Passive Diffusion Bag Samplers

Description

Both types of passive diffusion bag (PDB) samplers available today take advantage of semi-permeable membrane technology to gather contaminants from water. One type of PDB sampler is an equilibrium sampler. It typically contains reagent-grade organic-free water in a semi-permeable membrane (Figure 1). When this sampler is placed in contact with an ambient medium (contaminated water), contaminants diffuse across the semi-permeable membrane into the reagent-grade organic-free water. After some time, the bag is retrieved and the water inside is drained into a sampling vial for later analysis. This type of sampler can be used to monitor groundwater and determine contaminant entry points in groundwater/surface water interaction areas. In some designs, a 40-milliliter (mL) vial is placed in the bag to collect the volatile organic compounds that diffuse into the vial air, which is later analyzed. Another type of passive sampler contains a sorbent material that collects but does not release contaminants that come in contact with it inside the semi-permeable membrane. This is not an equilibrium sampler and provides a total concentration that can be used to obtain an average over the period it is deployed. The Semi-Permeable Membrane Device (SPMD) is an example of this kind of non-equilibrium passive sampler (Figure 2).

Figure 1. Typical polyethylene Passive Diffusion Bag Sampler with stainless steel cable and weight

Figure 2. Lipid containing semipermeable membrane device and a typical deployment apparatus. Courtesy USGS Columbia Environmental Research Center

Passive diffusion water sampling requires sufficient contact time between the chemical contaminants and the semi-permeable membrane for the chemical contaminants to reach equilibrium on both sides of the membrane. Non-equilibrium samplers, such as the SPDMs, need to be in contact with the sampling medium long enough to retain a sufficient quantity of contaminants to analyze the average contaminant concentration over time. Reported equilibration times range from 48 hours to 4 weeks, depending on the temperature and contaminant of interest.

Typical Uses

PDB samplers are generally used to reduce sampling costs primarily when long-term monitoring is required. They also are used to increase the number of discrete data points taken within a well screen and decrease the uncertainty of remedial design or optimization efforts.

PDB samplers can collect non-polar volatile organic compounds (VOC) in groundwater, surface water, and sediment pore water. They are most frequently used at sites with long-term VOC monitoring programs to collect low levels of chlorinated solvents, such as tetrachloroethene (PCE) and petroleum derivatives, such as benzene, toluene, ethylbenzene, and xylenes (BTEX), in groundwater. SPMD samplers are typically used to collect semivolatile organics in surface water and groundwater. Other non-equilibrium samplers that use charcoal or other similar sorbents are used for volatiles.

Figure 3. Deploying multiple PDB samplers can detect heterogeneity in contaminant concentrations within the screened interval

PDB groundwater sampling methods can be used to identify contaminated zones within wells with large screens by stringing a series of bags together across the screened interval (Figure 3). Contaminant concentrations can vary widely even within a 10-foot screening interval. Data from PDB samplers can be used to help isolate the zones where contamination is highest so that remedial systems can be designed appropriately.

Although PDB sampling methods reduce overall sampling costs dramatically in comparison to conventional methods, PDB technology has some significant limitations. The semi-permeable membrane can foul easily, and PDB samplers cannot accurately measure some chemical constituents, such as alcohols and ketones and chemicals greater than about 10 Angstroms are generally too large to pass through the polyethylene. They also are inadequate for the collection of natural attenuation parameters and other basic water quality indicators, such as redox potential, pH, and dissolved oxygen.

Theory of Operation

The semi-permeable container used in PDBs to hold reagent-grade organic-free water is made of low-density polyethylene (LDPE). The LDPE allows VOCs from the ambient medium to move into the container while preventing water from moving across the semi-permeable barrier. Given adequate time, equilibration of the VOCs between the contaminated, ambient medium and the sampling medium inside the PDB will occur. The concentration gradient between the ambient medium and the sampling medium drives the diffusion of contaminants from the groundwater into the sampler. Contaminants will typically migrate from a relatively high concentration in the ambient medium to a low or zero contaminant concentration in the sampler until equilibrium is reached. The necessary contact time between the sampler and the contaminated groundwater depends on the sampler size, membrane composition, groundwater flow rates, contaminant type, water temperature, and other chemical characteristics. If a decrease in contaminant concentrations occurs in the ambient water following equilibrium, the process will reverse, causing contaminants to migrate out of the bag.

The U.S. Geological Survey (USGS) in its User's Guide for Polyethylene-Based Passive Diffusion Bag Samplers (User’s Guide) reports that, under laboratory conditions, the concentrations of benzene, cis-1,2-dichloroethene, PCE, trichloroethene (TCE), toluene, naphthalene, 1,2-dibromoethane, and total xylenes equilibrated with an aqueous mixture of those compounds surrounding the sampler within 48 hours at 21°C. At 10°C, PCE and TCE equilibrated within 52 hours, while the other VOCs in the mixture required longer equilibration times. Chloroethane, cis-1,2-dichloroethene, trans-1,2-dichloroethene, and 1,1-dichloroethene equilibrated in 93 hours, while vinyl chloride, 1,1,1-trichloroethane, 1,2-dichloroethane, and 1,1-dichloroethane equilibrated within 166 hours.

Based on laboratory data and additional information collected during PDB field deployment, the USGS User's Guide indicates that a 2-week equilibration period is generally sufficient to capture most VOCs for environmental sampling purposes; however, smaller diameter samplers equilibrate faster than larger diameter samplers. When applying PDB samplers in ambient waters colder than previously tested (10°C) or when sampling compounds without sufficient corroborating data, a side-by-side comparison with a conventional methodology is advisable to justify the field equilibration time.

PDB samplers can remain in groundwater for extended time periods without substantial damage. According to the USGS, PDB samplers have been deployed routinely in a single well for up to 1 year without loss of performance. However, leaving bags in wells for an extended time may result in biofouling, which can restrict performance. In addition, by the nature of the operational principles governing performance, PDB samplers provide concentrations for only the most recent equilibration time, rather than an instantaneous snapshot of concentrations in a particular location or an average over deployment time.

The SPMD sampler operates differently from the typical PDB sampler. The SPMD sampler consists of a neutral, high molecular weight lipid (> 600 daltons), such as triolein, which is encased in a thin-walled (50-100 µm), lay-flat polyethylene membrane tube (Figure 2). The nonpolar chemicals that pass through the nonporous membrane concentrate in the lipid. SPMDs are designed to sample chemicals dissolved in surface water or groundwater, thereby mimicking the bioconcentration of organic contaminants in the fatty tissues of organisms. The SPMD enables sampling of trace concentrations of organic contaminant mixtures for toxicity assessments and toxicity identification evaluation (ITRC 2006).

System Components

A typical PDB equilibrium sampler for groundwater sampling consists of a 1- to 2-foot long LDPE tube, sealed at each end, and filled with laboratory-grade reagent water (Figure 4). PDB samplers are available either prefilled with deionized organic free water or unfilled. Unfilled samplers can be filled by the operator through a plug, which also allows for sample recovery. PDB samplers used in 2-inch-diameter wells are about 1.2 inches in diameter. Other sampler diameters are proportional to the size of the well. A polyethylene mesh is occasionally used to protect the sampler from abrasion.

Figure 4. Typical components for a single PDB sampler deployment

The PDB sampler is attached to a weighted line and lowered into position at the target sampling depth. If the sampler has an attachment point of sufficient strength, weights may be attached directly to the sampler. The line used to suspend the PDB must be strong enough to support the PDB sampler and the weights (Figure 4). The line should be non-buoyant and resistant to stretching. Examples of suitable lines are braided polyester, stainless steel wire, and teflon-coated stainless steel wire. Rope and wire that cannot be decontaminated prior to reuse could contribute to cross-contamination of future samples and therefore should not be reused.

A standard SPMD is 2.5 cm wide by 91.4 cm long, and it contains 1 mL of triolein. SPMDs of different sizes can be made by maintaining the ≈ 100 cm²/g SPMD ratio (ITRC 2006). They are typically deployed in rigid perforated canisters for protection.

Mode Of Operation

PDB samplers are deployed at the target horizon within a screened or open interval of a well that is between 5 and 10 feet in length. If the screened interval is greater than 10 feet in length, the most appropriate target horizon must be identified (Figure 5). Multiple PDB samplers or results from real-time measurements, such as those obtained using a membrane interface probe, can be useful when identifying the target horizon for monitoring. Chemical stratification caused by slight changes in stratigraphy may be significant even in wells completed in permeable aquifers.

Figure 5. Variations in TCE concentrations with depth

When each PDB sampler is retrieved from a well, it should be examined for biofilms, iron coatings, or tears in the membrane. All observations should be noted in the field log-book. Torn PDB membranes should be discarded before analysis.

Transfer of the water from inside the PDB to 40-mL volatile organic analysis (VOA) vials should occur immediately after the sampler is retrieved from the well. Failure to transfer the contents immediately might result in the loss of some contaminants that diffuse out of the bag. Some PDBs have a discharge device inserted into the bag so that the water may be easily poured into the VOA vial, while others require cutting the end of the bag with decontaminated scissors to release the contents inside. Samples in VOA vials should be preserved according to the requirements of the analytical method and stored at 4°C in accordance with standard analytical protocols.

SPMDs are transported to and from the sampling site in gas-tight metal cans. After field-deployed SPMDs are retrieved from the well, they should be stored frozen or at least on ice until processing. Chemical residues in the SPMD are recovered through organic solvent dialysis, which involves submersing the SPMD in an organic solvent, such as hexane. The analytes diffuse out into the hexane while the lipids remain inside the tubing. Following dialysis, all targeted chemicals are in the hexane and the used SPMD can be discarded (ITRC 2006).

Target Analytes

PDB samplers are generally used to detect low levels of VOCs. If contaminant concentrations are high, the sampling media of the PDB can become saturated and less representative of actual VOC concentrations. Oxygenated or more polar substances, such as the methyl ketones, tend not pass through the LDPE as effectively as less polar substances. Detectable sensitivities can be in the low parts per billion range. Typical groundwater parameters, such as dissolved oxygen, conductivity, and natural attenuation parameters, cannot be collected using PDB samplers. Table 1 provides a list of target analytes that have been tested under laboratory conditions for PDB sampling effectiveness. The results are valid provided the samplers are implemented under geological conditions that do not compromise the representativeness of the results. Table 2 provides a list of analytes that field data suggest should be amenable to PDB sampling.

SPMDs are used to sample hydrophobic, bioavailable semi-volatile organic chemicals, such as polychlorinated biphenyls (PCBs), polycyclic aromatic hydrocarbons (PAHs), organochlorine pesticides, dioxins and furans, selected organophosphate and pyrethroid pesticides, and many other nonpolar organic chemicals (ITRC 2006).

Performance-Related Issues

Method Reporting Limits

The size of the PDB used for sample collection may limit the use of analytical methods that require higher sample purge volumes to increase instrument sensitivities. Reporting limits can be lowered using a higher volume of purge water during the analysis. The usual volume required for VOC analysis using methods such as SW-846 method 8260B is around 25 milliliters per analysis. The project team can ensure the collection of a sufficient volume by using a larger sampler.

As with most types of analyses, the sensitivity can be driven by the presence of contaminants other than those targeted for the project. When relatively few contaminant species are present, maximum sensitivity is generally achievable. When complex mixtures of constituents, such as hydrocarbons with chlorinated solvents are present, however, bag performance and analytical sensitivities may become less optimal. Before PDB sample collection is selected as the preferred alternative when complex mixtures exist, a direct comparison between traditional methods and the PDB samplers should be considered.

Sampling Design Considerations

Geologic and hydrogeologic factors must be reviewed carefully before a PDB sampling scheme is designed. In general, equilibration times can be longer in low-permeability materials. Prior to choosing a PDB sampler, vertical flow data should be collected from the wells. When well screens are less than 5 feet, and the suspected vertical gradients are minimal (much less than 0.5 liter per minute [Church and Granato 1996]) in the formation, the bag sampler is usually placed in the middle of the screened interval. When screened intervals are greater than 5 feet, multiple samplers should be used to limit the potential for missing contaminants that slip into preferred pathways at specific depths. Where vertical flows are likely, or stratification appears to control contaminant distributions, alternative sampling methods, such as straddle packers, can be used to limit vertical mixing and ensure the representativeness of the data.

Quality Assurance/Quality Control

Prior to the final placement of PDB samplers in a well, the samplers must be prepared for use. Such handling can introduce systematic or other biases into the sampling results. Thus, an equipment blank should accompany the shipment of bags to and from the field. Acetone, a common laboratory contaminant, does not easily pass through the PDB samplers; therefore, the presence of acetone may indicate a source of laboratory-related artifacts. A longer sampler may be needed when additional quality control samples are collected as matrix spikes or duplicates.

Sample Throughput

Advantages

Multiple samplers, spaced vertically, can provide a vertical profile of groundwater samples at 1-foot intervals.

Passive diffusion sampling reduces or potentially eliminates purge water associated with well sampling, and it reduces the labor, logistical requirements, and expense of traditional sample collection. For example, sampling with a traditional monitoring well requires three well volumes of water to be purged prior to sampling. The cost for the traditional well sampling method includes the labor required for purging the well, the equipment (bailers or pumps), decontamination of the sampling equipment, and disposal of the purge water and decontamination solutions (investigation-derived waste). The low flow sampling technique will lower the purge water requirements but not reduce them entirely.

The relative ease of deploying and recovering passive diffusion samplers lowers the level of technical expertise involved and therefore the cost required to employ the technique.

Passive diffusion samplers are disposable, and thus they reduce the risk of cross-contamination that can result from incomplete decontamination of traditional samplers.

Because the sampling technique is passive and relies on diffusion of the contaminant, sampling pumps that would have to be decontaminated and would add to the cost are generally unnecessary.

The impacts of sediments on the sampling results are reduced by the small (less than 10 angstroms) pore size of LDPE, which does not allow sediment to pass into the bag.

When determining the contaminant flux between groundwater and surface water, PDBs can be buried in the sediments to measure pore water contamination.

Contaminant flux in surface waters can be measured using SPMDs, especially for bioaccumulative contaminants.

Limitations

Two mobilizations are required to place and later retrieve the samplers from wells.

Passive samplers do not provide direct or real-time data.

The number of compounds for which passive sampling can be used is limited.

Biofouling can make PDBs less effective.

PDB sampling in monitoring wells relies on the presence of an uninhibited horizontal water flow. Other factors, such as vertical flow, biofilms, or iron fouling may negatively affect the quality of PDB sampling data.

Well stratification can be an issue even in wells with small screened intervals. If PDB samplers are used to identify the highest potential concentration in a well, numerous linked samplers may be needed to decide on the optimal placement of the final sampler. This use can increase the initial analytical program costs.

Table 1. Laboratory Test Results Using PDB Samplers

Source: Vroblesky, D.A., and Campbell, T.R., 2001, Equilibration times, stability, and compound selectivity of diffusion samplers for collection of ground-water VOC concentrations: Advances in Environmental Research, v. 5, no. 1, p. 1-12.

Table 2. Contaminants Where Field Data Suggest that PDBs May be Useful in Their Collection

*The data set for this compound was relatively small (fewer than five instances of comparison), so the power of the classification (i.e., acceptable or unacceptable) is fairly low.

Adapted from ITRC. 2006. Technology Overview of Passive Sampler Technologies and Parsons. 2005. Final Results Report for the Demonstration of No-Purge Groundwater Sampling Devices at Former McClellan Air Force Base, California.

Cost Data

ITRC 2006 has published cost estimates for PDB and SPMD samplers. While the typical PDB sampler may cost only about $25, the customized deployment equipment costs about $60 per well when multiple PDBs are deployed. Deployment costs include a weight, poly-tether material, connections to the sampler, identification tag, well cap, and other miscellaneous expenses.

The cost associated with purchasing a typical 91.4-cm long SPMD for commercial purposes includes the membrane-triolein device ($50); SPMD holders, which contain the device; and the canister. Together the holders and canister cost about $250. However, the SPMD holders and canisters can be leased for a monthly rate. Extraction of the lipid, which is necessary prior to use when an ultra-trace level analysis is involved, costs an extra $5. The various methods that can be used to recover and fractionate analytes vary in costs (ITRC 2006).

Technology Quick Reference Summary

PDB sampling of groundwater reduces monitoring costs and improves the value of data collected at sites requiring long-term monitoring. PDB sampling is generally applicable when VOC concentrations (primarily chlorinated solvents and BTEX compounds) are in the low parts per million to parts per billion range and the mixture of contaminants is relatively simple and does not include hydrophilic substances, such as methyl-tert-butyl ether or methyl isobutyl ketones. The value of information obtained from a monitoring point can be increased by hanging numerous bags along screened intervals of less than 10 feet to identify heterogeneities that can affect the maximum measured concentration from a well and influence the remedial design. PDB technology has limited value in vertical gradients greater than 0.5 liter per minute. PDB results reflect a time average that can limit data use under some circumstances.

SPMD has found its greatest use in surface water sampling to collect persistent organic pollutants and determine if they are at levels that bioaccumulate. Since the adsorption of the chemicals in the SPMD is cumulative, the results of the analysis represent the total concentration collected over the period of exposure, which provides an estimate of the average concentration. In some applications a tracer is used with the adsorbent that bleeds at a predetermined rate into the water as it flows by. Analysis of the amount tracer remaining when the SPMD is removed can aid in estimating the contaminant flux.

References Cited

Church P.E. and Granato, G.E., 1996, Bias in ground-water data caused by well-bore flow in long-screen wells: Ground Water, v . 34, no. 2, p. 262-273.

ITRC. 2006. Technology Overview of Passive Sampler Technologies. 115 pp.

Additional Resources

ITRC Diffusion Sampler webpage

USEPA Field Tests of Nylon Screen Diffusion Samplers and Push Point Samplers for Detection of Metals in Sediment Pore Water

USEPA Vapor Diffusion Samplers for Sediments

USGS Diffusion Sampler webpages

• Toxic Substances Hydrology Program Highlights
• SPMD Basics

Disclaimer
http://clu-in.org/char/technologies/passdiff.cfm
Page Last Modified: August 20, 2008