12790 Seabreeze F.D. San Diego CA, 92130
Phone (858) 259-2252, FAX (858) 259-2384)
 |
 |
 |
 |
 |
 |
 |
|
INTRODUCTION
There is nothing under the sun that is new about soil sampling. That is to say, nothing new until EPA Method 5035 came to light (officially) in June of 1997. Until that time soil sampling for volatile organic compounds (VOCs) was pretty much an exercise of personal preference. The guidance was not so much on how you collected the sample but to make sure the sample jar had no headspace when you shipped it back to the lab. In fact when it comes to soil samplers it turns out that there are a plethora of them to choose from.
You could have Mechanical Sample Recovery such as Hand-held Power Augers, Solid Stem Flight Augers or Hollow-Stem Augers, Bucket Augers and even Backhoes. There are a variety of Non-Mechanical Samplers such as Screw-type Augers, Barrel Augers, Tube-type Samplers and yes our favorite the Bulk Sampler. In fact the Barrel Augers and Tube-type Samplers have over 16 varieties to choose from within just those two types of samplers. Unfortunately, none of these samplers address the issue of discrete soil sampling as referenced in EPA SW846, Method5035.
SCOPE OF VOC SOIL SAMPLING PROBLEM
By the mid to late '80s, questions were being raised about the accuracy of the VOC analyses. Were the laboratories actually testing representative samples from the field or were substantial losses occurring between the sample in situ and its arrival at the analytical instrument in the laboratory. Interested parties began to search for new ways to produce more accurate sampling and storage techniques as well as improved sample preparation procedures in the lab.
Studies began to show that significant losses of VOCs were occurring prior to arriving at the point of analysis. One study by Hanisch and McDevitt (1984) reported that if there were any headspace in the sample container, there would be VOC desorption from the soil particles into the headspace. Those VOCs would then be lost to the atmosphere when the cap was removed during subsampling for analysis at the laboratory. However, when compared to the analysis, handling activities and the sample collection itself introduce much larger errors (Barcelona, 1989). Negative bias, having a measured value less than the true value, is the most significant and most difficult to delineate and control. This error is caused primarily by loss through volatilization during soil sample collection, handling and storage (EPA, Lewis, Crockett, Siegrist, Zarrabi, 1991). It was determined by Siegrist and Jennsen (1990) that 81% of the VOCs were lost from samples held in glass jars sealed with Teflon™ lined caps compared, to samples immersed in methanol in jars.
SOLUTION TO LOSS OF VOCs
These were some of the factors that energized the designing of EPA SW846-Method 5035 and ASTM D4547-91. In short, Method 5035 calls for the preservation of soil VOCs in methanol. This of course meant that a hazardous substance (methanol) was going to be taken into the field. Methanol also created limitations for shipping samples to the laboratory. It was suggested that an alternative process should be made available to address the concerns of those who did not like the idea of using methanol in the field. The idea was promoted to develop short-term storage devices. The first device, a stainless steel design was tested by the EPA and accepted for 48-hour storage. This storage time would then allow most firms to ship their samples overnight to their respective laboratories for methanol preservation (and sonication, which is required in some states). SoilCore, Inc. has since then developed a disposable sampling/storage device, which substantially reduces costs and eliminates the need for cleaning the stainless steel sampler every time it is used.
Method 5035 also delineates samples as to high and low concentrations. For samples containing VOC concentrations greater than 200 ug/kg a 25g sample (or bulk sample or 5g sample)* is collected and preserved in methanol. For samples containing VOC concentrations less than 200 ug/kg a 5g sample** is collected and preserved in sodium bisulfate.
DISCUSSION
The key to the new sampling method though is being able to collect a soil sample without disturbing the soil matrix, which is otherwise known as a discrete sample. As soon as impacted soils are exposed to the atmosphere, soil gasses are released and volatilization of organic compounds begins. Siegrist and Jennsen (1990) demonstrated that when they used a discrete soil sample that was followed by immersion in methanol, the VOC losses were minimal. Therefore at least for most of us, accurate analyses begin with discrete sampling followed by methanol preservation (and someday, sonication). And while this writer has found very few field personnel who are happy with Method 5035, almost everyone agrees that discrete sampling and storage is much faster, easier and safer than discrete sampling with methanol preservation in the field.
The USEPA recommends and approves discrete soil sampling and storage. To that end SoilCore, Inc. has developed a simple yet advanced discrete soil sampler and storage device. Rather than having to buy two samplers when you need both a 5g and a 25g sample, the SoilCore™ Sampler provides both 5g and 25g on one sampler. This allows one cavity to be used for determining dry weight if you do not need both high and low level on the same sample. The combination of special manufacturing material and the internal configuration of the sampler allows for simplified extraction of the soil. By simply tapping it on the side of the methanol extraction jar the soil sample will easily fall into the jar.
* Check with your specific state to obtain appropriate sample collection data. ** The State of Michigan only accepts 25g samples preserved in methanol.
The summary of the procedure is that environmentally exposed sand was put into many samplers. Each sampler was then spiked with a stock standard solution in methanol, capped and refrigerated. A minimum of five samples were then put in four ounce jars, preserved with methanol and refrigerated. These would serve as the zero hour reference samples. The soil samples were then preserved in methanol after being stored for 63 hours in the SoilCore™ Samplers. Analyses were then conducted by EPA 8260. Table 1 compares the zero hour data to results after being in storage for 63 hours.
A total of 54 compounds were observed in this study and the recoveries ranged from 72.2% to 127.2%. 1,2,3-Trichloropropane had the lowest recovery while 1,1,2,2-Tetrachloroethane had a recovery of 127.2%. Three compounds had a recovery of less than 80%; 1,2,3-Trichloropropane 72.2%, n-Butylbenzene 75.4% and Isopropylbenzene at 78.2%. The only compound with a recovery greater than 120% was 1,2,3-Trichloropropane at 127.2%. There were 14 compounds having recoveries between 80.0% and 90.0%, and 5 compounds between 110.0% and 120.0%. Thirty-five of the compounds (65%) were between 90 and 110% recovery. The average recovery for all 54 compounds was 96.4%.
Three additional studies (Hewitt '98, Yazdani '98 and Yazdani '99) have substantiated, with their studies, that the 25g sampler has performed at a rate greater than 80% for recoveries from their zero hour analyses. As discrete sampler technology continues to evolve we anticipate the average recoveries will continue to increase. The current developmental goal is to take the design of the analytically perfect discrete soil sampler and figure out a way to make it easy to use out in the field.
One new area of development is independent studies being conducted by Dr. Alan Hewitt, SoilCore, Inc. and others, regarding freezing the samples in the SoilCore™ Sampler for up to fourteen days. If this freezing could be incorporated into Method 5035 it would go a long way in making the Method 5035 much easier to live with by removing that critical 48 hour period in which the sample must be preserved with methanol.
CONCLUSION
The most significant losses of VOCs occur during sample handling procedures. One of the best ways to greatly reduce those losses is to take the soil sample without disturbing the soil matrix. The best way to do that is to use a discrete soil sampler, such as the SoilCore™ Sampler, so that the soil matrix is not disturbed. The soil is then stored in the manner and for the length of time as described in Method 5035.
Table 1.
Percent Recovery Of Analytes From SoilCore Sampler.
Results of soil spikes after storing in SoilCore Sampler, using methanol preservation(ug/Kg).
Percent Recovery Of Analytes From SoilCore Sampler.
Results of soil spikes after storing in SoilCore Sampler, using methanol preservation(ug/Kg).
| VOCs By GC/MS - 8260 | |||
| Methanol Extract | 0-HR | 63-HR | Percent Recovery |
| Trichlorofluoromethane | 53.9 | 49.8 | 92.9 |
| 1,1-Dichloroethene | 114.5 | 115.0 | 100.4 |
| Methylene Chloride | 436.5 | 413.8 | 94.8 |
| Trans-1,2-Dichloroethene | 191.3 | 177.0 | 92.5 |
| 1,1-Dichloroethane | 236.8 | 226.1 | 95.5 |
| 2,2-Dichloropropane | 566.5 | 560.3 | 98.9 |
| Cis-1,2-Dichloroethene | 274.4 | 230.1 | 83.9 |
| Chloroform | 318.2 | 307.4 | 96.6 |
| Bromochloromethane | 329.0 | 362.4 | 110.2 |
| 1,1,1-Trichloroethane | 178.1 | 160.2 | 89.9 |
| 1,1-Dichloropropene | 227.0 | 191.8 | 84.5 |
| Carbon Tetrachloride | 163.5 | 156.2 | 95.5 |
| 1,2-Dichloroethane | 506.5 | 557.4 | 110.0 |
| Benzene | 358.9 | 333.7 | 93.0 |
| Trichloroethene | 245.9 | 252.4 | 102.6 |
| Bromodichloromethane | 435.6 | 395.2 | 90.7 |
| Dibromomethane | 430.3 | 497.1 | 115.5 |
| Cis-1,3-Dichloropropene | 550.9 | 576.4 | 104.6 |
| Toluene | 426.5 | 377.5 | 88.5 |
| 1,2-Dichloropropane | 481.0 | 424.1 | 88.2 |
| Trans-1,3-Dichloropropene | 601.6 | 721.6 | 107.2 |
| 1,1,2-Trichloroethane | 452.5 | 438.9 | 97.0 |
| 1,3-Dichloropropane | 466.9 | 597.8 | 105.0 |
| Tetrachloroethene | 258.7 | 255.6 | 98.8 |
| Chlorodibromomethane | 407.1 | 443.3 | 108.9 |
| 1,2-Dibromoethane (EDB) | 338.6 | 392.6 | 115.9 |
| Chlorobenzene | 418.8 | 394.2 | 94.1 |
| 1,1,1,2-Tetrachloroethane | 407.8 | 451.5 | 110.7 |
| Ethylbenzene | 325.6 | 392.3 | 108.6 |
| p-Xylene and m-Xylene | 876.1 | 905.9 | 103.4 |
| o-Xylene | 422.4 | 454.0 | 107.5 |
| Styrene | 408.0 | 430.0 | 105.4 |
| Bromoform | 501.3 | 523.0 | 104.3 |
| Isopropylbenzene | 388.2 | 303.5 | 78.2 |
| 1,1,2,2-Tetrachloroethane | 384.7 | 489.4 | 127.2 |
| 1,2,3-Trichloropropane | 307.7 | 222.2 | 72.2 |
| n-Propylbenzene | 502.7 | 440.9 | 87.7 |
| 2-Chlorotoluene & 4-Chlorotoluene | 1083.2 | 904.4 | 83.5 |
| Tert-Butylbenzene | 519.4 | 459.4 | 88.4 |
| 1,3,5-Trimethylbenzene | 477.7 | 479.7 | 100.4 |
| 1,2,4-Trimethylbenzene | 521.8 | 515.8 | 98.9 |
| Sec-Butylbenzene | 469.7 | 405.9 | 86.4 |
| p-Isopropyltoluene | 482.2 | 434.0 | 90.0 |
| 1,3-Dichlorobenzene | 539.4 | 482.8 | 89.5 |
| 1,4-Dichlorobenzene | 481.1 | 445.1 | 92.5 |
| n-Butylbenzene | 613.2 | 462.2 | 75.4 |
| 1,2-Dichlorobenzene | 538.3 | 456.3 | 84.8 |
| 1,2-Dibromo-3-Chloropropane | 705.1 | 651.5 | 92.4 |
| 1,2,4-Trichlorobenzene | 447.2 | 435.7 | 97.4 |
| Hexachlorobutadiene | 550.0 | 484.1 | 88.0 |
| Naphthalene | 522.6 | 569.9 | 109.1 |
| 1,2,3-Trichlorobenzene | 427.4 | 426.4 | 99.8 |
| 2-Methylnaphthalene | 485.4 | 424.0 | 87.4 |
| Bromobenzene | 555.1 | 455.4 | 82.0 |
| Analytical results provided to SoilCore, Inc. on December 10, 1998 |
REFERENCES CITED
Hanisch, R.C. and M.A. McDivitt, 1984. Protocols for Sampling and Analysis of Surface Impoundments and Land Treatment/Disposal Sites for VOCs. Technical Note EPA-EMB 68-02-3850, Work Assignment 11.
Barcelona, M.J. 1989. Overview of the Sampling Process. In: Keith, L.H. (Ed.), Principles of Environmental Sampling, American Chemical Society, Washington D.C., pp. 3-23.
U.S. EPA, 1991. Soil sampling and Analysis for Volatile Organic Compounds. Office of Research and Development, Office of Solid Waste and Emergency Response, Environmental Monitoring Systems Laboratory, Las Vegas, NV
Siegrist, R.L. and P.D. Jennsen, 1990. Evaluation of Sampling Method Effects on Volatile Organic Compound Measurements in Contaminated Soils. Env. Sci. Tech. 24:1387-1392.
Hewitt, A. D., "Studies with SoilCoreä Sampler and Sample Holders". US Army Corps of Engineers, Cold Regions Research and Engineering Laboratory, Hanover, NH 12/98.
Hewitt, A. D., "Storage and Preservation of Soil Samples for Volatile Organic Compound Analysis". Draft, US Army Corps of Engineers, Cold Regions Research and Engineering Laboratory, Hanover, NH 11/98.
U.S. EPA, Test Methods for Evaluating Solid Waste, SW846-Method 5035, Third Edition Update III, Office of Solid Waste and Emergency Response, Washington D.C.
U.S. EPA, Test Methods for Evaluating Solid Waste, SW846-Method 8260B, Third Edition Update III, Office of Solid Waste and Emergency Response, Washington D.C.
Wisconsin Department of Natural Resources, Validation Study for Sampling Devices Used to Collect Volatile Organic Compounds and Gasoline Range Organics in Soil. Bureau of Integrated Science Services, 1998.
Yazdani, S. 11/98. "Study to Document the SoilCoreä Sampler Discrete Sampling Device as an Equivalent Alternative to En Coreä in EPA SW846 Method 5035". SoilCore, Inc. Wyoming, MI 49548