In last month’s issue of Focus on the Environment, we discussed various ways of collecting or measuring air and soil gas to assess vapor intrusion (VI), or to locate subsurface contamination sources. The below table summarizes the capabilities, advantages, and applications for each method. Obviously, the ratings on the table are subjective, so I’ll explain my rationale and elaborate on merits and limitations of various systems. So, from top to bottom, starting with capabilities:
Constituents of Concern: All of the methods are capable of measuring VOC concentrations, but except for some of the lighter SVOCs, such as naphthalene, only sorbents are capable of collecting the heavier SVOCs, including PAHs.
Sample Duration: All of the electronic instruments provide real-time readings, but a hand-held PID/Multimeter typically would not provide continuous data over a period of days or weeks. A field GC or GC/MS can be configured to collect continuous data and to relay information or warnings via internet, but the cost is prohibitive for many or most projects. Advances in technology and reduction in cost will make this more feasible over time.
Grab samples, i.e., samples (or readings) of a few seconds or minutes in duration, can be collected by all methods except for passive sorbents. Passive sorbents are designed to absorb chemicals over longer durations, typically days or weeks. Time-integrated results provide average concentrations over an extended period. Tedlar® bags, glass vials, and sorbents collected via syringe are completed in a few seconds or minutes, and provide instantaneous data. The same is true of field PIDs and most portable GCs.
Other: All of the methods except passive sorbents routinely provide concentration data, i.e., results in terms of mass/volume, such as micrograms/cubic meter (ug/m3) or parts per billion volumetric (ppbv). Passive sorbents provide only a chemical’s mass, and the volume of air it represents must be estimated based on uptake (absorption) rates for each chemical. Uptake rates might not be known for all chemicals, and they might vary due to moisture and other variables. For soil gas, there also exists a risk of the starvation effect, in which a sorbent absorbs chemicals faster than the surrounding soil can replenish them, and the results will be biased low.
Power: Methods that require no electrical power can be operated anywhere. Battery-powered devices can also be used in remote areas, but batteries will have to be replaced or recharged occasionally, and some devices must be adjusted periodically as battery power declines. Tedlar® bags should not be filled via pump, due to the risk of cross contamination from the pump. Instead, the bags are usually placed in an air-tight “lung box”, and air is pumped from the lung box, causing the Tedlar® bag to fill.
Data Quality: Canisters evaluated by Method TO-15 are still regarded as the gold standard, and any other approach will probably have to be verified against TO-15 results for VI assessments. The PID scores lowest because it can provide consistent results, but they are only accurate after applying a compound-specific correction factor for the primary constituent, if known.
Discriminates Compounds: The PID has no ability to distinguish between compounds. Other electronic field instruments can recognize compounds with a high level of certainty, but they are usually calibrated for a limited number of compounds, compared to fixed equipment in laboratory settings.
Low Reporting Levels: The lowest score goes to PID/multimeters, because most cannot detect vapors at VISL concentrations, even in soil gas. However, we find that they are fully capable of locating source areas in commercial/industrial settings. Sorbents are capable of achieving the lowest reporting levels, especially for SVOCs, if one draws more air through the sorbent or leaves it in place for a greater duration. But exercise caution, because too large a sample or too long a duration can ruin sorbent samples.
Minimal Container Interference: Because Tedlar® bags are made of plastic, they tend to emit some chemicals and absorb others. They also have the shortest holding time of any laboratory sample, which is why they scored lowest. Glass containers score high because they are disposable and impermeable. However, some laboratories might reuse glass containers, so if cross contamination is a major concern, discuss this with the lab. Electronic field devices typically don’t use containers, so there’s no risk of container interference, which is good. (Arguably, they should get no ranking, so they were given three open bullets).
Regulatory Acceptance, USA: As mentioned earlier, canisters remain the gold standard in the eyes of most regulatory agencies in the U.S. Sorbents are widely used and accepted in Europe for some constituents, but their suitability for soil-gas and for multiple chemicals is lower. Vial samples are not widely used or widely accepted, in part because they are limited to grab sampling. But because soil gas does not vary greatly over the course of several days, the advantage of time-integrated soil-gas samples is limited. The reporting levels for glass vials might not meet some soil-gas VISLs, but because VISLs greatly exaggerate VI risk in most commercial/industrial settings, vial samples are, in my opinion, useful for soil-gas sampling. Moreover, vials are ideal for locating contaminant sources, where concentrations are typically far higher than VISLs.
Fast to Implement: Canisters don’t score well because they first must be transported from the lab, and in many cases, clean-certified prior to shipping. Use of the more complex electronic equipment is generally subject to the availability of equipment and qualified users, and equipment might require a fair amount of preparation prior to use. Glass vials, sorbents collected via syringe, and PIDs can be deployed with short notice, and unlike passive sorbents, they do not require two field deployments – one to install them and one to retrieve them.
Availability: Canisters and Tedlar® bags are widely used in the U.S., and many labs analyze them. PIDs are also widely available and most environmental consultants already have one. Unfortunately, only a limited number of laboratories analyze glass containers for air.
Simplicity: The GC/MS has distinct virtues, but simplicity isn’t one of them, so it, and fixed GC devices scored lowest in this category. Canisters, sorbents collected via pump, and field GCs and are simpler, but glass vials, Tedlar® bags, and sorbents collected passively or via syringe have few or no moving parts and take first place for simplicity and ease of use. PIDs/multimeters are more complex, but most environmental workers routinely use them, so they were also awarded three bullets.
Dependability: All of the devices are reasonably dependable, if used in the appropriate setting, so none were scored with a single bullet. But any of them can provide poor or no data. Canisters are prone to clogging and leaking, and residual contamination is a risk, especially for indoor air. Most of the problems can be avoided by using clean-certified canisters from a dependable lab, and conducting shut-in tests and leak tests prior to field work. Tedlar® bags aren’t prone to mechanical failures, but absorption and outgassing of chemicals can render the results unusable, as can the limitations of short holding times.
We have had no failures using glass vials. On one project, the analytical results included detections of acetone in every sample they ran that day, but they didn’t interfere with our constituents of concern. Acetone is a common lab contaminant, and it shows up in other types of samples as well.
Sorbent samples collected via pump can fail if the batteries run out, but this is more likely when collecting indoor-air over an extended period than with soil-gas sampling, which may be collected over a shorter period. Sorbent samples collected passively or via syringe are pretty much immune to mechanical problems but all sorbent samples are subject to failure if the sorbent doesn’t match the constituents of concern, or if the sorbents become saturated for one or more chemicals.
Electronic devices can also fail, but the risk is greater for complex equipment, such as a portable GC or GC/MS, than for a simpler device, such as the PID/multimeter. The PID scored better because most failures can be resolved in the field, and if not, a replacement is usually readily available.
Low Cost: Because highly complex equipment is expensive to own and operate, the field GC/MS and fixed GCs and GC/MSs received a low score, with the understanding that they’re a bargain if nothing else can resolve key problems. A PID/multimeter isn’t cheap, but most field workers already own one, and operating costs are low, so it received a high rating.
Requires Minimal Skill: Again, the field GC/MS and fixed GCs and GC/MSs score low due to their complexity. Tedlar® bags, glass vials, and sorbents used passively or with a syringe score high since they have few or no moving parts. PIDs/multimeters are somewhat more complex, but familiar to most field operators, so they also received high marks.
Requires Minimal Site Familiarity: Because of the need to match sorbents to the constituent of concern, and problems caused by exposing sorbents to too much or too little contamination, sorbents might not be the first choice to evaluate unfamiliar sites, unless they are regarded strictly as screening tools. Devices that collect actual air samples, i.e., canisters, vials, and bags can be analyzed with minimal knowledge of site conditions, so they scored well. Field PID/multimeters also scored well because they are generally usable in all conditions.
Ease in Shipping/Holding Time: Canisters have a 30-day holding time and need no refrigeration, which is good, but they’re awkward and costly to ship, so they were rated low. Tedlar bags® are cheaper and easier to transport, but due to their very short holding time and risk of damage in flight, they were also ranked at the bottom. Glass vials and sorbents received higher ratings due to their compactness, but vials were rated higher since they don’t need refrigeration. Electronic devices generally don’t need to be shipped, which is a plus, and earned them three open bullets.
This is where the pluses and minuses of each approach are combined to generate even more subjective overall scores. Thus far, we’ve focused primarily on soil gas, but here we’ll add some comments on indoor air, which must is often collected with soil gas. Here’s how I ranked them:
Soil-Gas Vapor Intrusion: Canisters get the highest mark, because despite their bulk and potential mechanical problems, they provide the most reliable data in all settings at a reasonable cost. Furthermore, the same type of 6-liter canisters are used for soil gas and indoor air. Tedlar bags can be used for soil gas, and they are capable of meeting soil-gas VISLs for most compounds, but their short holding time is a problem, and their use might be necessitate an onsite lab. Vials, in my opinion, would be fine for VI sampling – at least in commercial/industrial settings where subslab attenuation factors are generally far lower than 0.03. Unfortunately, vials have little or no support from regulatory agencies, so they were ranked low for VI analysis.
Sorbents are gaining favor for VI assessments, but questions regarding uptake rates might make passive sorbents inappropriate for comparing concentrations to screening levels, especially for less-commonly analyzed chemicals.
Again, the PID/multimeter is the most essential of the electronic instruments for VI assessments, but only as a supplemental tool, since its reporting level is far higher than soil-gas VISLs, and it cannot discriminate between compounds. Portable and fixed electronic instruments can be used for VI assessments, but usually require verification via canister samples.
Indoor-Air Vapor Intrusion: Despite their bulk, cost, and complexity, canisters remain the standard for VI assessments, especially for indoor air, so they are ranked at the top. But while soil gas is sometimes collected instantaneously as a grab sample, and sometimes collected over the course of hours as a time-integrated sample, indoor air is rarely collected as a grab sample, which precludes the use of Tedlar® bags, glass vials, sorbents filled via syringe, and hand-held PIDs. Limited data quality or insufficiently low reporting levels also make these methods inappropriate for indoor air.
Sorbents collected via pump can provide time-integrated samples and vapor-concentration data. But in the U.S. they are less familiar than canisters, and they are only gradually being accepted by regulatory agencies, so they received an intermediate score. Issues involving sorbent saturation and the need to match the sorbent to constituents of concern still exist, but to a lesser degree than for soil gas. Passive sorbents have the same limitations, together with questions concerning uptake rates and the ability to determine chemical concentrations from chemical masses.
The field PID/multimeter is of limited use for indoor-air assessments, since most compounds can be detected only at concentrations vastly higher than their VISLs. However, as the only instrument that indicates real-time LEL conditions, it’s useful even indoors, and it received a single bullet. Because a portable GC/MS, such as the Hapsite, has the ability to distinguish chemicals at low concentrations in real time, it has proven very useful for locating interior (background) sources of contamination, but due to its high cost and the need for an expert operator, it was ranked low overall. A fixed GC or GC/MS, especially one equipped to collect indoor-air data from various parts of the building, concurrently with soil gas, can excel at solving VI riddles. But due to its high cost, the need for an expert user, and the fact that its results must be verified with canister data, it will not be appropriate for most VI assessments, and it scored two bullet points.
Source Prospecting: By source prospecting, we mean collecting data from multiple soil-gas points to locate areas of high-level contamination, usually in commercial/industrial settings. It is not necessary to detect vapors at VISL levels, and certainly not at residential VISLs, to locate most sources.
Summa-type canisters can collect soil gas for source prospecting using subslab sampling methods, but they’re overkill. The low reporting levels and time-integrated samples that canisters can achieve are unnecessary, as the task can be accomplished by other means for less cost and effort. On the other hand, if the area of interest is beneath an occupied building, a VI assessment is probably needed, and canister data could serve both needs.
Tedlar® bags are up to the task of locating sources, and the possibility of chemical interference is less of a problem when dealing with high-concentration vapors near sources. But their short holding time and the need for a lung box to fill them result in a low overall score for bag samples.
In my experience, nothing compares to vial samples for locating sources. Their compactness, 30-day holding time, low cost, simplicity, and ability to distinguish compounds at near-VISL levels makes them perfect for locating contamination sources. We have often screened for contamination beneath building floors or exterior pavement with a PID/multimeter, supplemented by glass vial samples from a smaller number of hot spots. The combination of field screening and limited lab analysis enables us to assess large areas at relatively low cost, and it provides concentration data for approximately 50 chemicals. And again, the ability to measure low O2 and elevated LEL in soil gas, which often indicates the breakdown of non-chlorinated compounds near sources, is unique to the PID/multimeter.
Sorbent samples are well suited to source prospecting. Soil-gas concentrations aren’t especially critical for locating gross contamination – it’s more important to figure out which chemicals are present and where concentrations are highest. Time-integrated data isn’t needed either, so it generally makes more sense to fill sorbents with a syringe than a pump. Passive sorbents have the additional advantage that they can be collected in unpaved areas, but on the down side, they provide no real-time data, and they can only be evaluated by paying for laboratory analysis at each and every sample location.
Portable GCs could be useful for source prospecting, but because their high cost, the need for calibration of the proper compounds, and their need for expert users, they were awarded only two bullet points. The Hapsite or equivalent field GC/MS could be used for source prospecting, but its high cost and need for expertise makes it a poor choice overall. Of course a fixed GC or GC/MS would not be used to measure contamination in multiple locations, unless it was outfitted with a lot of tubing and some kind of manifold system.
There are no magic bullets. Each of the methods is well suited to some situations and not others, depending on the constituents of concern, vapor concentrations, site familiarity, budget, and other factors. Understanding the differences will help you select the most appropriate method to answer the question at hand.
Published in Cox-Colvin’s November 2018 Focus on the Environment newsletter.
Mort Schmidt is a Senior Scientist with Cox-Colvin & Associates, Inc. He received his BS and MS degrees in Geology and Mineralogy from The Ohio State University, and has been a Cox-Colvin & Associates employee since 1997. His areas of expertise include vapor intrusion and contaminant investigation and analysis, and he currently serves as Cox-Colvin's Practice Leader - Vapor Intrusion Services. Mort is a Certified Professional Geologist with AIPG and is a registered Geologist in Indiana. Craig Cox is a principal and co-founder of Cox-Colvin & Associates, Inc., and holds degrees in geology and mineralogy from the Ohio State University and hydrogeology from the Colorado School of Mines. Mr. Cox has over 30 years of experience managing large environmental project implemented under CERCLA and state voluntary action programs. In addition, Mr. Cox has developed a variety of software products including Data Inspector, an internet-enabled environmental database application. Mr. Cox is a Certified Professional Geologist (CPG) with AIPG and is a Certified Professional (CP) under Ohio EPA's Voluntary Action Program.