Current approaches for long-term groundwater monitoring programs rely on water sampling and analysis using traditional decades-old protocols that are time-consuming and costly. Complying with the requirements of these monitoring programs comprise a significant portion of life-cycle remediation costs for the Department of Defense (DoD). This project involved basic research on an alternative groundwater sampling approach—vapor-phase groundwater monitoring—that relies on a different set of physical processes and analytical instruments to provide reliable and accurate long-term monitoring for volatile organic compounds (VOCs) more rapidly and at a lower overall cost. The objective was to evaluate the utility of on-site vapor-phase analysis of samples from a groundwater monitoring well as an alternative to off-site analysis of groundwater samples.

All investigations completed as part of this project were designed to test the principle that the VOC concentration measured in a vapor-phase sample in equilibrium with affected groundwater can be used to accurately determine the VOC concentration in the associated groundwater at or below maximum contaminant levels (MCLs). Two key hypotheses were developed to support this principle: (1) Portable vapor-phase monitoring instruments can be used to accurately determine VOC concentrations in water under equilibrium conditions; and (2) In-well mixing is sufficient in some or all groundwater monitoring wells to establish equilibrium partitioning conditions between affected groundwater and in-well headspace vapors. These hypotheses were tested through a series of laboratory and field-based programs, consisting of a laboratory-based study to validate analytical equipment and to identify promising methods, three distinct phases of field-based studies to test various sampling and collection methods and to examine design and well-specific factors that influenced performance, and a combined modeling-field study that focused on the influence of seasonal temperature gradients on vertical stratification of concentration within monitoring wells. A variety of vapor-phase sampling and/or analysis techniques were tested, including direct sampling and analysis from the headspace of a capped monitoring well; several different permutations of submerged passive vapor diffusion samplers, all of which are gas-permeable but water tight; and “field equilibration” of groundwater (collected using low-flow techniques) in a vial, followed by on-site analysis of the equilibrated headspace. A combination of quantitative methods was used to evaluate vapor-phase based concentration data to more conventional (baseline) groundwater concentration data. These evaluation methods and metrics included linear regression, relative percent difference, coefficient of variation, ANOVA, and parametric and non-parametric statistical tests for significance. The vast majority of the validation data were collected in the field, with approximately 1,100 concentration datapoints collected during the various field programs.

The project findings confirmed that existing field-portable vapor-phase monitoring equipment are sufficiently accurate, precise, and sensitive for calculating equivalent VOC concentrations in groundwater. Specifically, a field-portable gas chromatograph (GC) demonstrated the highest performance of the analytical devices that were tested. Alternative field instruments for vapor-phase analysis—a simple PID-based handheld meter and the HAPSITE with GC/MS capabilities—were also tested during one or more of the field programs. These instruments did not perform as strongly as the field GC with respect to accuracy and precision, although the HAPSITE did prove useful in terms of identifying a higher number of constituents at lower detection limits. Vapor-phase sampling and analysis methods proved easy to implement and can be tailored to site-specific needs, including multi-level sampling. Collecting vapor samples from the well headspace was not an effective method for determining groundwater concentrations under the tested conditions. In part, this was due to the influence of some degree of vertical stratification of concentrations within the well network, such that the vapor sample collected from the well headspace was in equilibrium with water that was typically not representative of the water collected for low-flow sampling.  Instead, low-flow groundwater concentrations could be most reasonably estimated using submerged passive vapor diffusion samplers or field equilibration of collected groundwater. Because these latter two methods collect samples within the screened interval of the well, they are not as reliant on in-well mixing to overcome stratification as is the simpler headspace method. A combination of modeling and field data were used to show that seasonal temperature gradients have the potential to contribute significant variability to monitoring data, including both conventional and vapor-phase based methods. In particular, they can promote or diminish vertical stratification within the well during different periods. Of the other well and aquifer-specific factors that were investigated, only the presence of a confining aquifer significantly contributed (negatively) to variability. A year-long, multi-event evaluation demonstrated that vapor-phase based monitoring methods are no more variable than conventional groundwater monitoring methods, with both types subject to similar spatial and temporal variability that can be difficult to reduce.

What Was Learned 

• Existing commercially available field-portable vapor-phase monitoring equipment are sufficiently accurate, precise, and sensitive for calculating equivalent VOC concentrations in groundwater down to part per billion levels.
• VOC groundwater concentrations can be reasonably and reliably estimated using submerged passive vapor samplers. Both a simple passive vapor sampler constructed of a 40-mL vial in plastic and the Haas Balloon Sampler worked well.  Field equilibration of conventional collected groundwater samples followed by on-site vapor analysis using a field GC also worked well. 
• A field-portable GC demonstrated the highest performance of the analytical devices that were tested. Simple PID instruments did not work well for this application.
• Vapor-phase sampling and analysis methods are easy to implement and can be tailored to site-specific needs. 

What Doesn’t Work

• Collecting vapor samples from a sealed monitoring well headspace was not an effective method for determining groundwater concentrations under the tested conditions due to stratification in wells.
• Vapor-phase based monitoring methods are no more variable than conventional groundwater monitoring methods, including low flow sampling. 

Things to Watch Out For

• Although not a strong factor in this study, seasonal temperature gradients have the potential to significantly alter monitoring data, including both conventional and vapor-phase based methods.
• Vertical stratification can be an important contributing factor to variability and limits the utility of the well-headspace vapor-phase based monitoring approach.
• Other well and aquifer-specific factors can contribute to variability and influence the performance of vapor-phase based monitoring methods.

Key Conclusions

• Passive vapor sampling methods represent a promising approach for field-based estimation of groundwater concentrations.
• Vapor-phase based methods represent a significant cost savings (36% or more) relative to conventional groundwater monitoring approaches.

The development of reliable vapor-phase based monitoring approaches is designed to aid DoD with several key goals in long-term monitoring optimization. First, it entails a less cost and time-intensive method for analyzing specific contaminants of concern, including most chlorinated hydrocarbons. Further, it can utilize inexpensive and cost-effective tools during the data collection process. Finally, it represents a simple approach that would be easy to implement at a majority of DoD sites nationwide. All of these factors work to significantly reduce the cost liabilities associated with groundwater monitoring while providing a more sustainable long-term approach.

Extensive cost modeling demonstrated that groundwater monitoring could be completed at a cost savings of at least 36% when on-site vapor-based monitoring was completed using a rented GC. This represents a savings of several hundred dollars per sample for typical monitoring programs.  Sensitivity analysis was used to examine the impact of the number of samples per event and per well on overall cost.  In particular, using passive vapor samplers to perform multi-level monitoring (i.e., increasing the number of samples per location) shifts the economics sharply in the favor of vapor-phase based methods. The vapor-phase monitoring methods are straightforward and can be implemented by DoD and other stakeholders with limited additional training and expense. Consequently, there are no technical limitations for larger-scale use.