pp. 161-172. In D. T. Hage, and L. Wallace, Ede. Proceedings of the SymyoBiu~ on the Development and Usage of Personal Monitors for Exposure and Bealth Effect Studies. U.S. Environmental Protection Agency (Environmental Monitoring and Support Laboratory, and Health Effects Research Laboratory} Report No. EPA-600/9-79-032. Washington, D.C.: U.S. Government Printing Office, 1979. 6. Binder, R. E., C. A. Mitchell, B. R. Bosein, and A. Bouhuys. Importance of the indoor environment in air pollution exposure. Arch. Environ. Health 31:277-279, 1976. 7. - _ Bridge, D. P., and M. corn. Contribution to the assessment of exposure of non-emokers to air pollution from cigarette and cigar smoke in occupied spaces. Environ. Res. 5:192-209, 1972. 8. Brief, R. S. Simple way to determine air contaminants. Air Eng. 2:39-41, 1960. 9. Brooks, J. J. , D. S. West, D. J. David, and J. D. Mulik. A combination sorbent system for broad range organic sampling in air, pp. 383-412. In D. T. Mage, and L. Wallace, Eds. Proceedings of the Symposium on the Development and Usage of Personal Monitors for Exposure and Health Effect Studies. U.S. Environmental Protection Agency (Environmental Monitoring and Support Laboratory, and Health Effects Research Laboratory) Report No. EPA-600/9-79-032. Washington, D.C.: U.S. Government Printing Office, 1979. 10. Burgess, W., L. DiBerardinis, and F. E. Speizer. Exposure to automobile exhaust. III. An environmental assessment. Arch. Environ. Health 26:325-329, 1973. Calder, K. L. A Numerical Analysis of the Protection Afforded by Buildings against BW Aerosol Attack. Biological Warfare I`aboratories Technical Study No. 2. Fort Detrick, Md.: Office of the Deputy Commander for Scientif ic Activities, 1957. 24 pp. Calder, X. Le A Simple Mathematical Model for the Penetration of Forest Canopy by Aerosols. Biological Laboratories Technical Study No. 37. Fort Detrick, Md.: Office of the Scientific Director, 1961. 13. Campbell, J. E. The development of ~ passive dosimeter for airborne aniline vapors. Am. Ind. Hyg. Assoc. J. 41 :180-184, 1980 . 14. Carnow, B. W., M. B. Lepper' R. B. Shekelle, and J. Stamler. Chicago air pollution study. SO2 levels and acute illness in patients with chronic bronchopulm~nery disease. Arch. Environ. Health 18: 768-776, 1969. Constance, J. D. Mixing factor is guide to ventilation. Power 114 (2) :56-57, 1970. 16. Cortese, A. D., and J. D: Spengler. Ability of fixed monitoring stations to represent carbon monoxide exposure. J. Air Pollut. Control Assoc. 26:1144-1150, 1976. 17. Cote, W. A., and K. Holcombe. The influence of sir conditioning systems on indoor pollutant levels, pp. I-20, Session 5, Paper 3. In J. W. White, Ed. Proceedings of the First Conference on Natural Gas Research and Technology. Chicago: Institute of Gas Technology, 1971.

294 18. Dockery, D. W. and J. D. Spengler. Personal exposure to respirable particulates and sulfated: Measurement and prediction, pp. 111-129 (includes discussion'. In D. T. Mage and I`. Wallace, ads. Proceedings of the Symposium on the Development and Usage of Personal Monitors for Exposure and Health Effect Studies. U.S. En~riroDenta1 Protection Agency {33nvironmente1 Monitoring and Support Laboratory, and health Effects Research Laboratory Report No. EPA-600/9-79-032. Washington, D.C.: O.S. Go~rerr~ent Printing Office, 1979. 1 9. Dockery, D. W., and J. D. Spengler . Personal exposure to respirable particulates and sulfates versus Client measurements. Paper No. 77-44.6 presented at the 70th Annual Meeting of the Air Pollution Control Association, June 20-24, 1977, Toronto, Canada. 20. Dockery, D. W., J. D. Spengler, M. P. Reed, and J. Ware. Relationships Among Personal, Indoor and Outdoor NO2 Measurements. Paper presented at the 73rd Annual Meeting of the Air Pollution Control Association, Montreal, Canada, June 23-27, 1980. 21. Dorman, R. G. Filtration, pp. 195-222. In C. N. Davies, Ed. Aerosol Science. New York: Academic Press, Inc., 1966. 22. Drivas, P. J., P. G. Sim~onds, and F. H. Shair. Experimentation characterization of ventilation systems in buildings. Environ. Sci. Technol. 6:609-614, 1972. 23. Duan, N. Microenvironment Types: A Model for Buman Exposures to Air Pollution. S-MS Technical Report. Stanford, Cal.: Stanford University, Department of Statistics, 1981. (in preparation) 24. Environmental Measurements Incorporated. Radon Sampling, pp. 6-8. Brochure No. 2, Annapolis, Md.: Environmental Measurements Incorporated, 1978. 25. Fleischer, R. L., W. A. Giard, A. Mogro-Campero, L. G. Turner, H. W. Alter, and J. E. Gingrich. Dosimetry of environmental radon: Methods and theory for low~se, integrated measurements. Health Physics 39: 957-962, 1980. 26. Fuchs, N. A. The Mechanics of Are-: Bale. New York: Pergamon Press, 1964;,408 pp. 27. Fugas, M. Assessment of total exposure to an air pollutant. Paper no. 38.S in Proceedings of the International conference on Environmental Sensing and Assesmuent, vol. 2, September 14-19, 1975, Las Vegas, Nevada. Institute of Electrical and Electronics Engineers Publication No. t75-CE 1004-1 INCEST. 28. Fug ad, M., B. Wilder, R. Paukovic, J. Rook, and D. Steiner-'kreb. Concentration levels and particle size distribution of lead in the air of an urban and an industrial area 88 ~ basis for the calculation of population exposure. pp. 961-968. In D. Barth, A. Berlin, R. Engel, P. Recht, and J. Smeets, Eds. Proceedings. International Symposium on Environmental Bealth Aspects of Lead' Amsterdam, October 2-6, 1972. Luxembourg: Commission of the European Communities, Center for Information and Documentation, 1973. 29. George, A. C. A passive environmental radon-monitor. pp. 25-30. In A. J. Breslin, Ed. Radon Workshop, February 1977. Energy Research and Development Administration, Health and Safety

295 Laboratory Report HASL-325 (CONF. 770231). New York: U.S. Energy Research and Development Administration, 1977 . 30. Georgii, H. -W. Investigation of the air exchange between rooms and the air outside. Arch. Meteorol. Geophys. Bioklimatol. (Ser. B) 5 :191-214, 1954. ~ in German; English sunonary) 31. Gilath, C. Ventilation and air pollution studies using radioactive tracers. A critical review. Int. J. Appl. Radiat. Isot. 28 :847-854, 1917. 32. Gillespie, J. C., and L. B. Daniel. A new sampling tool for monitoring exposures to toxic gases and vapors, pp. 479-491. In D. T. Mage and L. Wallace, Eds. Proceedings of the Symposium on the Development and Usage of Personal Monitor s for Exposure and Health Effect Studies. U. S. Environmental Protection Agency (Environmental Monitoring and Support Laboratory, and Health Effects Research Laboratory) Report No. EPA 600/9-79-032. Washington, D.C.: U.S. Government Printing Office, 1979. 33. Gold, A., T. J. Smith, C. E. Dube, and J. J. Cafarella. Solid sorbent for acrolein and formaldehyde in air, pp. 425-435. In D. T. page and L. Wallace, Eds. Proceedings of the Symposium on the Development and Usage of Personal Monitors for Exposure and Health Effect Studies. U. S. Environmental Protection Agency (Environmental Monitoring and Support Laboratory, and Health Ef feats Research Laboratory) Report No. EPA-600/9-79-032. Washington, D.C.: U.S. Government Printing Office, 1979. 34. Goldmuntz, L. A. Extent to Which EPA Monitoring Station Measurements Reflect Human Exposure to Carbon Monoxide. National Science Foundation Grant No. STP 75-21384. Washington, D.C.: Economics & Science Planning, Inc., 1976. 35. Gosselink, D. W., D. L. Braun, H. E. Mullins, and S. T. Rodriguez. A new personal organic vapor monitor with in situ sample elusion, pp. 365-382. In D. T. Mage and L. Wallace, Eds. Proceedings of the Symposium on the Development and Usage of Personal Monitors for Exposure and Health Effect Studies. U.S. Environmental Protection Agency (Environmental Monitoring and Support Laboratory, and Health Effects Research Laboratory) Report No. EPA-600/9-79-032. Washington, D.C.: O.S. Government Printing Of f ice, 1979. 36. Green, H. L., and W. R. Lane. Particulate Clouds: DuStS, Smokes and Mists. 2nd ed. Belfast: Universities Press, 1964. 479 pp. Hardy, J. K., P. K. Dasgupta, K. D. Reiazoer, and P. W. West. A personal chlorine monitor utilizing permeation sampling. Environ. Sci. Technol. 13:1090-1093, 1979. 38. Harrison, O. Development Strategy for Pollutant Dosimetry. U.S. Environmental Protection Agency Report No. 600/2-76-034. Research Triangle Park, N.C.: U.S. Environmental Protection Agency, 1976. 39. Harrje, D. T., C. M. Bunt, S. J. Tredo, and Ne J. Malik. Automated Instrumentation for an Infiltration Measurement in Buildings. Center for Environmental Studies Report No. 13. Princeton, N. J.: Princeton University, 1975. Hitchin, E. R., and C. B. Wilson. A review of experimental techniques for the investigation of natural ventilation in buildings. Build. Sci. 2:59-82, 1967. 40.

296 41. Bollingdele-Saith, P. As Authors' reply. Ann. Occup. Bye. 22:86, 1979. 42 . Bollowell, C. D., J. V. Berk , and G . W. Traynor . Impact of reduced infiltration and ventilation on indoor air quality. ASEIRAE J. 21~7) :49-53, 1979. 4 3 0 Boriuchi, B. Don-222 and its daughters in buildings at Uranium City, Saskatchewan. In J. E. Turner, C. F. Boloway, and A. S. Loebl, Eds. Radon Workshop on Dosimetry for Radon and Radon Daughters. Oak Ridge National Laboratories Report No. 0BNL-4348. 44. Bunt, C. H. The Control of Airborne Dust in Post Office Facilities. Final Report. AS Project 4213417. Bureau of Research and Engineering. Washington, D.C.: Post Office Department, 1970. 45. Bunt, C. M., and D. M. Lurch. Air infiltration measurements in a four-bedroom townhouse using sulfur hexafluoride as a tracer gas. AS~RAE Trans. 81 {Pt. 13:186-201, 1975. (includes discussion) 46. Bunt, C. H. , B. C. Cadoff, and F. J. Powell. Indoor Air Pollution Status Report. National Bureau of Standards Report 10 591. Washington, D.C.: U.S. Department of Commerce, National Bureau of Standards, Building Research Division, 1971. 47. International Energy Agency. Methods of measuring infiltration and building tightness, pp. 22-48. In Draft Program Plan. Air Inf filtration in Buildings . U. S . Department of Energy Report No. DOE/CS-0099-D . Washington , D. C .: U. S . Department of Energy , Division of Buildings and Community Systems, 1979. 48. Intersociety Committee. Methods of Air Sampling and Analysis, pp. 117-129. Washington, D.C.: American Public Health Association, 1972. 49. Jackson, D. L., and V. A. Newill. The strengths and weaknesses of population studies in assessing environmental health effects, pp. 161-175. In Proceedings. International Symposium. Recent Advances in the Assessment of the Bealth Effects of Environmental Pollution, Paris, (France), June 24-28, 1974. Vol. I. I`uxembourg: Commission of the European Communities, Directorate General Scientific and Technical Information and Information Management, 1975. 50. Kahn, A., R. B. Rutledge, G. L. Davis, J. A. Altes, G. E. Gantner, C . A. Thornton ~ and N . D . Wallace . Carboxyhemoglobi n sources in the metropolitan St. Louis population. Arch. Environ. Bealth 29 s 127-135, 1974. 51. Kusuda, T. Control of ventilation to conserve energy while maintaining acceptable indoor air quality. ASERAE Trans. 82(Pt. 1} :3~169-1181, 1976. Kusuda, T., C. H. Bunt, and P. E. McNall. Radioactivity (radon and daughter products) as ~ potential factor in building ventilation. AS=AE J. 21~7) :30-34, 1979. Lawrence Berkeley Laboratory, Enviror~ental Instrumentation Group. Gases, pp. l-S1. In Instrumentation for Environmental Monitoring. Vol. I. Air. Lawrence Berkeley Laboratory Report I,BL-1. Berkeley, Cal e Lawrence Berkeley Laboratory, 1976. 54. Lidwell, O. M., and J. E. Lovelock. Some methods of measuring ventilation. J. Bye. 44: 326-332, 1946.

297 Linct1, A. L. Evaluation of Ambient Air Quality by Personnel Monitoring . Cleveland, Ohio : CRC Press , Inc., 1974 . 226 pp. 56. L~ppmann, M. ~Respirable~ dust sampling, pp. G-1-—i-23. In Air Sampling Instruments for Evaluation of Atmospheric Contaminants. 5th ed. Cincinnati: American Conterence of Governmental Industrial Hyg ienists, 1978. 5/. Lou, B. Y. H., and K .W. Lee. Efficiency of n~e~rane and nucleopore ~ alters for submicrometer aerosols . Environ. Scz . Technol. lo :345-350, 1976. 58. Loo, B. W., R. S. Adachi, C. P. CorK, F. S. Goulding, J. M. Jaklev~c, D. A. Landis, and W. L. Searles . A Second Generation Dichotomous Sampler for Large-Scale Monitoring of Airborne Particulate Matter. Lawrence Berkeley Laboratory Report No. LBL-8725. Berkeley, Cal.: Lawrence Berkeley Laboratory, 1979. 18 PP 5 9. Lucas, H. F. Alpha scintillation radon counting, pp. 69-156. In E. D. Harward, Ed. Workshop on Methods for Measuring Radiation in and around uranium Mills, May 23-26, Albuquerque, New Mexico. Program Report 3-9. Washington, D.C.: Atomic Industrial Forum, 1977. 60. Mage, D. T., and L. Wallace, Ems. Proceedings of the Symposium on the Development and Usage of Personal Monitors for Exposure and Health Effect Studies, January 22-24, 1979, Chapel Hill, N.C. U.S. Environmental Protection Agency (Environmental Monitoring and Support Laboratory, and Health Effects Research Laboratory) Report No. EPA-600/9-79-032. Washington, D.C.: U.S. Government Printing Office, 1979. 525 pp. 61. Marple, V. A. Simiation of respirable penetration characteristics by inertial impactions J. Aerosol Sci. 9:125-134, 1978. Mazur , J . F., R. L. Bamberger , G. E. Podolak , and G. G. Esposito . Development and evaluation of an ammonia dosimeter. Am. Inct. Byg. Assoc. J. 39 :149 - 753, 1978. McDermott, D. L., K. D. Rei~zner, and P. W. West. Development of long-term sulf ur dioxide monitor using permeation sampling . Environ. Sci . 1`echnol. 13 :1087-lU9U, lo I9. 64. Miller, M. E. Laboratory Fume hood s and the Exhaust Gas Re-entry Problem. Pasadena: Calitornza Institute of Technology, Department ot Chemical Eng~neer~nq, Master of Science thesis, 1979. 6 5. Molly, G. H. A Theory of Chemical Attack ot Tanks and Enclosed Fortit Locations . Report for the Chemical Corps. Edgewise Arsenal, Md.: Army Chemical Center, Chemical and Radiological Laborator yes, 19 5 3 . 6 6. Hilly, G. 11., and S . D. Thayer. Techniques tor Dosage Prediction. Vol. I. Development ot Techniques. Travelers Research Center Report No. T=247, 1967. 6 7. Morgan, M. G., and S. C. Morris. Inc~iv~dua1 Air Pollution Monitors: An Assessment of National Research Needs. Brookhaven National Laboratory Report No. 50482. Upton, N. Y.: Brooknaven National Laboratory, 1976. 35 pp. 66. Morgan, M. G., and S. C. Morris. Individual Air Pollution Monitors. a. Examina~c~on ot Some Nonoccupational Research and

298 Regulatory Uses and Needs. Brookhaven National Laboratory Report No. 50637. Upton, N. Y.: Brookhaven National Laboratory, ]977 . 11 pp. 6 9 . - schandrea=, D. J ., Ed. Indoor Air Pollution in the Residential Environment. volume II. Field Monitoring Protocol, Indoor Episodic Pollutant Release Exper iments and Numer Cal Analyses . U.S. Environmental Protection Agency (Environmental Monitoring and Support Laboratory) Report Ho. EPA-600/7-78-229b. Research Triangle Park, N.C.: U.S. Environmental Protection Agency, 19780 70. - schandreas, D. J., and S. S. Morse. Exposure estimation and mobility patterns. Paper No. 79-14.4 presented at the 72nd Annual Meeting of the Air Pollution Control Association, Cincinnati' Ohio, June 24-29, 1979. 71. Moachandrea=, D. J., D. J. Pelton, D. J. Sibbett, J. W. C. Stark, and J. E. McFadden. Comparison of Indoor~Outdoor Concentrations of Atmospheric Pollutants. Field Monitoring Protocol. Scientific Report RP1309. GEOMET Report No. E-721. Gaithersburg, Md.: GEOMET, Inc., July 1978. 103 pp. 72 . - schandreas , D. J ., and J . W. C . Stark . The GEOMET Indoor~Outdoor Air Pollution Model. U.S. Environmental Protection Agency Report No. EPA-600/7-78-106. Research Triangle Park , N.C.: U . S . Environmental Protection Agency, Environmental Monitor ing and Suppor t Laboratory, 1978 . 75 pp. 73. - achandreas, D. J ., J. W. C. Stark, J. E. McFadden, and S . S . Morse. Indoor Air Pollution in the Residential Environment. Final Report for U. S . Environmental Protection Agency, Environmental Research Center and U.S. Department of Housing and Urban Development, Off ice of Policy Development and Research under EPA Contract - . 68-02-2294, GEOMET Report No. EEt~688e Gaithersburg, Hd.: GEOME!, Inc., 1978. 74. National Research Council, Committee on Biologic Effects of Atmospher ic Pollutants. I`ead . Airborne Lead in Perspective, p . 216. Washington, D.C.: National Academy of Sciences, 1972. 7S. NeLms, L. B., R. D. Rei~zner , and P. W. West. Personal vinyl chloride monitoring device with permeation technique for sampling . Anal. Chem. 49: 994-998, 1977. 76. Ott, W. R. A Field Quality Assurance Evaluation of Two Personal Monitors for Carbon Monoxide. Stanford, Cal.: Stanford University, Department of Statistical Unpublished report) 77. Ott, W. R. Concepts of Burn Exposure to Environmental Pollution. SIMS Technical Report No. 32. Stanford, Cal.: Stanford University' Department of Statistics, 1980. 78. Ott, W. R. Development of Activity Pattern Models for Human Exposure Monitoring. Innovative Research Program Proposal to U.S. Environmental Protection Agency, July 1, 1979. 79. Ott, W. R., and D. T. Mage. A method for simulating the true human exposure of critical population groups to air pollutants, pp. 2097-2107. In Proceedings. International Symposium. Recent Advances in the Assessment of the Health Effects of Environmental Pollution. Paris, {France), June 24-28, 1974. Vol. TV. Luxembourg: Commission of the European Communities, Directorate General Scientific and Technical Information and Information Management, 1975.

299 80. Palmes, E. D. Effect of wind on diffusion samplers. Ann. Hyg . 22: 85, 1979. 81. PaLmes, E. D. Personal Samplers for CO, NO and NO2 in Air. U.S. Bureau of Mines Open File Report 92-77 . Washington, D.C.: U. S . Department of Interior, Bureau of Mines, 1977. 22 pp. Palme s, E. D . Personal sampler for measurement of ambient levels of NO2, pp. 57-64. In D. T. Mage and L. Wallace, Eds. Proceedings of the Symposium on the Development and Usage of Personal Monitors for Exposure and Health Effect Studies. U.S. Environmental Protection Agency (Environmenta1 Monitoring and Support Laboratory, and Health Effects Research Laboratory) Report No. EPA-600/9-79-032. hashingon, D. C.: U. S . Government Printing Off ice, 1979. 83. Palmes, E. D., and A. F. Gunnison. Personal monitoring device for gaseous contaminants. Am. Ind. Hyg. Assoc. J. 34: 78-81, 1973. 8 4. Palsnes, E. D . A. F. Gunnison, J. Di~5attio, and C . Tomczyk . Personal sampler for nitrogen dioxide. Am. Ind. Hyg. Assoc. J. 3 7: 570-577, 1976. Palmes, E. D., and C. Tomazyk. Personal sampler for NOX . Am. Ind. Byg. Assoc. J. 40: 588-591, 1979. 8 6. Palmes, E. D., C. Tomczyk , and J. DiMattio. Average NO2 concentrations in dwellings with gas or electric stoves. Atmos. Environ. 11:869-872, 1977. 8 7. Pich, J. Theory of aerosol f iltration by f ibrous and membrane f ilters, pp. 223-285. In C. N. Dairies, Ed. Aerosol Science. New York : Academic Press, Inc., 1966. 88. Reiszner, K. D., and P. W. West. Collection and determination of sulfur dioxide incorporating permeation and West-Gaeke procedure. Environ. Sci. Technol. 7: 526-532, 1973. 8 9. Repace, J. L., and A. H. L~wrey. Indoor air pollution, tobacco smoke, and public health. Science 208: 464-472, 1980. 90. Sabersky, R. H., D. A. Sine~r,a, and F. B. Shair. Concentrations, decay rates, and removal of ozone and their relation to establishing clean indoor air. Environ. Sci. Technol. 7: 347-353, 1973. 91. Saltzman, B. E. Direct reading calorimetric indicators, pp. S-1--S-39. In Air Sampling Instruments for Evaluation of Atmospheric Contaminants. 5th ed. Cincinnati: American Conference of Governmental Industr ial Hyg ienists, 19 78 . 92. Sasaki, E., J. Schienle, T. Shibata, and F. H. Shair. Influence of meteorology upon the reentry of fume hood exhaust. Pasadena: California Institute of Technology, Department of Chemical Eng ineering, 1981 . (unpublished) 9 3 . schuette, F. J. Plastic bags for collection of gas samples. Atmos. Environ . 1: 515-519, 1967. 9 4 . Shair, F. H. Relating Indoor Pollutant Concentrations of Ozone and Sulfur Dioxide to Those Outside: A. Determination of the Rate Constants Associated with the Beterogeneous Lc~s';es of Ozone and Sulfur Dioxide inside Laboratory-Office Buildings, B. Economic Reduction of Indoor Concentrations of Smog Pollutants through Selective Filtration of the Make-up Airstream. Research Project

300 142 Financial Report to ASEIItAE. A - erican Society of Heating, Refrigerating and Air~Conditioning Engineers Technical Paper No. 2618. Chicago: American Society of Beating, Refrigerating and Air~Conditioning Engineers, Inc., 1979. 16 pp. plus appendices. ASERAE Trans. (in press, 1981) 95. Shair, F. B., and K. L. Heitner. A theoretical model for relating indoor pollutant concentrations to those outside. Environ. Sci. Technol. 8 :444-451, 1974. 96. Shair , F. R., D. W. Wolbrink, L. O. Bowen, C. E. Neelley, and X. E. Sampeel. Influence of mechanical ventilation on moisture content of bathroom air. ASHRAE J. 21~7) :54-60, 1979. 97. Shaw, R. W. and R. K. Stevens. The tandem filter package, pp. 131-143. In D. T. Hage and L. Wallace, Eds. Proceedings of the S;y~posium on the Development and Usage of Personal Monitors for Exposure and Bealth Effect Studies. U.S. Environmental Protection Agency (Environmental Monitoring and Support Laboratory, and Health Ef fects Research Laboratory) Report too. EPA-600/9-79-032. Washington, D.C.: U.S. Government Printing Office, 1979. 98. Spitz, B. B., and M. E. Wrenn. The Diurnal Variation of the Radon-222 Concentrations in Residential Structures in Grand Junction, Colorado. Second Workshop on the Natural Radiation Envirorment, 1974. 99. Stewart, R. D., E. D. Baretta, L. R. Platte, E. B. Stewart, J. B. Kalbfleisch, B. van Yserloo, and A. R. Rum. Carboxyhe~lobin levels in American blood donors. J. Am. Hed. Assoc. 229:1187-1195, 1974. 100. Stewart, R. D., C. L. Bake, A. flu, T. A. Stewart, and J. B. Kalbfleisch. Carboxyhemoglobin trend in Chicago blood donors, 1970-1974. Arch. Environ. Bealth 31: 280-286, 1976. 101. Turk, A. Measurements of odorous vapors in test chambers: Theoretical. ASERAE J. 5~10~:55-58, 1963. 102. Turner, W. A., J. D. Spengler, D. W. Dockery, and S.D. Colane . Design and performance of ~ reliable personal monitoring Hem for Despicable particulates, pp. 101-109. In D. T. Mage and I. Wallace, Eds. Proceedings of the Symposium on the Development and Usage of Personal Monitors for Exposure and Health Ef feet Studies. U.S. Environmental Protection Agency (Environmenta1 Monitoring and Support Laboratory, and Bealth Effects Resee -'ch Laboratory) Report No. EPA 600/9-79-032. Washington, D.C.: U. S. Government Printing Off ice, 1979. 103. U.S. Environmenta1 Protection Agen=, Environmental Monitoring Systems Laboratory. List of Designated Reference and E uivalent Methods. Research Tr tangle Park: U. S. Environmental Protection AgenQ, February 15, 19 80 . 22 pp. 104. Wallace, LO Personal air quality monitors: Uses in studies of human e" osure, pp. 7-18. In D. T. Mage and L. Wallace, Ede. Proceedings of the Symposium on the Development and Usage of Personal Monitors for Exposure and Bealth Effect Studies. U.S. Environmental Protection Agency {Environmental Monitoring and Support Laboratory, and Bealth Effects Research Laboratory) Report No. EPA-600/9-79-032. Washington, D.C.: U.S. Government Printing Office, 1979.

301 105. West, P. W., and R. D. Reiaxner. Field tests of a permeation-type personal monitor for vinyl chloride. Ant. Ind. Byg. Assoc. J. 39: 645-650, 1978. 106. West, P. W., and R. D. Reiazner. Personal Monitor for Nitroqan Dioxide . U. S . Environmental Protection Agency Report ate. EPA-600/2-78-001. Research Triangle Park, N.C.: U.S. Environmental Protection Agency, 1978 . 33 pp. 107. Went, P. W., and IC. D. Reiazner. Personal monitoring by means of gas permeation, pp. 461-471 . In D. T. Mage, and L. Wallace, ads . Proceeding. of the Symposium on the Development and Deage of Personal Monitors for Exposure and Health Effect Studies. U.S. Environmental Protection Agency (Environmental Monitoring and Support Laboratory, and Health Effects Research Laboratory} Report No. EPA 600/9-79-032 . Washington, D.C.: U.S. Government Printing Office, 1979. 108. Willeke, R. Selection and design of an aerosol sampler simulating Despicable penetration. Am. Ind . Byg . Assoc. J . 39: 317-321, 1978 . Woods, J. E. Ventilation, health and energy consumption: A states report. ASHRAE J. 21(1):23-27, 1979. Woods, J. E., R. G. Nevins, and E. L. Besch. Analysis of thermal and ventilation requirements for laboratory animal cage environments. ASERAE: Trans. 81tPt. 1~:45-66, 1975. Wright, G. R., S. Jewcsyk , J . Onrot, P. Tomlinson , and R. ~ . Shephard. Carbon monoxide in the urban atmosphere. Hazards to the pedestrian and the street~worker. Arch. Environ. Health 30: 123-129, 1975. Zukoski, E. E., and T. Kubota. Two~layer modeling of smoke movement in building f ires . Fire Mater . 4 :17-27, 1980 . .

Next: VII. Health Effects of Indoor Pollution »
Indoor Pollutants Get This Book
×
Buy Paperback | $140.00
MyNAP members save 10% online.
Login or Register to save!
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!