Main Content

VAPOR INTRUSION FUNDAMENTALS – TYPES OF AIR SAMPLES – PART 1

In the United States, EPA Toxic Organics Method TO-15 has long been regarded as the gold standard for analyzing Volatile Organic Compounds (VOCs) in air matrices. Method TO-15 specifies not only the analytical procedure (Gas Chromatography/Mass Spectrometry – GC/MS), but also the container type, specifically evacuated stainless-steel canisters, often called “Summa cans” (actually a brand name). Canisters are normally equipped with regulators to control the fill time, typically between 5 minutes and 24 hours. Flow regulators are complex, and prone to leaking, clogging, and flowing at the wrong rate, especially when subjected to rough handling. Additionally, canisters are bulky, typically ranging in size from ½ liter to 6 liters. A canister and regulator combination costs close to $1,000, so the laboratory’s inventory might be limited, and equipment must be cleaned and tested between samples, which the user sees as rental costs. But when canisters work, they work well, and they provide a direct indication of vapor concentrations in air or soil gas.

Glass vials and Tedlar^® bags are useful for soil-gas sampling, but they collect only grab samples, so they cannot be used for indoor air, which is normally collected over the course of 8 or 24 hours. Glass containers are especially useful for locating sources of contamination, but they might not provide sufficiently low reporting levels for VI assessments. Glass containers and Tedlar^® bags have the advantage of being disposable, which reduces the risk of cross contamination, but vials have the longer holding time.

Sorbent devices differ in that they don’t actually collect air. Sorbents collect chemicals from the air (or in some cases water), onto activated carbon, polyurethane foam, or other media, making them essentially air samples, but without the air.

There are two methods for sorbent sampling – active and passive. The most common active sorbent samples consist of ¼-inch diameter stainless-steel tubes (shown below). Air may be drawn through them with a pump or syringe. For VI purposes, sorbent samples are most often analyzed via EPA Method TO-17, which, like TO-15, includes analysis via GC/MS.

Air is not drawn through passive sorbent samples. Passive sorbents simply soak up chemicals from surrounding air via molecular diffusion. In Europe, passive sorbents have been used for years to sample indoor air, but the use of passive samplers for soil gas is relatively new. There are several designs, including a modified version of the TO-17 tube, but none of them can fit into the small-diameter hole typically used for subslab sampling.

Besides the advantages of compactness, simplicity, and low equipment costs, sorbent samples are superior to canisters for measuring VOCs at very low concentrations. Lower detection levels are achieved by drawing more air though active samplers or by increasing the exposure time of passive samplers. The ability to collect large air samples with sorbents also enables them to collect heavier semi-volatile organic compounds (SVOCs), which are often associated with Petroleum Hydrocarbons (PHCs). Sorbents also can be used to sample mercury vapor, which cannot be sampled via evacuated canisters. Passive sorbents are advantageous for long-term sampling, and can provide average concentrations over a period of two or more weeks, compared to evacuated canisters which seldom have a sampling duration greater than 48 hours.

However, sorbent samplers have serious limitations, one of which is that calculating the volume of air represented by sorbents is not necessarily straight forward. When a laboratory analyzes sorbent samples, they measure the mass of a chemical in the sorbent. With active sorbents, the sample collector tells the lab how much air was drawn through the sample, and the lab divides the chemical mass by the air volume to calculate concentrations. However, active samples are often collected with battery-powered pumps, and if the pump slows down due to battery depletion or filter clogging, the actual air volume might not be known.

In the case of passive sorbents, chemicals enter the sorbent through molecular diffusion, not air flow, and the lab must determine uptake rates specific to the chemical and sampling device. The lab analyzes a sample’s chemical masses, and factors in uptake rates to calculate concentrations, which again leads to uncertainties. On the plus side, because chemicals enter passive sorbents via molecular diffusion, passive sorbents can sometimes be buried in tight or wet soils that would not yield sufficient soil gas for active sampling. Consequently, passive samples can be buried in a grid configuration to locate contamination sources, but the results might be more qualitative than quantitative, and it might not be possible to evaluate them relative to screening levels or other limits.

Another limitation to sorbents is that they trap various chemicals differently, and it’s necessary to use the appropriate sorbent for the chemicals present. It’s also critical to expose the sampling device to the proper amount of air. If the sampler is exposed to too little air, some chemicals won’t be detected. If it’s exposed to too much air, the sorbent becomes saturated, and higher concentrations of some chemicals won’t register. These issues hamper the usefulness of sorbents at poorly understood sites. Moisture and temperature also affect the sorptive capacity of sorbents, which may necessitate the use of moisture and temperature corrections. These limitations can be avoided to some degree by collecting multiple samples with different sorbents, air volumes, and exposure times, but this could offset money saved by avoiding canisters. And while passive sorbents work well with heavier compounds, including some Poly-Aromatic Hydrocarbons (PAHs), they perform relatively poorly with some lighter compounds, especially vinyl chloride.

Perhaps the greatest limitation to passive sorbents is that they don’t lend themselves to the collection of real-time field data. Using common hand-held electronic instruments and conventional soil-gas sampling points, VOCs, Oxygen (O₂), Carbon Dioxide (CO₂), and Lower Explosive Limits (LELs) in soil gas can be measured instantly. VOC concentrations can be measured with varying degrees of precision, depending on the sophistication of the measuring device. Basic hand-held meters can provide levels of total VOCs, O₂, CO₂, and LEL, while portable gas chromatographs (GCs) or GC/MSs can provide concentrations of individual chemicals. The ability to collect real-time data is especially useful for source prospecting, because the area of interest can be modified or expanded in response to field data, without having to wait for analytical results.

Hand-held Photo-Ionization Detectors (PIDs) generally are not sufficiently sensitive to detect soil gas at EPA’s Vapor Intrusion Screening Levels (VISLs), but they are excellent supplemental tools that most consultants already own, and their sensitivity is adequate to locate major source areas. Some multi-gas meters combine the functions of PID, LEL, O₂, and LEL meters, and in addition to providing useful information on soil gas, they can be used to purge soil gas prior to sampling. The common approach of purging three volumes of soil gas prior to sampling is based on investigations in which soil gas was sampled after purging one, two, three, or more volumes of “dead” air in the sample train and sample point. The investigations found that in most settings, soil-gas concentrations stabilized after purging three volumes of “dead” air. A more direct approach is to monitor O₂ and PID readings while purging with a multi-gas meter, and sampling immediately after they stabilize.

More sophisticated instruments have the ability to detect vapors at lower concentrations and to distinguish between compounds. Field GCs, such as the Frog 4000 can distinguish between multiple compounds, and can reach lower detection levels than most PIDs. A portable GC/MS, such as the Hapsite, is capable of detecting vapors at even lower concentrations, and can detect most or all VI constituents at soil-gas VISLs. The GC/MS also has the potential to recognize compounds for which it was not calibrated, due to its ability to recognize spectral lines corresponding to the weight of molecular fragments. However, due to the limitations of working in field conditions, a portable GC or GC/MS is less reliable than a permanently-mounted lab instrument, and the results are regarded as screening data that must be verified with TO-15 samples. A portable GC or GC/MS is also costly compared to a multi-gas meter, and requires more expertise to operate. Additionally, because a new Hapsite costs over $100,000, they are typically rented, and they may be subject to rough use and maintenance problems.

Finally, a GC or GC/MS can be used in the fixed mode to provide data continuously over long periods of time. This can be useful for some chemicals such as trichloroethene (TCE), which can potentially cause health problems with short exposures. The ability to collect data continuously over long periods makes electronic monitoring capable of detecting short-term spikes in concentrations, but it is very costly and requires considerable expertise, making it unsuitable for most VI evaluations. Additionally, while the cost of continuous monitoring might sometimes be justified for sampling indoor air, it will rarely or never make sense for soil-gas sampling.

In next month’s issue of Focus on the Environment, we’ll discuss the capabilities, advantages, and applications of each method in greater detail, and provide a table that may simplify the decision-making process when deciding what kind of samples to collect.