The purpose of this Advisory Opinion is to raise awareness of x-ray fluorescence (XRF) technology and its application in the Northeast.⁽¹⁾ This Advisory Opinion is intended to communicate Technology Review Committee (TRC) interest in the use of XRF technology to potential users of hazardous waste site characterization technology, such as consultants, as well as to project managers within the various state site cleanup programs. The Advisory Opinion is also intended to educate consultants and the state regulators who oversee projects about the factors that can affect the proper use of XRF technology.

All seven of the Northeast states participated in the development of this Advisory Opinion consensus statement. In addition, the technical information was reviewed by U.S. EPA Region I and a vendor of XRF technology. However, it should be noted that this Advisory Opinion is not intended to be an "approval" of this technology. The appropriateness of the use of XRF technology will need to be determined on a site-by-site basis. Potential users should contact officials in the state in which the project is located to determine if there are any state-specific requirements that could apply.

Project Background:

Regulatory and institutional barriers to the adoption of innovative hazardous waste site assessment technologies can result in increased expenditures to evaluate and remediate contaminated sites. Because innovative technologies have the potential to clean up and protect the environment and the public's health in a more cost-effective and efficient manner, finding ways to encourage their increased use is crucial.

Recognizing the need to overcome barriers to the acceptance of technology innovation, the six New England States, EPA Region I - New England, the Northeast Waste Management Officials' Association (NEWMOA) and the New England Governors' Conference signed a Memorandum of Agreement (MOA) in March 1998 to promote interstate regulatory cooperation for waste site assessment and cleanup technologies. Subsequently, NEWMOA has worked closely with EPA Region I and the Northeast Hazardous Substances Research Center (NHSRC) to increase the understanding of the factors that discourage the use of innovative technologies. NEWMOA held meetings and conference calls with NEWMOA's Waste Site Cleanup Workgroup and co-sponsored, with NHSRC, a Stakeholders Workshop held in May 1998 called "Increasing the Use of Innovative Technologies on Small Hazardous Waste and Petroleum Sites." The focus of this Workshop was on building consensus among the stakeholders regarding measures to reduce or eliminate obstacles to the use of innovative site assessment technologies.

At the May 1998 Stakeholder Workshop, participants identified the lack of an interstate forum in the Northeast to actively review technologies and communicate both public and private sector use of innovative technologies as a major impediment to the overall marketability of the newer field analytical, characterization and monitoring technologies. To address this need, NEWMOA has established the TRC, made up of one or more staff members from each of the Northeast states to coordinate state review, issue advisory opinions and disseminate information on the use of innovative technologies.

Overview of Technology:

XRF is a non-destructive analytical technique used to determine the metals composition of environmental samples, primarily soils. XRF can also be used to determine the metals concentration in water or air by filtering a known quantity and then using XRF to analyze the dried filter medium. The primary advantage of XRF analysis over laboratory analysis is that analytical results can be generated in real-time allowing decision-making in the field regarding the need for additional sampling or further remediation (provided that proper data validation procedures are followed). The time required for analysis depends upon the element(s) targeted; the concentration of a single element is typically determined in 30 seconds or less, and the analysis of several elements usually requires 2-5 minutes. Another advantage of XRF analysis over standard laboratory analysis is that the procedure does not involve altering the soil sample (other than mixing and possibly grinding) so no investigation derived wastes are generated as they are when extraction with solvent or acid is performed. Because the sample is not destroyed by the analysis, the same sample that was analyzed in the field can be sent to the laboratory for confirmatory analysis.

XRF analyzers emit X-rays that irradiate the sample and excite the electrons of the element(s) present. As these excited electrons return to their normal state they give off energy that is detected by the XRF equipment and the pattern is analyzed to determine the element. Generally, XRF technology can detect elements with an atomic number of 16 (sulfur) through 94 (plutonium). Different XRF analyzers use different elements to generate x-rays, with some using iron-55, cobalt-57, cadmium-109, or americium-241. The metals that can be detected by a particular XRF analyzer depend upon the element used to generate the x-rays:

• cobalt-57 is used to detect lead
• iron-55 is used to detect calcium, chromium, potassium, sulfur, titanium and vanadium
• cadmium-109 is used to detect arsenic, chromium, cobalt, copper, iron, lead, manganese, mercury, molybdenum, nickel, plutonium, rubidium, selenium, strontium, uranium, vanadium, zinc and zirconium
• americium-241 is used to detect antimony, barium, cadmium, silver and tin

Field XRF analyzers are generally relatively lightweight hand held instruments that gather data. After data collection, the analyzer is connected to a computer for data analysis and storage. XRF units utilize sophisticated software for internal calibration, to account for the natural decay of the radioactive x-ray source over time, and to analyze the data input from the probe. The software is all menu driven and can output the data in several formats so the user does not need to manipulate the data. Generally field XRF units utilize solid state detectors of lithium drifted silicon (Si(Li)), mercuric iodide or silicon pin-diode, with Si(Li) detectors the most precise. However, instruments that have a Si(Li) detector require liquid nitrogen to cool the detector, which can add time and expense to a project because of the need to: locate a source of the liquid nitrogen near the site; purchase the liquid nitrogen each day; and fill the internal dewar of the instrument and allowing the Si(Li) detector to cool down prior to analysis.

Some field XRF analyzers can be placed directly on the soil surface for in situ measurements. The analyzer measures the metal content of the sample over a surface area of approximately one square centimeter (1 cm²) area to a depth of approximately 2 millimeters (2 mm). Other field XRF analyzers require that soil samples are collected and placed in a sample cup that is then placed in a covered sample chamber for analysis. Most field XRFs can perform both in situ and ex situ analysis. In situ analysis provides qualitative data. Ex situ analysis can provide semi-quantitative or quantitative data depending upon the amount of soil preparation and the calibration standards used. Due to the inherent heterogeneity of soil, ex situ analysis is the preferred method because the soil can be homogenized to provide a sample that is more representative of the location from which it was collected. Average sample throughput for ex situ analysis generally varies from 50 to 100 samples per day, depending on the number of analytes, the particular analyzer used and the amount of soil preparation performed. In situ analysis allows a greater number of analyses at a given site because little or no sample preparation is performed.

Several different XRF analyzers have been evaluated in the EPA's Environmental Technology Verification (ETV) Program. The Technology Verification Statements for the evaluated XRF analyzers as well as more information about the ETV Program itself can be obtained at http://www.epa.gov/etv or by calling U.S. EPA Region I at (617) 575-CEIT. XRF technology has been used for site characterization or cleanup monitoring at over 35 Superfund sites, including several Department of Defense (DOD) and Department of Energy (DOE) facilities.⁽²⁾ U.S. EPA Region I has used XRF technology at several sites in New England and published Standard Operating Procedure for Elemental Analysis Using the X-MET 920 Field X-Ray Fluorescence Analyzer in October 1996. The guidelines can be obtained at http://www.epa.gov/ region01/measure/xray/xrayfluor.html or by calling (617) 575-CEIT. In addition, several Northeast states have successfully used XRF technology during site characterization and/or remediation, including Massachusetts, New Hampshire, New York and Vermont.

Recommendations:

The TRC has determined that, if used properly, XRF technology can provide useful data that should improve site characterization and/or cleanup verification. Potential users of XRF technology are strongly urged to consult U.S. EPA Region I's Standard Operating Procedure for Elemental Analysis Using the X-MET 920 Field X-Ray Fluorescence Analyzer (October 1996) and technology vendors prior to planning the field effort. Although the EPA Region I document was developed for the specific make of analyzer they owned at the time,⁽³⁾ much of the information in it is relevant to XRF in general. The project planning methodology outlined in Figure 1 of the document is attached to this Advisory Opinion. The intended use of the data and data quality objectives (DQOs) must be determined prior to the field event and agreed to by the appropriate regulatory authority. The TRC recommends the following items to improve or insure product performance; however, users should recognize that a particular XRF analyzer might have additional requirements:
 

1. Personnel who use XRF analyzers must be qualified and receive formal training including an explanation of XRF theory, how the specific analyzer works, quality assurance/quality control requirements and methods, and how to interpret the results. Most operators can be trained in one or two days. XRF vendors typically offer this type of training for a fee. Users should carefully follow the manufacturers instructions. In addition, because the analyzers use a form of radiation, a specific operator license is required by some states and includes a license fee. Typically, XRF vendors can guide the user through the required licencing procedure.
2. Soil samples must be collected and handled following standard procedures to promote consistency and comparability of results. In situ analysis requires that any nonrepresentative debris, such as rocks and leaves is removed and the soil surface is smooth so the probe window can make good contact with the soil surface. For semi-quantitative analysis every effort should be made to homogenize the soil. At a minimum, the moisture content of the soil must be low (see Item #5 below), large nonrepresentative debris removed, and the soil sample placed in a plastic bag for hand mixing prior to analysis. For quantitative data the sample must be oven dried, grinded with a mortar and pestle and passed through a 60-mesh sieve. Sections 11 and 12 of EPA Region I's Standard Operating Procedure for Elemental Analysis Using the X-MET 920 Field X-Ray Fluorescence Analyzer (October 1996) detail sample preparation requirements.
3. QA/QC requirements vary with different DQOs and regulatory authorities and should be agreed upon prior to the field event. Users are expected to record the results of the QA/QC sample analyses and evaluate them daily to ensure that the analyses and QA/QC checks meet the criteria established in the vendor literature and the project DQOs. In order to meet U.S. EPA QA/QC standards for semi-quantitative and quantitative data, the following QA/QC measures are usually required.⁽⁴⁾

• Energy calibration check should be performed at the beginning and end of each work day, and also after the batteries are changed or the instrument was shut off, or under any other circumstances recommended by the manufacturer. A pure element such as iron, lead or copper is analyzed to determine whether the characteristic x-ray lines are shifting.
• An instrument blank should be analyzed once a day or every 20 samples, whichever is more frequent. The instrument blank can be silicon dioxide, a Teflon block, a quartz block, "clean" silica sand or lithium carbonate.
• When sample preparation is performed using non-dedicated equipment (i.e. the grinding and sieving required for quantitative analysis), a method blank should be analyzed once a day or every 20 samples, whichever is more frequent. The method blank can be "clean" silica sand or lithium carbonate that undergoes the same sample preparation procedures as the environmental samples.
• A calibration verification check should be performed at the beginning and end of each day. The verification check standard(s) should contain all of the target elements at concentrations that are preferably near the action levels for the site. For semi-quantitative analysis, standard reference materials (SRM) obtained from the National Institute of Standards and Technology (NIST) are preferred. However, if the SRM does not contain a primary metal of concern, or if the concentration of the SRM is below the detection limit or not reasonably close to the concentration range of concern, then a commercially prepared standard containing the appropriate elements(s) at the appropriate concentration(s) should be used. The results of the calibration verification check should be within the limits specified for the SRM standard or ±30 percent of the value specified for the commercially prepared standard. For quantitative analysis, site specific calibration standards (SSCS) are required. Section 8.1 of EPA Region I's Standard Operating Procedure for Elemental Analysis Using the X-MET 920 Field X-Ray Fluorescence Analyzer (October 1996) details SSCS preparation requirements.
• A precision test should be performed at least once a day by calculating the relative standard deviation (RSD) between 7 to 10 replicate measurements of a standard. The standard analyzed for the precision test should contain the primary target element(s) for the site at a concentration near the action level. The RSD for the precision test should be less than 20 percent for each element, provided that the action level is suitably above the method detection limit.
• Duplicate samples should be analyzed (with the field XRF) at a minimum frequency of 1 per 20 or once per day, whichever is more frequent. Duplicates are two portions of the same sample that have been prepared and homogenized together, and then split and analyzed in the same manner.
• Field duplicate samples should be analyzed (with the field XRF) at a minimum frequency of 1 per 20 or once per day, whichever is more frequent. Field duplicates are two different samples taken from the same location that are prepared and homogenized separately, but using the same equipment and technique.
• A confirmatory sample should be collected and sent for laboratory analysis at a minimum frequency of 1 per 20 or once per day, whichever is more frequent. The confirmatory sample can be a split of the homogenized soil or the actual sample that underwent XRF analysis.

4. The half life of cadmium-109, iron-55 and cobalt-57 is relatively short (1.3, 2.7 and 0.75 years, respectively). Therefore, these source units must be replaced or reconditioned on a regular basis according to the manufacturers recommended schedule. The half-life of americium-241 is 458 years and therefore, replacement is not required.
5. A high moisture content can interfere with the analysis. In situ analysis cannot occur if there is ponded water. For ex situ semi-quantitative analysis, if the moisture content is greater than approximately 20 percent, then the sample should be air dried, or preferably dried in a convection or toaster oven. Oven drying is always part of the sample preparation protocol for quantitative analysis. Microwave drying should not be used because it can increase the variability of results and arcing can occur when metal fragments are present in the sample.
6. The presence of certain metals can interfere with the analysis of certain other metals. For example, iron tends to absorb copper x-rays, while chromium levels are enhanced in the presence of iron. The user should be aware of these types of matrix effects. Vendors can typically provide the necessary information during the planning stage to anticipate matrix effects. Moreover, the effects can be corrected mathematically through the software of the XRF instrument.

The NEWMOA Technology Review Committee has issued this Advisory Opinion on this 21^st day of September 1999.
 

For More Information Please Contact:
 

In Connecticut:  Christine Lacas  Department of Environmental Protection  Bureau of Water Management  79 Elm Street  Hartford, CT 06106  (860) 424-3766	In Maine:  Mark Hyland  Department of Environmental Protection  Bureau of Remediation and Waste Management  17 State House Station  Augusta, ME 04333  (207) 287-7673
In Massachusetts:  Dorothy Allen  Department of Environmental Protection  Bureau of Waste Site Cleanup  One Winter Street  Boston, MA 02108  (617) 292-5795	In New Hampshire:  Robert Minicucci  Department of Environmental Services  Waste Management Division  6 Hazen Drive  Concord, NH 03301  (603) 271-2941
In New York:  James Harringtion  Department of Environmental Conservation  Division of Environmental Remediation  50 Wolf Road  Albany, NY 12233  (518) 457-0337	In Rhode Island:  Laurie Grandchamp  Department of Environmental Management  Office of Waste Management  235 Promenade Street  Providence, RI 02908  (401) 222-2797
In Vermont:  Richard Spiese  Department of Environmental Conservation  Waste Management Division  103 South Main Street  Waterbury, VT 05671  (802) 241-3888	At NEWMOA:  William Cass  NEWMOA  129 Portland Street, 6th Floor  Boston, MA 02114  (617) 367-8558, ext. 301
At EPA Region I:  Carol Kilbride  U. S. EPA  Center for Environmental Industry and Technology  One Congress Street, Suite 1100  Boston, MA 02114  (617) 918-1831

 

Footnotes:

1. In this document, the Northeast states are: Connecticut, Maine, Massachusetts, New Hampshire, New York, Rhode Island and Vermont.

2. U.S. EPA, Field Analytical and Site Characterization Technologies, Summary of Applications, EPA-542-R-97-011, November 1997.

3. Note: U.S. EPA Region I has subsequently purchased additional XRF analyzers from other manufacturers.

4. QA/QC requirements for qualitative data are less extensive and should be agreed upon with regulatory authorities prior to the field event. In some instances, less rigorous QA/QC might be accepted by regulatory authorities for semi-quantitative data.