«RATIONALE FOR PRESERVATION OF SOIL AND SEDIMENT SAMPLES FOR DETERMINATION OF VOLATILE ORGANIC COMPOUNDS OCTOBER 2005 Connecticut DEP QA/QC Workgroup ...»
RATIONALE FOR PRESERVATION OF SOIL AND SEDIMENT
SAMPLES FOR DETERMINATION OF VOLATILE ORGANIC
Connecticut DEP QA/QC Workgroup
Given the large number of manufacturers and vendors of scientific equipment, glassware,
reagents, and supplies, it is not feasible to cite all possible sources for these materials. Thus,
the mention of trade names or commercial products in this paper is for illustrative purposes only, and does not constitute an endorsement or exclusive recommendation for use by the State of Connecticut or the authors. Glassware, reagents, supplies, equipment, etc., other than those listed in this paper may be employed provided that method performance appropriate for the intended purpose is achieved.
BACKGROUND Since the early 1980s, environmental laboratories have routinely used the dynamic headspace, or purge and trap, procedure as a means of concentrating and introducing the volatile organic compounds in a sample into a gas chromatograph for analysis. Soil and sediment samples (hereafter referred to as soil samples) were typically collected in the field, placed in glass jars sealed with Teflon® liners and solid caps, and sent to the laboratory for analysis. This is referred to as the “dirt in a jar” procedure. Once in the laboratory, samples would be held at 4º C until analysis. The standard holding time for VOCs in soil samples is 14 days from collection, meaning that the samples must be analyzed in this time-frame or the results are considered invalid due to losses of the volatiles in the sample.
On the day of analysis, samples would typically be brought to room temperature. A sub-sample would be weighed out in a fume hood into a sparging flask or test tube, which would then be attached to the purge and trap instrument. Water, surrogates, and internal standards would be added, and the sample would be purged to transfer the volatiles onto the trap. The trap would then be ballistically heated, and the volatiles determined using either a gas chromatograph (GC) or a gas chromatograph/mass spectrometer (GC/MS). The focus of this paper is on the steps that occur prior to the purging of the sample.
Based on informal surveys with Connecticut-based laboratories, approximately 50% of the soil samples currently submitted for VOC determination are “dirt in a jar”-type samples, with no preservation other than cooling at 4º C.
As the environmental testing industry acquired more experience and expertise, a phenomenon was noted that samples classified as “hot” in the field, either due to field screening or odor, would come back from the laboratory with only low levels (200 ppb) of VOCs present. People began to suspect that the current sampling and sample storage procedures were not adequate.
In May 1993, EPA published a report entitled, Behavior and Determination of Volatile Organic Compounds in Soil: A Literature Review (1). This report noted that substantial losses of volatile organics occur when samples are stored at 4º C in only a few days. Furthermore, laboratory handling of the sample can also create substantial losses in VOC concentrations, averaging 60%.
Studies have shown that VOCs are also subject to microbial degradation even when stored at 4º C (2). The combination of losses due to poor sample seals, headspace, and microbial action will lead to rapid loss of VOCs in general, but even if the sample containers are completely sealed, significant losses of certain analytes, notably the aromatics such as benzene, toluene, etc., will occur due to microbial degradation (2). Thus, in order to store samples for any appreciable length of time, it was determined that the samples should be sealed from the atmosphere and steps taken to limit any biological activity.
Sample collection and field handling were also determined to be critical issues. In 1996, Alan Hewitt and Nicole Lukash of the U.S. Army Corps of Engineers Cold Regions Research and Engineering Laboratory (CRREL) conducted a study (3) in which various types of sample collection devices and techniques were evaluated. The results of this study are significant. Some
key findings were as follows:
1. When sampling soils for VOCs it is important to keep the sub-sample soil structure intact (i.e.
the sample should remain as a plug, and not be broken apart).
2. Soil samples exposed to air for two minutes in a sealed plastic bag showed a 90% loss in certain analytes (e.g. trichloroethene or TCE). This indicates that any handling of a soil sample with spoons, trowels, or any type of field compositing will result in significant losses of volatile analytes.
3. Samples collected using some type of coring device (e.g. a plastic syringe with the tip cut off), allowed for an undisturbed plug of sample to be collected and easily added to a 40-milliliter (ml) volatile organic analysis (VOA) vial.
4. Samples collected in core liners and uncapped showed significant losses in less than 40 minutes (e.g. 90% loss of TCE). Samples collected in core liners and capped with either Teflon® or aluminum foil had significant losses in 5 days.
5. Samples collected with cut-off syringes and added to VOA vials containing 5 mls water, with a Teflon®-lined, butyl rubber septum showed only slight losses (5-15%) when stored at 4º C after 28 days.
After the results of this study were published, the first commercial sampling device (the En Core® sampler) was introduced. This device uses Viton® o-rings to provide a hermetic seal that will prevent losses of volatiles due to evaporation. However, due to potential microbial degradation, samples must be transferred to a medium that will inhibit degradation (such as sodium bisulfate or methanol) or frozen at -12º C within 48 hours of collection.
As a result of these and other studies, EPA developed and published Method 5035. This method calls for collection of samples using a coring device and either immediately transferring the sample to a pre-weighed VOA vial (a vial that uses a Teflon®-lined, rubber or silicone septum as a seal) containing either water, a sodium bisulfate solution, or methanol. Samples collected in water must either be analyzed within 48 hours of collection or frozen to prevent microbial degradation. The bisulfate solution has a low pH, which effectively stops any microbial degradation. The methanol has two functions. It stops the microbial degradation, and the volatile organic compounds are soluble in it. Thus, samples collected in methanol can be opened for short periods of time (e.g. less than 1 to 2 minutes) in the laboratory without fear of analyte loss.
Samples collected in bisulfate or water should not be opened, as analyte loss will rapidly occur.
The typical holding time for soils preserved with either methanol or sodium bisulfate is 14 days from collection.
Another option is collecting the sample using a coring device such as the En Core® sampler, and shipping the device to the laboratory. At the laboratory the sample can be transferred to the VOA vial (containing bisulfate, water, or methanol). Regardless of whether the sample is collected in methanol, bisulfate, or kept in the En Core®-type device, the sample may be frozen to extend the holding time.
Sample Collection Considerations
The first step in obtaining meaningful volatile organic compound concentrations is to obtain a sample that is representative of in-situ conditions. It is obvious that the major problem with collecting samples for this class of analyte is that they are volatile, and easily lost from a sample.
Therefore, the following steps should be taken to limit any losses of analytes:
1. The sample should not be exposed to air. Exposure of a minute or two will result in significant losses. VOCs will be lost even in a sealed container if there is headspace present and the container is opened. In general, compositing of soil samples for volatile organics is not recommended.
2. A coring device should be used so that the structure of the sample is maintained. Coring devices can be syringes, En Core®-type devices, etc. The device should be able to extrude the sample in a plug, thus preserving the soil structure.
3. Samples should be collected as soon as possible after the soil is exposed. Any delay in collecting the sample will result in losses of VOCs. Prior to sampling, the technician should scrape away several inches of soil to obtain a fresh sample. The coring device should be inserted directly into the soil.
4. If samples are to be chemically preserved, the preservation should occur immediately after collection. The container must already contain the preservative, and then sealed immediately after the sample is added. The chilling process must then be begun immediately. Samples should not be collected from one area of the site, brought to a central staging area, and then added to the storage container.
5. The contact surfaces of the sample container must be cleaned of all soil particles to insure a hermetic seal. Any sample received at the laboratory without a good seal should be rejected.
In October 1998, the U.S. Army Corps of Engineers (USACE) published a paper describing sample collection and preparation strategies (4). This paper goes into much detail and rationale for collecting soil samples for volatiles and is strongly recommended reading.
An Appraisal of Preservation Options It has been shown that collection of soil samples using the “dirt in a jar” method will lead to significant losses of volatile organic analytes. This section describes the various preservation options currently available and their strengths and weaknesses.
Preservation with Sodium Bisulfate Sodium bisulfate is an acidic salt, which in aqueous solution lowers the pH to inhibit any microbial degradation of the VOCs. Many chemical suppliers offer VOA vials with a sodium bisulfate solution and a small magnetic stir bar already present. Samples collected in the field can be added directly to pre-weighed vials with little difficulty. The vial is then either weighed in the field or back at the laboratory to obtain the exact sample weight. A separate container must be submitted for percent solids determination. The concentration range, assuming a 5-gram sample, is 1 to 200 μg/kg (ppb). Once the sample is placed in the vial, and the vial sealed, it is not opened until after the analysis is completed. Assuming the vial is sealed correctly and stored at 4º C, there will be either minimal or no loss of analytes.
There have been some identified weaknesses with this procedure. Regardless of the type of sample container used, the contact area (e.g. the threads and top of VOA vials or the o-rings and contact surfaces of En Core®-type devices) must be free of soil particles. An airtight seal is critical. Being acidic, the sodium bisulfate solution will corrode the laboratory purge and trap unit over time. This is not a significant issue, but may require addressing by the laboratory every 3 to 5 years. Since the sample can only be analyzed once per vial, samples are usually collected in duplicate in case of breakage or other problems.
Perhaps the most significant issue is that certain naturally occurring compounds (humic acids, etc.) will decompose when exposed to the bisulfate solution and form ketones, notably acetone.
The amount of acetone formed is extremely matrix dependent, but may be produced in significant concentrations. When using sodium bisulfate as a preservative, the data user must keep this in mind when evaluating the data. Also, it has been found that acidification with sodium bisulfate causes the loss of styrene (5). Therefore, if styrene is a compound of concern, sodium bisulfate should not be used.
Another significant issue with sodium bisulfate is for samples which react with the acid and effervesce. The effervescing will result in significant losses of volatile organics, and in such cases the sodium bisulfate cannot be used.
Preservation with Methanol Methanol, or methyl alcohol, is commonly used for preservation of soil samples. Equal amounts of soil mass and methanol volume are commonly used (e.g. 10 grams soil added to 10 mls methanol). Again, a pre-weighed VOA vial containing methanol would be sent to the field. A sample would be extruded into the vial, the vial sealed, and reweighed either in the field or back at the laboratory. A separate container must be submitted for percent solids determination.
Aliquots of the methanol extract are analyzed using the same procedure as for aqueous samples.
(e.g. a 100-μL aliquot of the methanol is added to 5 mls of reagent water and analyzed the same way as a water sample). The concentration range, assuming a 10-gram sample and 10 mls methanol, is approximately 250 to 10,000 μg/kg (ppb). Since the methanol extract can be diluted, the upper range can be extended further.
The most serious limitation with this procedure is the relatively high detection limit. To counteract this, most projects call for collection of both sodium bisulfate-preserved and methanol-preserved samples from the same sampling point. If the sample has high concentrations, the methanol-preserved vials are analyzed. If the sample is low, then the bisulfate-preserved vial is analyzed.
The sample matrix itself can affect the solubility of the volatile organics in methanol. It might take several hours for concentrations to stabilize (5). Some procedures call for sonicating the vial containing the methanol/soil mixture for 30 minutes prior to analysis. Samples immersed in methanol should be sonicated at 40º C for 30 minutes as a precaution.