Objectives

With this Web Project, we hope we have succeeded in providing:

1. Background information on the emerging concern over VOCs in drinking water as a contributor to indoor air quality problems.
2. Information on tools that have been developed to predict indoor VOC concentrations..
3. Technologies and methodologies that can be employed to mitigate indoor air pollution resulting from VOCs in drinking water.
4. An experts perspective on VOC indoor air pollution.
5. A few fun animations relating to the subject.

Table of Contents

Background	Mathematical Model
Sources of VOCs in Drinking Water	Direct Measurement of VOCs in Indoor Air
Current Regulatory Environment	Victim of VOC Exposure!!!
Treatment Technologies	Comments from Dr. John C. Little on Indoor Air Pollution
Indoor Mass Balance Model	References

Background

Concern and consequently regulation, of potentially harmful substances in drinking water supplies have historically focused on the ingestion pathway. However, scientists, engineers, and regulatory agencies are beginning to understand that vaporization and subsequent inhalation of organic contaminants in drinking water can also present significant health risks. The Total Exposure Methodology (TEAM) of the United States Environmental Protection Agency (USEPA) studied personal exposures of 600 residents of seven U.S. cites during the period of 1981 through 1984. Indoor exposure to chlorinated hydrocarbons, toluene, benzene, and eight other volatile organic compounds (VOCs) exceeded outdoor exposure by factors of two to five. A TEAM study completed in New Jersey found that the median drinking water chloroform concentration was 67 mg/L and the median residential indoor air choloform concentration was 3.2 mg/m³. Although the study did not conclude chloroform in air resulted only from drinking water, the authors hypothesized that hot showers were a source. Recent studies have also shown that 50% or more of common contaminants of water supplies such as chloroform and trichloroethylene will volatilize and constitute inhalation exposures at the point of water use as well as to inhabitants elsewhere as gas phase contaminants are transported throughout the home.[1] McKone has shown that human exposure attributable to inhaling air contaminated with VOCs from drinking water may be as much a six times higher than that incurred by consuming 2 liters of the same drinking water.[2]

Sources of VOCs in Drinking Water

VOCs can enter the drinking water supply system via two routes:

1. A water source can be contaminated from spills and industrial activities. Few traditional water treatment systems are designed to remove synthetic organic chemicals from the water supply.
2. Naturally occurring organic material (NOM) can react with chlorine during the disinfection process to create chlorinated chemicals commonly referred to as disinfection byproducts (DBP). Because chloroform and trihalomethanes are suspected carcinogens, these substances are DBPs of concern.

Regulatory Environment

The Safe Drinking Water Act (SDWA) of 1974 required the administrator of the USEPA to establish minimum national standards for controlling the presence of contaminants in drinking water. Congress amended the SDWA in 1986. The 1986 amendments require the USEPA to set maximum contaminant levels (MCLs) and maximum contaminant level goals (MCLGLs). If it is not economically or technically feasible to monitor a contaminant level, the USEPA can require a treatment technology in place of an MCL. Water supplies subject to the SDWA must meet MCLs; MCLGs are goals only; not mandates. Water treatment operations subject to the SDWA must implement treatment and monitoring systems insure SDWA requirements are being met. Current regulatory procedures establish MCLs, MCLGs, and mandated treatment technologies based on ingestion of 2 liters per day.[3] Dosages obtained through inhalation are not currently considered when establishing MCLs, MCLGs, and mandated treatment technologies.

Treatment Technologies

The SDWA amendments declare granular activated carbon (GAC) as the "best available control technology" (BAT) for removing synthetic organic chemicals (SOCs) from drinking water. This means that any technology used to remove SOCs from drinking water must be at least as effective as GAC. Most VOCs are absorbable to some extent. More weakly adsorbing compounds can often be removed by air stripping.[4] In addition to water plant processes, many devices for "purifying" water at the point of residential use are currently marketed in the United States. Most of these devices use GAC as a sorbent.

Indoor Mass Balance Model

As a result of growing concerns over human health impacts associated with indoor VOC exposures, the USEPA has recently began to study the issue. A major study was initiated by the USEPA in 1995. The primary objective of this study was evaluation of exposures and risks associated with inhalation of contaminants in drinking water. A portion of this project is currently under way at Virginia Tech.

Virginia Tech faculty and students are developing PC-based, mathematical models and simulations that will assist the USEPA and other regulatory agencies estimate inhalation exposure to drinking water contaminants. The mathematical models utilize a three-compartment model developed by McKone (1987) to represent a typical household.

The equations from the assembled mathematical model are incorporated into a program, written in the programming language C. The concentrations of the chemical contaminants in each compartment are evaluated and the exposure dose is calculated. A two-dimensional Monte Carlo analysis is used to estimate the uncertainties associated with the exposure predictions. The analysis is currently carried out for radon and is to be extended to include radon progeny and the entire range of chemical contaminants found in drinking water supplies.

At the completion of this project, a user-friendly computer model will be available that can be used to estimate the exposure and associated uncertainties for ingestion, inhalation and dermal sorption of chemical contaminants in drinking water. The software will serve the following purposes:

• Provide researchers and public health officials with a tool to estimate exposure to the full range of chemical contaminants found in drinking water; and
• Allow public health officials to respond to concerns about sensitive sub-populations and to facilitate further refinements to the exposure prediction procedure by allowing easy incorporation of additional data as it becomes available.

The following sections provide more detailed information on the functionality of the models

Exposure Pathways

Human exposure to chemical substances occur primarily through:

• ingestion of drinking water.
• inhalation of air in contact with contaminated water.
• sorption of contaminants by the skin.

Traditionally it has been assumed that the major route of exposure to drinking water contaminants is by ingestion, however studies on the relative importance of these pathways indicate that exposure to volatile chemicals through routes other than direct ingestion may be as large as or larger than exposure from ingestion alone. This indicates that there is clearly a need to develop exposure models for the entire range of chemical contaminants found in drinking water and for the three primary exposure pathways.

Mathematical Model

McKone (McKone, 1987) developed a compartment model of a typical household to simulate transfer and distribution of volatile chemicals inside the home. This model consisted of three compartments: the shower, the bathroom and the remainder household volume. Figure 1 illustrates the constituents of the model and indicates the mass flow pathways between the different compartments of the model.

A mathematical model was assembled for the mass transfer between the three compartments in the house which comprised of the following equations:

where


Ci(t)	= the concentration in compartment i (mg/L),
Vi	= the volume of compartment i (L),
Qi(t)	= the source term of compartment i (mg/min),
qij	= air exchange rates from compartment i to compartment j (L/min),
	(subscripts s, b, a refer to the shower, the bathroom and the remaining household compartments respectively).

The air exchange rates (q_ij) are evaluated from design variables such as volume of each compartment and residence time of air in each compartment. The source terms (Q_i(t)) are used to account for the influx of volatile contaminants from the drinking water in each compartment. The source terms are evaluated using the volatilization fraction, i.e., the fractional mass of volatile contaminant transferred from the aqueous phase to the gaseous phase. In this context some basic principles of mass transfer, which are used to calculate the source terms, are listed below.

where H = Henry's constant, Cw = concentration of contaminant in water.	One of the fundamental principles in Mass Transfer Theory is the Henry’s Law, which illustrates the equilibrium between the gas phase and the liquid phase concentrations. The Henry’s Constant denoted by H is a measure of the volatility of a compound. Higher the H the higher the gas phase concentration.
where KOL = Overall liquid phase mass transfer coefficient, Cg = concentration of contaminant in the air (gas phase).	The mass transfer between the liquid and the gas phase is given by the Mass Transfer Flux. The KOL used to calculate this flux is the overall liquid side mass transfer coefficient.
where KL = Liquid phase mass transfer coefficient, KG = Gas phase mass transfer coefficient.	KOL is estimated using the Two Resistance Theory which expresses this KOL as a function of the liquid and gas side mass transfer coefficients. The phase that controls the equilibrium is determined again by the value of H.

Therefore applying the above principles, general equation for the source term in the mass transfer model is :

where

Some of the important variables used in the model are :

• the number of people in a house,
  
• the volume of each compartment of the house,
  
• the residence time in each compartment,
  
• the water usage in each compartment,
  
• the transfer efficiency of the volatile contaminant (or the mass transfer coefficient)
• in each of the three compartments,
  
• and the duration the compartment is in use.
  

The above set of simultaneous differential equations is solved by numerical integration with time using a program in C, to calculate the concentrations of the volatile contaminants in each compartment as a function of time. Using these concentrations the exposure to that particular volatile contaminant is evaluated. Currently the C program evaluates exposure through the inhalation pathway. This exposure is then used to determine the extent of risk posed by inhalation of any volatile organic compounds that might be present in air that is in contact with contaminated drinking water.

Uncertainty Analysis

In an environmental risk assessment, there are generally three broad sources of uncertainty:

• Scenario uncertainty involves ascertaining the basic appropriateness of the facts, and the inferences used to select the exposure scenarios of concern. For example, selection of exposure scenarios such as inhalation.
• Model uncertainty refers to the uncertainty and potential for error introduced by a mathematical model's simplified representation of an exposure scenario or a dose-response relationship.
• Variable uncertainty refers to the inability to determine accurate values for the variables in an exposure or risk model, due to factors such as measurement error, systematic error or random error.

There are two fundamentally different types of uncertainty :

• Natural or stochastic uncertainty is due to variation or heterogeneity among different members of a set or population. For example, there are differences between people in the amount of water they drink and the amount of time they spend indoors, differences between houses in room size and ventilation rate, and differences between water systems in radon concentration. Because of this variability, there is no unique answer to a question such as "What is the risk from chloroform in water?". The answer takes the form of a range of possible values, most commonly described in terms of statistics such as the average, median, geometric mean etc.
• The second type of uncertainty results from lack of knowledge about the parameters of the variables of a model or system. It is therefore possible to only provide a range of alternative estimates of the true (but unknown) value of a parameter. This type of uncertainty can be reduced by gathering better data. Examples of knowledge uncertainties in this analysis include uncertainty around the true mean and/or the standard deviation of variables such as VOC concentration in water.

Monte Carlo Simulation

Continuous stochastic variables are most conveniently and completely described in terms of probability density functions (PDFs). Each variable in the mathematical model described above has an associated PDF. Each PDF requires one or more parameters, to calculate the value of a variable. These parameters in turn have their own PDFs because of inherent uncertainty in the data collected. Hence a two dimensional Monte Carlo Method is used in the program. The program developed by the project team has all the PDFs for the variables built in and uses the Monte Carlo Method to conduct uncertainty analysis. In the outer loop of this two dimensional simulation the parameters for each variable are evaluated and in the inner loop the actual value of the variable itself is calculated. These values of the design variables are then used to solve the mass transfer model and predict exposure. Further uncertainty analysis are conducted to compute percentiles of the exposure.[6]

Program Inputs and Outputs

Since the PDFs for the design variables are built into the program, the only inputs required to be entered are the number of outer and inner loops. A typical case of input needed for the program is shown in Figure 2 below.

Figure 2: Program Input.

A typical output of design variables generated by the program is as shown in the Figure 3 below.

Figure 3: Output values of design variables from the C program.

Figure 4 below illustrates the uncertainty analysis results from the program.

Figure 4: Output values of exposure percentiles from the C program.

Figure 5 shows the exposure results from the program for the three volatile organic compounds considered.

Figure 5: Exposure Results for VOCs

Compound	Exposure
CCl3F	1016
CHCl3	828
DBCP	180

Results

Concentration profiles for some VOCs obtained through computer modeling are shown in the following figures. The compounds chosen were - Fluorotrichloromethane (CCl₃F, H=3.0), Chloroform (CHCl₃, H=0.12) and 1,2-dibromo-3-propane (DBCP H=0.005) - selected for the wide difference in the values of their Henry's constants.

Figure 6: Concentrations in the three compartments for Fluorotrichloromethane,
Chloroform, and 1,2-Dibromo-3-propane

The above graphs indicate that compounds of higher volatility are present at higher concentrations in the gas phase. Also the concentration profiles exhibit a characteristic trend in each compartment, this is due to the source terms which come into play at different times during a day.

Direct Measurement of VOCs in Indoor Air

Provided that mathematical models can reasonably predict concentrations of VOCs in indoor air, they can be an extremely valuable tool. Direct measurement is expensive and time consuming. Extremely small concentrations (< 1 mg/m³) of VOCs in indoor air can adversely affect human health. These concentrations can be below detection limits of most analytical techniques. Therefore accurate analysis requires concentrating samples before analysis.

One frequently employed method is an adsorption technique. Indoor air is metered through a glass tube packed with sorbent (frequently GAC) for a period of several hours. The VOCs deposit on the sorbent and must be removed prior to analysis. Solvent extraction or thermal desorption are commonly employed for VOC removal. Using an appropriate analytical instrument and methodology, the total mass of the VOC captured in the sorbent can be determined. Dividing VOC mass by the total volume of air metered through the sorbent packed tube yields the concentration (mass/volume) of VOCs that were in the indoor sample. Two sorbent packed tubes and sampling lines used to capture a VOCs sample from an environmental chamber are shown in the photograph below.

Figure 7: VOC sampling apparatus.

References

[1]	Ram, N.M., Christman, R.F., and Cantor, K.P., "Significance and Treatment of Volatile Organic Compounds in Water Supplies," Lewis Publishers, Chelsea, Michigan, 1990, p. 485.
[2]	McKone, T.E., "Human Exposure to Volatile Organic Compounds in Household Tap Water: The Indoor Inhalation Pathway," Environ. Sci. Technol., 1987, Vol. 21, No. 12, pp 1194-1201.
[3]	Ram, Christman, and Cantor, pp. 3-9.
[4]	Ram, Christman, and Cantor, p. 229.
[5]	Wilkes, C.R., et al., "Inhalation Exposure Model for Volatile Chemicals from Indoor Uses of Water," Atmospheric Environment, 1992, Vol. 26A, No. 12, pp 2227-2236.
[6]	EPA Report, Office of Water, U.S.E.P.A., 1995

Water Treatment Primer Table of Contents

Previous Topic

Next Topic

You are visitor number to our Web Page ! This is one crazy counter ! We hope you enjoyed this visit !

Send comments or suggestions to the Student Authors by clicking on their photographs !

Niranjan Aiyagari	Steve S. Cox	Karpagam Sankaran

Faculty Advisor:

Daniel L. Gallagher :
Copyright © 1997 Daniel Gallagher
Last Modified: 12-07-1997