Since Part 1 of this topic, Active Soil-Gas Method, was printed in October 2002, U.S. EPA has initiated a series of informational workshops/conferences that address the soil-gas upward-migration risk pathway. (See http://www.clu-in.org/conf/tio/vapor/resource.cfm for more information.) While the active soil-vapor method is discussed in detail during these workshops, there is little discussion on the surface flux-chamber method. Yet, based on some of the default approaches recommended by U.S. EPA, the surface flux-chamber method may be the best method to use in some situations. Why is the method not discussed in the guidance? Primarily because of a lack of familiarity, experience, and understanding by the environmental community, including regulators, consultants, and contractors. So, let's take a look at this field technique and see when and how it can aid in the assessment of this risk pathway.
Let me start by making two important points concerning surface flux chambers:
There is currently no published or official U.S. EPA method for surface flux chambers. There is a published study performed under contract with EPA that gives a recommended protocol, but it is not regulatory guidance.
There is no one right way to perform a flux-chamber survey. Like any field technique, there are variations of the method-the suitability of each depends on the project goals.
Direct measurement of compound fluxes has been commonly performed in the oceanographic, soil science, and natural resource exploration (i.e., petroleum and minerals) communities for many years. The approach has not been as readily applied to environmental risk assessment.
In the mid 1980s, Radian Corporation, under contract to U.S. EPA, performed a series of testing programs on the method that were summarized in a users guide (Kienbusch, 1986). The method described in this document has often been incorrectly labeled as the official U.S. EPA flux-chamber method. While the document gives a thorough treatment of one flux-chamber approach, including a comprehensive treatment of statistical sampling, it is a recommended protocol only, has several limitations for risk-based applications as described further in this article, and is a difficult read for the inexperienced user.
Subsequent documents by Radian for EPA on air emissions at Superfund sites contain more general discussions on flux chamber methods and applications (Eklund & Schmidt, 1990).
Currently, risk due to the upward flux of vapor-phase contaminants into an overlying structure is assessed either from direct measurements of indoor air or by the collection of groundwater and/or soil-gas data and the application of a predictive transport model or attenuation factor. Both approaches have limitations.
Surface-flux chambers installed on site.
The determination of upward contaminant flux from the measurement of indoor air is subject to such complications as contributions from the natural background of contaminants in ambient air (especially in urban locations), contributions from sources from within the structure, and temporal and spatial variations. Further, the process is often a logistical headache, especially when the measurements are performed in private residences.
For these and other reasons, U.S. EPA currently recommends collecting subsurface groundwater or soil-gas data prior to the measurement of indoor air concentrations (OSWER Draft Guidance for Evaluating the Vapor Intrusion to Indoor Air Pathway from Groundwater and Soils, November 29, 2002, www.epa.gov/correctiveaction/eis/vapor.htm).
The determination of upward contaminant flux using groundwater or soil-gas data requires the application of a predictive model or attenuation factor to compute the contaminant concentration in an overlying room. Attenuation factors, commonly referred to as alpha factors (α), are defined as the concentration of indoor air to either measured soil-gas concentration (soil-gas alpha) or indoor air to a calculated soil-gas value from groundwater concentrations using the compound-specific Henry's Constant (groundwater alpha).
At present, attenuation factors predicted by the models have yet to be thoroughly validated with field data. Until such time that a sufficient data base is accumulated to test the model-derived values, U.S. EPA is recommending the use of default attenuation factors in its vapor intrusion guidance that are conservative and may be overprotective by up to several orders of magnitude. The ramification is an increased likelihood of falsely concluding that there may be a risk when the assessment is based on subsurface data, especially if site-specific data are not available.
The flux-chamber approach provides a direct measurement of the subsurface contaminant flux and therefore alleviates the uncertainty introduced from the existing predictive flux models or the use of an overly conservative alpha factor. Assuming proper placement, as described below, fluxes measured by this approach should, in theory, be reflective of all of the subsurface fate and transport processes that are operative and difficult to model (e.g., phase partitioning, bioattenuation, preferential pathways, and advective flow).
If flux chambers can solve some of the problems of the other approaches, then why not use them? Because, as with any method, flux chambers are not applicable to all situations and they have their share of limitations that must be understood before attempting to employ them on a site.
Flux chambers are not applicable to every type of structure or site. For example, the use of chambers in basements or any other subterranean enclosure is not practical because the four walls of the basement could also be a source of vapor flux. Also, flux chamber results from undeveloped lots may or may not be representative of fluxes into a future structure. On one hand, the measured flux could be over-estimated because there is no building foundation impeding the flux; on the other hand, the measured flux could be under-estimated for reasons such as the lack of pressure-induced advective flow caused by the heating or ventilation system in the overlying structure.
While factors influencing the results from this method include adequate coverage, measurement time, and temporal variations (these factors also influence indoor air and soil-gas results), the two greatest concerns I have heard voiced from skeptics on this method are:
Doubt as to whether chambers measure the actual flux into a structure due to our inability to place chambers in the location of highest vapor intrusion. Experience from radon intrusion studies over the years has shown that in many structures, especially older ones, the most permeable zones into basements and slabs are at the junction between the structure footing and the slab/floor (i.e., near the walls) or from conduits (e.g., utility lines and pipes) protruding through the walls or slab. Because chamber designs preclude measurements in such locations, the concern is that measured fluxes will be lower than actual fluxes in such situations.
Concerns as to whether the air-flow conditions inside a chamber match the air-flow conditions in a room. If the air flow in the chamber is more restricted, fluxes could be reduced. If the air-flow conditions in the chamber are higher than in the room, measured fluxes could be over-estimated if upward advection is created or under-estimated if chamber air is pushed downward into the subsurface.
Structures with basements, older construction, and structures containing many conduits through the slab, walls, or floor are not likely to be good candidates for flux chambers. Structures with newer slab-on-grade construction are most applicable for flux chambers. Chambers are applicable to undeveloped lots, as long as effects caused by a future building are considered when interpreting the results. Enough chamber measurements should be made to ensure that spatial variations around the building footprint due to potential preferential pathways (e.g., near the footing and slab junction) are adequately covered. Finally, chamber measurements should be made for a period of time sufficient to ensure that any temporal variations in flux are averaged.
There are basically two different types of flux-chamber methods: (a) the Static-(Closed) Chamber Method and (b) the Dynamic-Chamber Method. Both methods offer advantages and disadvantages as described below.
In this method, there is no introduction of gas into the chamber during the incubation period. Contaminants flux into the trapped and stagnant chamber volume and the contaminant concentration builds up over time. Discrete samples for analysis are withdrawn either at the end of the incubation period or, preferably, at regular intervals during the incubation period. In essence, the chamber acts like a "mini-room," except there is no air exchange, which provides a time-integrated sample, similar to a Summa canister collected over a specified time period.
The equipment is very simple, consisting essentially of a collection container with sampling ports. (See Figure 1.) Chambers have been made from 55-gallon drums (metal or plastic), Summa canisters, galvanized cans, bowls, and pots. More important than the type of container is the chamber material. For most VOCs, the chamber should be constructed of an inert, non-adsorbing material, such as polished stainless steel with a minimum of rough adsorbing sites. (Teflon is not a good choice due to adsorption on its surface.)
This method offers many operational advantages over the dynamic method including the following:
Croom = Cchamber * Hchamber/Hroom
Flux = Cchamber * Vchamber/Achamber * T
Where: C refers to concentration
H refers to height
V refers to volume
A refers to area
T refers to incubation time
For example, a measured concentration of 10 µg/m3 after an 8-hour period in a 10-inch high chamber would be equivalent to a concentration of 1 µg/m3 in an 8-foot high room. This value can be compared directly to tabulated acceptable room concentrations for the applicable risk level and allowed room air-exchange rate. Or the value can be easily converted to a flux for input into an exposure model.
There is one major disadvantage to the static method: If chamber concentrations build up to a significant fraction of the subsurface concentration, the flux will be impeded. By Fick's Law, the flux is directly related to concentration gradient; hence, for example, a 20 percent reduction in concentration gradient will lead to a 20 percent reduction in flux.
For sites where emissions are known to be high (e.g., near landfills, compost piles), the flux reduction caused by concentration build-up could be significant. But, for most upward risk applications, concentration build-up will most likely not be significant. For example, existing case studies indicate that the attenuation factors are less than 0.01 for chlorinated solvent sites and less than 0.001 for hydrocarbon sites. The corresponding concentration build-up in a static chamber would be 20 percent and 2 percent of the subsurface soilgas concentration, respectively, for these two attenuation factors.
Any reduction in the measured flux can be identified and corrected for by measuring the chamber concentration periodically during the incubation period. If required detection levels can be achieved, I recommend on-site analysis to enable real-time feedback. Alternatively, multiple samples can be collected from the chamber over the incubation time for off-site analysis. If the measured concentration in the chamber is within 25 percent of the subsurface soil-gas concentration, then it is possible the measured flux was underestimated.
This is the method described in the Radian's Users Guide. In this method, an inlet gas (sweep gas) is continuously introduced into the chamber during the incubation period and an equivalent amount of the chamber gas is allowed to escape. The system is assumed to reach a steady-state concentration after four or five chamber-residence times, where one residence time equals the chamber volume divided by the sweep-gas flow rate.
At steady state, the contaminant concentration in the outlet gas is equivalent to the concentration in the chamber. The concentration in the outlet gas is monitored with a meter, or a sample of the outlet gas is collected for analysis, depending on the required detection level for the contaminants of concern. For risk-based applications requiring low detection levels, the typical approach is to collect a batch sample of the outlet gas for off-site analysis after steady-state conditions have been reached (approximately 30 minutes for the nominal conditions given in the Radian report).
The major advantage this method offers is that, except in the most extreme cases, there is little chance for the chamber concentration to build up to a significant fraction of the subsurface concentration due to the inflow and outflow of the sweep gas. Hence, there is very little chance that the measured flux will be impeded by concentration build-up in the chamber.
This method has a number of operational and technical disadvantages, including the following:
As discussed, reliable flux measurements can be made with both chamber techniques. For vapor intrusion applications, where low fluxes are likely to be detected, the static-chamber method offers more advantages and fewer disadvantages over the dynamic-chamber method.
This conclusion was also stated in a subsequent document by Radian to EPA Superfund group (Eklund and Schmidt, 1990). If high fluxes are expected (e.g., chlorinated solvent concentrations near the surface greater than 1,000 times allowable ambient air values), collect multiple samples from static chambers over the deployment period to detect any flux reduction due to potential concentration build-up.
If the dynamic method is used, the output-gas flow (not pressure) must be measured to ensure that the sweep air is not escaping underneath the chamber and impeding the natural flux.
If previous soil-gas data do not exist, the collection of corresponding soil-gas samples near the flux chambers is advised to substantiate the presence of target contaminants in the subsurface, especially at chlorinated solvent sites, where vapor clouds are more common.
The following are some of the key factors to consider when using either flux-chamber method:
Deploy multiple chambers in any program to provide representation of the area of interest and to determine precision. Chambers should be located in areas where surface features suggest possible conduits to the subsurface (e.g., cracks, drains) and close to the external walls near the junction of the footings and slab. At least one chamber should be deployed in the area of maximum subsurface contaminant concentration, if it has been identified from a previous subsurface investigation.
Keep the following in mind: You wouldn't consider proposing or accepting a site-assessment report with only one analysis from one or two borings, would you? So why would you accept only one or two flux-chamber measurements to characterize this risk pathway?
If the dynamic method is used, samples should be collected con- tinuously over the incubation period in a canister equipped with a flow regulator.
Barometric pressure has also been documented to have an effect on emission rates-highest emission rates are found during periods of lower atmospheric pressure. Programs should be avoided during any period of extreme high or low barometric pressure.
Temperature effects have been found to be relatively minor on emission rates, unless the flux is from soil contamination immediately at the surface. The greater effect of ambient temperature will likely be due to changes in the vapor flow below a structure caused by heating/cooling or ventilation systems in the building. Due to this latter issue, flux measurements collected over one or more seasons may be appropriate for locations with large seasonal variations in temperature (high or low).
The following are some of the special concerns associated with using either flux-chamber method:
Unfortunately, there are no regulatory guidance documents governing fluxchamber protocols. The Radian document, referenced previously, is the most comprehensive document, but it only deals with the dynamic method. San Diego County has some limited guidance regarding flux chambers in its Site Assessment Manual, most of which I included in this article (http://www.co.san-diego.ca.us/cnty/cntydepts/landuse/env_health/lwq/sam/pdf_files/presentations/soil-vapor_guide.pdf). Most other papers on flux chamber methods are case studies from vendors supplying the service, conference proceedings, or from other disciplines. Three recent papers comparing fluxes measured with chambers to fluxes estimated by models are by Menatti and Fall (2002), Richter and Schmidt (2002) and Frez et. al (1998).
Flux chambers should be considered to be another valid tool for upward vapor risk assessment, in addition to soil-gas data and indoor-air data. Which method to use on a given site depends upon the site-specific goals and the logistical limitations. In my view, the active soil-gas method described in LUSTline #42 offers less uncertainty and more versatility than the other methods for most situations. However, in situations where active soil-gas data are not definitive or can't be collected, and reliable indoor air samples cannot be collected due to background issues or other logistical reasons, then flux chambers may be the best approach.
Several reviewers of this article prior to publication raised the issue as to whether burial of adsorbent tubes into the cracks of the slab, utility conduits, or room edges might be another viable alternative. In my opinion, such an approach is not useful for quantitative results because one does not know the volume of air that passes through the adsorbent while it is emplaced. Without this knowledge, concentrations cannot be computed. However, one could use this approach as a screening method to decide where the areas of highest flux into a structure are to assist in locating the chambers.
Blayne Hartman, Ph.D., is a principal of HP Labs and the founder of TEG. He has lectured on soil vapor methods and data interpretation to over 20 state agencies and to all of the U.S.EPA regions. Blayne has contributed numerous articles to LUSTLine and authored chapters in three textbooks on soil vapor methods and analysis. This is his fourth article for LUSTLine on upward vapor migration. For more information, e-mail Blayne at email@example.com or check out his Web page at www.HandPmg.com.
Kienbusch, M.R. (1986), Measurement of Gaseous Emission Rates from Land Surfaces Using An Emission Isolation Flux Chamber, Users Guide, EPA/600/8-86/008, NTIS #PB86-223161)
Eklund, Bart and Charles Schmidt (1990), Estimation of Baseline Air Emissions at Superfund Sites, Air/Superfund National Technical Guidance Study Series, Vol II, August 1990, EPA-450/1-89- 002a
Frez, W.A., Tolbert, J.N., Hartman, B., and T.R. Kline (1998), Determining Risk Based Remediation Requirements Using Rapid Flux Chamber Technology, in proceedings, Remediation of Chlorinated and Recalcitrant Compounds, Monterey, CA, Battelle Press
Menatti, J.A. and E.W. Fall (2002), A Comparison of Surface Emission Flux Chamber Measurements to Modeled Emissions from Subsurface Contamination, 95th Annual Conference of the Air & Waste Management Association, June 2002, Baltimore, Maryland, Paper No. 42734
Richter, R.O. and C.E. Schmidt (2002), Assessing Realistic Risk to Indoor Occupants from Subsurface VOC Contamination, in proceedings, Symposium on Air Quality Measurement Methods and Technology, November