Description:

XRF spectrometry uses primarily x-rays to irradiate a sample, which causes elements in the sample to emit secondary radiation of a characteristic wavelength. Two basic types of detectors are used to detect and analyze the secondary radiation:

1. Wavelength-dispersive XRF spectrometry uses a crystal to diffract the x-rays, as the ranges of angular positions are scanned using a proportional or scintillation detector (an extremely sensitive instrument that can be used to detect alpha, beta, gamma, and x-radiation).


2. Energy-dispersive XRF spectrometry uses a solid-state, Si(Li) detector from which peaks representing pulse-height distributions of the x-ray spectra can be analyzed.

The elements in the sample are identified by the wavelengths of the emitted x-rays while the concentrations of the elements are determined by the intensity of the x-rays. Sample preparation is minimal compared to conventional analytical techniques. XRF spectrometry allows for simultaneous determination of several elements. Portable energy-dispersive XRF instruments are now available, and the more accurate wave length XRF instruments can be used in mobile laboratories. The portable energy-dispersive XRF instruments can be used for scanning the ground surface to determine the prescence of metals without collecting a sample for analysis.

Analytes:

7. Metals

Media:

Limitations:

• Laboratory use with liquid samples requires preconcentration or precipitation, which is time consuming.
• Soil texture and moisture content may affect performance.
• Target analytes next to each other on the periodic table may not be separated adequately because they emit similar x-ray wavelengths.
• Unsuitable in ambient temperatures outside 30 to 100° F range.
• Surface scanning does not indicate whether metals of interest are present under the surface due to shallow depth of penetration. Collection and processing of soil by obtaining a vertical profile and grinding samples is generally required to obtain reproducible reading using a portable probe.
• Does not differentiate between different valences.

ASTM Standards/EPA Methods:

No applicable ASTM standards or EPA methods are cited for this technology.

Past use of Technique/Technology:

Data for these projects comes from the Field Analytical and Site Characterization Technologies Summary of Applications (EPA-542-R-97-011)

Number of sites: 39

• Bristol Sandblasting - RI
• Brockton Gas - MA
• Carroll Products - RI
• Cohen Property - MA
• Finberg Field - MA
• Goldfedders - CT
• Hatherway and Patterson - MA
• Kearsarge - NH
• Lake Success Business Park, Remington Arms - Bedford, MA
• New Hampshire Plating Co. - Merrimack, NH
• Precision Chrome Plating Corporation - RI
• RAE Battery - CT
• Raymark - CT
• Shapiro Site - MA
• Sparkling Fiber - NH
• Unidentified Site - CT
• Surrette Battery - NH
• West Street Property - MA
• Vega Baja Solid Waste Disposal Site - PR
• Hebelka
• Mid-Atlantic Wood Preserves - MD
• Palmerton Zinc - Palmerton, PA
• Unidentified Site; active manufacturing facility
• Lockheed Martin Advanced Recorders - Sarasota, FL
• Old Citgo Refinery - Bossier City, LA
• St. Charles Metal Finishing Company - St. Charles, MO
• Tri-State Jasper County, MO
• Unidentified Site; 15 abandoned or inactive smelter sites
• California Gulch Superfund Site - Leadville, CO
• China Lake NAWS - Ridgecrest, CA
• Concord Naval Weapons Station - Concord, CA
• Defense Distribution Region West, Sharpe Depot - Lanthrop, CA
• Defense Distribution Region West - location unknown
• Mare Island Naval Shipyard - Vallejo, CA
• Sacramento Army Depot - Sacramento, CA
• Verdese Carter Park - Oakland, CA
• McCarty's Pacific Hide and Fur - Pocatello, ID
• Umatilla Army Depot - Hermiston, OR
  

Documented Past Use

• The results of an immunoassay test kit for PCBs in soil were compared to confirmatory laboratory results using the Contract Laboratory Program (CLP) method for PCBs. The results of this demonstration have been published by EPA in an ITER, document number EPA/540/R-95/517.

• The immunoassay test kit was calibrated in the semiquantitative mode in two ways: using three Aroclor 1242 standards (5, 10, and 50 mg/kg) and using one Aroclor 1248 standard (10 mg/kg). When using the Aroclor 1242 calibration, the immunoassay test kit analyzed 52 samples. When compared to the confirmatory laboratory results, the test kit produced 28 correct results, 24 false positives, and no false negatives. When using the Aroclor 1248 calibration, the immunoassay test kit analyzed 94 samples. When compared to the confirmatory laboratory results, the test kit produced 75 correct results, 19 false positives, and no false negatives. Most semiquantitative test kits are designed to be conservative, and the results of this demonstration confirmed that fact.

• The immunoassay test kit was calibrated with three Aroclor 1242 standards (5, 10, and 50 mg/kg) in the quantitative mode. The immunoassay test kit and confirmatory laboratory produced detectable PCB results for 89 soil samples. A linear regression analysis was used to compare the two data sets. The regression line that was calculated had a r² of 0.87, a y-intercept of 17.8 mg/kg, and a slope of 0.76. The regression analysis indicated that there was a good correlation between the two data sets, but that 10 to 20 percent of soil samples from an investigation would have to be sent off site to correct the results for a better match to the confirmatory laboratory results. Also, when the quantitative immunoassay data was plotted against the confirmatory laboratory results using 10 mg/kg as a critical action level, no false negatives occurred.
  
  
• The results of three immunoassay test kits for PCP in soil and water were compared to confirmatory laboratory results using EPA SW-846 methods for PCP. Samples were collected from two sites to evaluate the effect of two different carrier solvents for PCP: diesel fuel and a butane-isopropyl ether mixture. The results of this demonstration have been published by EPA in an ITER, document number EPA/540/R-95/514.



• Penta RISc^® Test System-This was a semiquantitative test kit that produced results in the following concentration ranges: (1) below 0.5 mg/kg, (2) between 0.5 and 5 mg/kg, (3) between 5 and 50 mg/kg, and (4) greater than 50 mg/kg. The comparability study with the confirmatory laboratory consisted of 114 pairs of data for the soil samples. Overall, 83 of 114 times, or 73 percent of the time, the immunoassay test kit was correct. Of the other 31 times, the technology gave 21 false positive results (18 percent) and 10 false negative results (9 percent). All of the false negatives were produced from samples containing less than 10 mg/kg of PCP. The immunoassay test kit had a higher percentage of correct results from the soil samples that had diesel fuel as a carrier. In a comparison of 19 water sample results, the immunoassay test kit produced nine correct results, eight false positive results, and two false negative results.



• Penta RaPID Assay^®-This was a quantitative test kit that used a calibration curve generated using three standards for soil and water. The linear range for the water samples is 0.10 to 10.0 micrograms per liter (g/L) and 0.10 to 10.0 mg/kg for soil samples. A regression analysis was conducted on the immunoassay versus confirmatory laboratory data sets on 90 soil samples collected from the two sites. The regression of the entire data set produced a r² of 0.81, a y-intercept of 28 mg/kg and a slope of 0.43. This regression analysis indicated a good correlation between the two data sets, but the slope indicates that the immunoassay test kit was underestimating PCP concentrations, especially for samples that had PCP concentrations greater than 100 mg/kg. This immunoassay test kit had a higher r² (0.90) for soil samples with diesel fuel as the PCP carrier than for soil samples with the butane-isopropyl ether mixture as a carrier, which had a r² of 0.65. A regression analysis on 19 water samples produced a r² of 0.75 between the immunoassay test kit and the confirmatory laboratory results.



• EnviroGard Test Kit-During the demonstration, this test kit was found to produce a large number of false negative results and poor precision when high concentrations of PCP were found in a sample. As a result, the manufacturer changed part of the analytical protocol for the test kit. The revised test kit was never evaluated. This test kit is no longer commercially-available.
  
  
• A quantitative PAH test kit was used to analyze 35 soil samples and 10 water samples. It was speculated during the use of the immunoassay test kit that the high concentrations of the straight chain hydrocarbons were causing interferences for the samples. When the analytical results were received from the confirmatory laboratory, it was discovered that most of the immunoassay test results were an order of magnitude higher or more than the immunoassay results producing a high degree of false positives. This occurrence was most likely due to the interference from the other petroleum constituents.

• A semiquantitative TPH test kit was used to analyze groundwater samples suspected of containing petroleum contamination. Nearly all the immunoassay results showed no detectable petroleum contamination in the water samples. However, the confirmatory laboratory found detectable levels of heavier petroleum products (for example, motor oil) in most of the water samples. This occurrence produced a high false negative rate for the immunoassay test kit. The problem was that the TPH immunoassay test kit primarily responds to the lighter aromatic constituents in the petroleum products which were not present in the heavy petroleum products at this site. This was a good example of the importance of knowing the contamination prior to the use of an immunoassay test kit.
  
  
• At the end of fiscal year 1998, EPA anticipates having available for distribution the final report summarizing the results of a demonstration of Strategic Diagnostics, Inc.'s (SDI) Ohmicron RaPID assay system for PCB analysis. The demonstration, conducted under EPA's ETV Program, was designed to detect and measure PCBs in soil and solvent extracts. The results of the demonstration showed that the system for PCB analysis can provide useful, cost-effective data for environmental problem-solving and decision-making. As with any technology solution, the user ultimately must determine whether it is appropriate for the application and the project DQOs.

Verification/Evaluation Reports:

SITE (MMTP)
ETV
CSCT
Cal/Cert

Vendor/Instrument Information: