Title: Introduction Text: Passive diffusion bag (PDB) samplers can be used as an alternative method for sampling groundwater monitoring wells for volatile organic compounds (VOCs). Some of the main advantages of using PDBs instead of conventional sampling methods include lower costs, little or no turbidity present in samples, and the ability to vertically profile a monitoring well by collecting multiple samples. This Web Tool reviews the principles behind the use of PDBs and documents their deployment at Navy sites to facilitate the sharing of lessons learned among NAVFAC stakeholders. PDBs are typically composed of low-density polyethylene (LDPE) and contain deionized water. Samplers are placed in monitoring wells at a predetermined optimal depth or multiple depths, and remain in the wells until equilibrium has been reached between the water in the monitoring well and the water inside the PDB sampler. The bags are then removed from the wells and the water inside the sampling bags is bottled and sent for laboratory analysis.

Title: Theory of Diffusion Text: The theory of diffusion explains how PDB samplers can be used to sample monitoring wells. PDBs work through the process of diffusion as it drives molecules towards equilibrium conditions between the contaminant concentration in groundwater and in the water inside the PDB sampler. The theory of diffusion states that chemicals tend to migrate from higher concentration to lower concentration until equilibrium is reached. Fick's Law of Diffusion states that the rate of diffusion mass transfer through a unit area is proportional to the difference in concentrations divided by the distance separating these concentrations. The equation shown represents Fick's Law. The constant D refers to the diffusivity of a given contaminant.

Title: Use and Operation of Passive Diffusion Samplers Text: As groundwater naturally flows, chemicals diffuse across the LDPE bag sampler (typically 4 mils thick) until chemical concentrations inside the bag are equal to those in the groundwater surrounding the monitoring well. A minimal deployment time for PDB sampling of VOCs is about two weeks. Samplers have been left in wells for as long as three months without compromising the integrity of the bag. If monitoring is conducted quarterly, a three month deployment would allow bags to be retrieved and replaced in the same field event. The sampling bags vary in size but typically measure around 2 feet in length and 1-1/2 inches in diameter. Bags can be designed to accommodate different well sizes and sampling requirements; however, they should not exceed 5 feet in length. If the screened interval is greater than 5 feet in a monitoring well, it is recommended that samplers be installed at multiple depths within the monitoring well with one PDB sampler per five feet of screen. The LDPE sampling bag is commonly covered by a protective mesh casing to prevent damage to the bag when deployed and recovered from the well.

Title: Chemical Applicability (1 of 2) Text: Laboratory tests involving known contaminant concentrations have been performed to evaluate the performance of PDB samplers with many VOCs. VOCs for which PDBs have shown favorable results in laboratory tests is shown to the left. Field studies have also been performed and suggest that additional VOCs can likely be sampled using PDB sampling. VOCs that have not produced favorable results in laboratory studies using LDPE diffusion samplers include acetone, methyl-tert-butyl ether (MTBE), methyl isobutyl ketone (MIBK), and styrene. LDPE diffusion samplers cannot be used to sample for metals and other inorganics. Different types of diffusion samplers, including regenerated cellulose dialysis membrane (RCDM) samplers are being studied for additional chemical applicability. Case Study 1 discusses the validation demonstration that is currently being conducted using RCDM sampling bags.

Title: Hydrogeologic Conditions (2 of 2) Text: There are many hydrogeologic factors that must be considered when determining the applicability of passive diffusion sampling at a specific site. Fluctuating groundwater tables can affect the integrity of the sample collected due to the potential for the sampling bag to come in contact with soil gas or ambient air, which could allow escape of VOCs from the sampling bag. Other factors that could affect the quality of the collected sample include groundwater velocity, hydraulic conductivity, and hydraulic gradient. No definite values have been established, but the following site conditions may not be appropriate for diffusion sampling techniques: Groundwater velocity less than 0.5 ft/day (1.76E-04 cm/sec) Hydraulic Conductivity less than 1.0E-05 cm/sec Hydraulic gradient less than 0.001 In addition, the deployment of PDBs in fractured bedrock can be challenging given low flow within the fracture and/or improper placement of the bag and screened interval within the fracture. More information on these limitations is provided below.

Title: Regulatory Issues (1 of 2) Text: The Interstate Technology Regulatory Council (ITRC) conducted an online survey for state regulators in May 2003. The survey was conducted to identify any regulations that might prevent use of PDB samplers. A total of 54 responses from 23 states were received, none of which identified any rules or regulations that would prevent use of PDB samplers. All 23 states had sites that use PDB sampling techniques. Nine of the 23 states indicated that they had guidance for using PDB sampling techniques. It is recommended that use of passive sampling techniques be discussed with regulators prior to implementation. Some regulators may prefer that a site-specific correlation study be performed.

Title: Regulatory Issues (2 of 2) Text: The following Web sites provide detailed guidance on regulatory requirements and proper sampling techniques: ITRC Technical and Regulatory Guidance for Using Polyethylene Diffusion Bag Samplers to Monitor Volatile Organic Compounds in Groundwater USGS User’s Guide for Polyethylene-Based Passive Diffusion Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1: Deployment, Recovery, Data Interpretation and Quality Control and Assurance USGS User’s Guide for Polyethylene-Based Passive Diffusion Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 2: Field Investigations

Title: Transitioning to PDB Sampling Text: A site-specific correlation study is recommended during a transition from low-flow groundwater sampling or other conventional techniques to PDB sampling because of potential limitations of PDB samplers. However, the potential cost savings of using PDB samplers outweighs the investment of a correlation study. In most cases, the correlation study will pay for itself within a year if PDB samplers are found to be a viable sampling method. Actual results of a correlation study are provided in Case Study 2 below. Common steps for transitioning to PDB samplers are as follows: Ensure PDB samplers are appropriate to sample COCs at the site Collect groundwater samples concurrently using PDB samplers and the sampling technique currently in use at the site Develop a set of acceptance criteria which are statistical tests to determine the success of results Perform a statistical comparison of the sampling results from PDBs versus the current sampling technique and evaluate according to the acceptance criteria

Title: Advantages Text: There are many advantages to using PDB sampling over low-flow purge and other conventional sampling methods. Some of the advantages include: PDB samplers are easy to deploy and recover PDB sampling almost or entirely eliminates purge water, reducing or eliminating the cost for its disposal Eliminates turbidity in samples Vertical profiling of contaminants in a monitoring well can be determined using multiple diffusion samplers within one well without mixing Costs are much lower using passive sampling techniques due to reduced labor and lower equipment costs Equipment decontamination is not necessary because sampling bags are disposable There is less wear and tear on the monitoring well There is no need to collect and analyze equipment blanks because equipment does not need to be decontaminated

Title: Limitations Text: There are a few limitations associated with PDB sampling, which include: PDB samplers are not appropriate for site characterization or final site closeout confirmation sampling. LDPE PDB samplers are only effective at sampling specific VOCs and cannot be used for metals and inorganics. However, new and innovative sampling bag materials are currently being studied for additional chemicals including metals, as discussed in Case Study 1 below. The temperature of the groundwater generally must be greater than 10 degrees Celsius (50 degrees Fahrenheit) Vertical groundwater gradients may interfere with the collection of a representative sample from the desired screened interval. Groundwater may experience vertical transport within a monitoring well prior to collection of a sample. There are some limitations to using PDBs in fractured bedrock, primarily with regard to deploying the PDB at the proper depth and the rate of groundwater flow. within the fracture.

Title: Depth Limitations Text: If the screened interval is large or encompasses several fractures, there is a possibility that PDB samplers may not capture the area of interest. However, this also is an issue when using low flow sampling. Therefore, as long as the PDB sampler is placed at a depth that encompasses the inlet depth of the low flow sampler, this should not be an issue.

Title: Flow Limitations Text: The other potential limitation to using PDBs in fractured bedrock is a low flow rate, but leaving the PDB in the well for a slightly longer duration than normal (>14 days) may help resolve this.

Title: Summary Text: Both laboratory and field studies have validated the use of PDB sampling as a viable method for groundwater sampling. Although this sampling method may not be appropriate for all chemicals or phases of site remediation, it can be highly effective at many sites, both technically and economically. The use of PDB samplers is gaining wider acceptance by end-users and the regulatory community with increased usage across the United States. Passive diffusion sampling techniques have been utilized at over 20 Naval facilities nationwide.

Title: Text: No text. Menu placeholder.

Title: Case Study 1 - Project Background Text: Groundwater samples were collected from three Department of Defense (DoD) contaminated sites for bench-scale studies to test a novel PDB material. This demonstration study was sponsored by ESTCP and conducted by NAVFAC personnel. The three study sites are listed below: Site 1: Naval Air Engineering Station (NAES) Lakehurst, New Jersey with trace metals and benzene, toluene, ethylbenzene, and xylenes (BTEX). Site 2: Naval Construction Battalion Center (NCBC) Port Hueneme and Point Mugu, California with BTEX, MTBE, and trace metals. Site 3: Naval Air Warfare Center (NAWC), Trenton, New Jersey, with chlorinated VOCs and monitored natural attenuation parameters.

Title: Demonstration Plan (1 of 2) Text: LDPE diffusion samplers are limited to collection of samples that will only be analyzed for VOCs. Many sites require analysis of one or more additional chemicals, including inorganics, metals, highly soluble VOCs, or semivolatiles, which may prevents the use of LDPE diffusion samplers. Another type of sampler has been developed which is composed of RCDM. RCDM diffusion samplers allow passage of both organic and inorganic chemicals into the bag. RCDM samplers have been tested in the field and show some success. The three main objectives of this study were the following: 1) to determine if the RCDM sampler would collect statistically valid samples for both organic and inorganic compounds; 2) to determine the length of time for samplers to reach equilibrium; and 3) to compare the results from samples collected using RCDM diffusion samplers with those of samples collected using other techniques.

Title: Demonstration Plan (2 of 2) Text: Monitoring wells chosen for each site for the study were well characterized in terms of well construction, hydrogeology surrounding the well, and contaminant concentrations in historical samples. Groundwater was extracted from select monitoring wells and brought back to the laboratory for bench scale testing. Demonstrations were performed on groundwater collected from Sites 1 (Test 1) and 2 (Test 2). Future plans include a bench-scale demonstration of groundwater collected from Site 3. Additionally, in the future, passive diffusion samplers will be deployed for field demonstration at the same DoD sites. Deployment durations for field demonstration will be determined from bench-scale results.

Title: Results - Lakehurst, NJ (1 of 2) Text: For Site 1 (Test 1), Lakehurst, New Jersey, bench-scale tests included trace metals and cations as the chemicals of interest including: total chromium (Cr), hexavalent chromium (Cr), ferrous iron (Fe), manganese (Mn), aluminum (Al), copper (Cu), lead (Pb), arsenic (As), zinc (Zn), nickel (Ni), cadmium (Cd), mercury (Hg), calcium (Ca), magnesium (Mg), sodium (Na), and potassium (K). Results indicated that 95% equilibrium was achieved in one day for As; three days for Al, Cr, Fe, K, and Na; seven days for Ba, Ca, Cd, Cr, Cu, Mg, Mn, Ni, Pb, and Zn; and 28 days for Hg. For triplicate samples, the coefficient of variation measured below 4% for most cations/metals and approximately 7% for Hg.

Title: Results - Port Hueneme/Point Mugu (2 of 2) Text: Test 2 was conducted using groundwater collected from monitoring wells at Port Hueneme and Point Mugu. The chemicals of concern were VOCs, which included 58 chemicals. The combined coefficient of variation for 50 of the 58 VOCs was less than 5%. The coefficient of variation for all of the VOCs was less than 18%. Additionally, field vertical profiling was performed in well CBC10 for VOCs using PDB samplers. The polyethylene diffusion samplers were deployed at five different sampling depths. Results for benzene and MTBE indicated greater chemical concentrations at shallower depths of approximately 11 feet below ground surface (bgs) decreasing to nondetect at approximately 24 feet bgs.

Title: Conclusions Text: As indicated by the bench-scale results, RCDM diffusion samplers can be utilized for sampling more chemicals than LDPE diffusion samplers, including metals and inorganic compounds. However, additional testing is necessary to include more chemicals and to determine optimal deployment times. Future testing will include additional chemicals and field demonstrations of RCDM samplers at two more sites. This Web tool will be updated with additional data as it becomes available. Please sign up to receive T2 updates to keep informed of changes made to the tools.

Title: Text: No text. Menu placeholder.

Title: Case Study 2 - Former Naval Air Warfare Center Text: The former NAWC Warminster, Pennsylvania is a closed facility that ceased operations in September 1996. Groundwater in three operable units at NAWC (OU-1A, OU-3, OU-4) is currently being treated via separate pump-and-treat systems. Wells in all areas are being monitored either quarterly, semiannually, or annually according to a long-term performance monitoring plan (LTPMP).

Title: Sample Collection Text: The PDB sampling Plan consisted of deployment of one LDPE sampler for every five feet of screened interval in each monitoring well, with a maximum of three samplers per well. The samplers were retrieved after equilibrating for a 14-day period. For quality assurance, duplicate samples were collected for 10% of the samples. For the first quarter of diffusion sampling, samples using conventional low-flow purging also were collected immediately after the passive samplers were retrieved from monitoring wells. For the second quarter, low-flow purge samples were collected from monitoring wells that did not have results that met the data evaluation criteria in the first quarter.

Title: Data Evaluation Criteria Text: For each sampling pair collected, the resulting contaminant concentrations were evaluated. PDB sampling was considered viable for a monitoring well if ranges (based on sampling variability) for individual chemical concentrations for the two samples overlapped or if the concentrations collected from the PDB sampler were greater than those collected using the low-flow purge method. If multiple samplers were deployed for a given monitoring well, the sampler exhibiting the greatest concentration was used in the comparison. If a monitoring well did not pass the criteria, reevaluation was performed in the next quarter. If a majority of the wells within an operable unit showed favorable results, future groundwater monitoring was performed using passive diffusion sampling techniques.

Title: Results - OU-1A (1 of 3) Text: A total of 37 of the 39 monitoring wells in OU-1A were sampled using both sampling techniques, with a total of 94 diffusion samplers being deployed. The primary contaminants for OU-1A were tetrachloroethene (PCE), trichloroethene (TCE), and carbon tetrachloride (CCl4), which were detected in 18 paired samples. There were a total of 352 paired contaminant detections, which had an overall r-value of 0.87, indicating strong correlation.

Title: Results - OU-3 (2 of 3) Text: For OU-3, 11 wells were sampled using both sampling techniques, with a total of 27 diffusion samplers being deployed. Only three contaminants were detected in all of the samples collected. All detections were from low-flow purge samples. PDB samplers did not have any contaminants detected. Therefore, PDB sampling did not prove to be a viable sampling method for OU-3. Detected concentrations in the low-flow purge samples were very low, measuring just below or above the method detection limit of 5 ppb. The low concentrations in the 5 ppb range may have contributed to the contaminants lack of detection in the PDB samplers in OU-3. Low concentrations were a big factor, but other unidentified site-specific conditions may have also contributed.

Title: Results - OU-4 (3 of 3) Text: A total of 25 of the 26 monitoring wells at OU-4 were sampled using both sampling techniques. A total of 62 passive diffusion samplers were deployed in monitoring wells. There were 62 contaminants detected in pairs using both sampling techniques. The primary contaminant was TCE, which was detected in six of the sampling pairs. The overall r-value for paired samples was 0.86, which indicated a strong correlation.

Title: Conclusions - NAWC Warminster Text: Based on the results for the first quarter of groundwater monitoring, PDB sampling was approved as a viable method for future groundwater sampling in OU-1A and OU-4, but not in OU-3. In the second quarter of groundwater monitoring, select wells from OU-1A and OU-4 were sampled using both sampling techniques in order to determine the appropriate diffusion sampler deployment depths and times. The reduction in costs of approximately 50% was realized by using the PDB samplers compared to conventional low-flow purge techniques.

Title: Text: No text. Menu placeholder.

Title: Case Study 3 - Naval Air Station North Island, CA Text: Naval Air Station (NAS) North Island in San Diego, California has been an active base since 1917. Groundwater contamination at the site has resulted from historic activities that released chlorinated VOCs and petroleum hydrocarbons including free-phase JP-5 jet fuel and Stoddard Solvent. Free-phase JP-5 jet fuel and Stoddard Solvent are present in monitoring wells. The purpose of the investigation at NAS North Island was to test passive diffusion samplers and to determine whether their use was viable at the base. Passive diffusion samplers were placed in 15 monitoring wells at multiple depths. Two samplers were also placed in free-phase buckets of JP-5 jet fuel and Stoddard solvent.

Title: Stoddard Solvent Text: Stoddard solvent is a colorless, flammable liquid that smells like kerosene. It is a petroleum mixture that is also known as dry cleaning safety solvent or petroleum solvent.

Title: Sampling Plan Text: Tygon tubing was attached to each diffusion sampler and extended to the land surface to allow a groundwater sample to be collected using the low-flow purge technique directly adjacent to where the diffusion sampler was deployed. Several methods of low-flow purge were utilized during the study including the bladder pump, peristaltic pump, and dedicated bladder pump. Diffusion samplers were allowed to remain in monitoring wells for a period of 65 to 71 days. Low-flow purge sample collection was performed immediately after the PDB samplers were retrieved from the wells. During low-flow purge sampling, water quality parameters were monitored until stable. All samples were analyzed for VOCs by Environmental Protection Agency (EPA) method 8260B. Duplicate samples were collected for 10% of the samples.

Title: Data Evaluation Text: Individual chemical concentrations for diffusion sampling were compared to chemical concentrations obtained from low-flow sampling by use of a bladder pump and a peristaltic pump. Integrity of the passive diffusion samplers deployed in the free-phase JP-5 jet fuel and Stoddard solvent also were evaluated.

Title: Results: Bladder Pump Comparison (1 of 2) Text: Tests that showed the most direct comparison were the results collected from low-flow sampling using a bladder pump to the results collected using PDBs. Trichloroethene (TCE) and 1,1-dichloroethene (1,1-DCE) concentrations in monitoring wells MW-9 and MW-5D differed by between 3 to 12 percent, which is within an acceptable sample-collection variability. The percent difference for tetrachloroethene (PCE) was a little higher (around 21%). Data from monitoring well MW-5 also showed favorable TCE results, which differed by approximately 17%. Monitoring well MW-12 showed favorable results for TCE; however, the concentration of cis-1,2-dichloroethene (cDCE) from the PDB sample measured 78% lower than the concentration from the low-flow purge sample. It is possible that the difference could be attributed to in-well mixing.

Title: Results: Peristaltic Pump Comparison (2 of 2) Text: Low-flow purging was also conducted using a peristaltic pump to compare the vertical distribution of chemicals within individual monitoring wells. Results from monitoring wells indicated similar vertical distribution of chemical concentrations using PDB sampling and low-flow sampling. However, several samples indicated lower concentrations in water collected using low-flow sampling techniques, which could possibly be attributed to loss of VOCs due to degassing during sample collection or to in-well mixing. VOC stratification was observed in monitoring wells. In some cases VOC concentrations increased with depth and in other cases concentrations decreased. The results indicated that when using PDB sampling it is important to select the proper sampling depth as concentrations can vary significantly with depth.

Title: Conclusions (1 of 2) Text: The diffusion samplers that were deployed in free-phase JP-5 jet fuel and Stoddard solvent did not show indications of integrity loss during the two months of deployment. Contaminant concentrations inside the passive diffusion samplers measured less than those of the free-phase product. However, this can be expected due to the fact that free-phase product is not in aqueous solution and the diffusion sampler does contain an aqueous solution. Click the 'View Site' button to view the Naval Air Station, North Island site. After the map zooms in, roll over the site map and click an area to zoom in on that part of the site map.

Title: Conclusions (2 of 2) Text: Contaminant concentrations obtained from passive diffusion samplers compared to samples collected using a bladder pump showed a general correlation. Some samples showed higher contaminant concentrations in the diffusion samplers, which may be attributed to mixing in the well and volatilization from bladder pump use. Based on the results, diffusion sampling is a viable alternative to sampling using a bladder pump in most wells at NAS North Island. Low-flow pumping using a peristaltic pump was used to show vertical distribution of contaminants at multiple depths within monitoring wells and to compare them to diffusion samples collected at the same depths. Again, lower contaminant concentrations were observed in some of the low-flow purge samples due to mixing and volatilization during sample collection. However, sampling results supported the use of diffusion samplers to demonstrate chemical stratification within a monitoring well.

Title: References Text: For more information regarding PDB samplers, please visit the NFESC ERB website: Diffusion Sampler ERB Webpage References Battelle. 2002. Final Diffusion Sampling Plan for Long-Term Environmental Monitoring at Operable Units 1A, 3, and 4 at Former NAWC, Warminster, PA. January. Battelle. 2002. Final Quarterly Monitoring Report for the February 2002 Performance Monitoring Event (Second Quarter FY 2002) for Long-Term Environmental Monitoring at Operable Units 1A, 3, and 4 at Former NAWC, Warminster, PA. July. Environmental Security Technology Certification Program (ESTCP). 2004. Demonstration and Validation of a Regenerated Cellulose Dialysis Membrane Diffusion Sampler for Monitoring Groundwater Quality and Remediation Progress at DoD Sites . Interstate Technology & Regulatory Council Diffusion Sampler Team. 2004. Technical and Regulatory Guidance for Using Polyethylene Diffusion Bag Samplers to Monitor Volatile Organic Compounds in Groundwater. February. USGS. 2001. USGS User’s Guide for Polyethylene-Based Passive Diffusion Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 1: Deployment, Recovery, Data Interpretation and Quality Control and Assurance USGS. 2001. USGS User’s Guide for Polyethylene-Based Passive Diffusion Samplers to Obtain Volatile Organic Compound Concentrations in Wells Part 2: Field Investigations. USGS, Naval Facilities Engineering Service Center, Battelle. 2003. Pre-Demonstration Test Plan for Demonstration and Validation of a Regenerated Cellulose Dialysis Membrane Diffusion Sampler for Monitoring Ground-water Quality and Remediation Progress at DoD Sites. October 19. USGS, Southwestern Division Naval Facilities Engineering Command. 2000. Diffusion Sampler Testing at Naval Air Station North Island, San Diego County, California, November 1999 to January 2000.

Title: Contact Information Text: For more information about Passive Diffusion Sampler Tools, please contact: NFESC POC (805) 982-1656 PRTH_NFESCT2@navy.mil