Final Report: Application of PGNAA Remote Sensing Methods to Real-Time, Non-Intrusive Determination of Contaminant Profiles in Soils

EPA Grant Number: R825549C054
Subproject: this is subproject number 054 , established and managed by the Center Director under grant R825549
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).

Center: HSRC (1989) - Great Plains/Rocky Mountain HSRC
Center Director: Erickson, Larry E.
Title: Application of PGNAA Remote Sensing Methods to Real-Time, Non-Intrusive Determination of Contaminant Profiles in Soils
Investigators: Faw, Richard E. , Shultis, J. K.
Institution: Kansas State University
EPA Project Officer: Hahn, Intaek
Project Period: May 18, 1995 through August 9, 1997
Project Amount: Refer to main center abstract for funding details.
RFA: Hazardous Substance Research Centers - HSRC (1989) RFA Text |  Recipients Lists
Research Category: Analysis/Treatment of Contaminated Soil , Land and Waste Management

Objective:

There are three principal tasks in the proposed effort. Goals of each are as follows:

  • Task 1 - Neutron transport in soil: Monte Carlo and/or discrete ordinates transport calculations will be performed to generate representative thermal and fast neutron fluence profiles in soil, resulting from isotropically incident 14-MeV neutrons within a fixed cone of collimation. Factors to be accounted for are:

      - soil composition
      - soil density
      - soil moisture
      - collimation angle

  • Task 2 - Capture gamma-ray transport in soil: The goal is to generate a library of gamma-ray fluences above the soil resulting from neutron capture in the soil. Neutron fluences and collimation angles determined in Task 1 will be used. A broad selection of concentration profiles in the soil will be made, including both continuous distributions and discrete layers. Factors to be specified in the library are:

      - soil composition
      - soil density and moisture content
      - spectrometer collimation angle
      - differentiation between capture gamma rays induced by thermal neutrons and fast neutrons

  • Task 3 - Application of modern inversion methods: In this phase of the project, modern inversion methods will be applied to the PGNAA soil contamination problem. In particular, the Bayes' statistical formulation based on the maximum entropy inversion approach will be applied to the present problem. During this phase, the maximum entropy method will be applied to the catalogue of spectra developed in Task 2 to assess the method's robustness for this particular application.
    Another key aspect of this phase of the project is the development of rigorous methods for estimating uncertainties in a predicted profile as a consequence of uncertainties in measured data, discretization errors in the physical model, and even from model assumptions. It is proposed to use the Latin Hypercube Sampling (LHS) method to assess the uncertainties in the estimated concentration profiles. This method can be used to propagate uncertainties from any number of input parameters, and thus is ideally suited for the PGNAA soil contamination problem. In this way, the sensitivity to uncertainties in the many data used to generate concentration profiles can be quantified and optimized measured strategies developed. The optimization of field measurements is a key aspect of using PGNAA for routine site analyses.
  • Summary/Accomplishments (Outputs/Outcomes):

  • PGNAA methodology is of immediate and direct interests to any organization concerned with remediation of hazardous wastes in soils. Specifically, the proposed research is directed toward meeting the technology needs for characterization, monitoring, and sensor technology, as prescribed by the U.S. Department of Energy Office of Waste Management.

    The PGNAA methodology is ideally suited to measurement of soil concentration profiles of the heavy-metal contaminants associated with mining and metallurgical enterprises. Specifically, the proposed research supports the Department of Energy efforts to develop methods for remote high precision characterization of buried waste.

    The proposed research is concerned with in situ hazardous-waste site characterization for and diagnosis of the need for decontamination.

    It deals with analysis of the factors affecting speed and accuracy of the PGNAA method and how both soil conditions and neutron-source/gamma-ray detector geometry need to be accounted for in the optimization of data collection and analysis.

    The proposed research will lead to improved precision in site characterization, thereby minimizing quantities of soil to be decontaminated.

    The work output of the proposed research is documentation of methodology. That documentation will take the form of reports and articles describing, in detail, methods of implementing optimization procedures.

    Investigators will make every effort to work directly with commercial and government enterprises in the implementation of the optimization methods developed for data collection and analysis.

    The proposed research was carried out over a two-year period. During the first year, a thorough literature review was be completed. Industry and national laboratory needs and interest were surveyed, leading to establishment of the scope of soil conditions and contaminant profiles to be addressed in the course of the research. Trial calculations were performed at Kansas State University, with sample spectra provided to personnel at Los Alamos to be used in the implementation of inversion methods to determine contamination profiles.

    During the second year, the full catalog of capture gamma-ray spectra was prepared and documented. The inversion methods were verified and the results documented.

    This project investigates the feasibility of rapid, in situ, determination of vertical contaminant profiles in soil using the method of prompt gamma ray neutron activation analysis (PGNAA). Fast neutrons are directed into the soil from a fixed source above the soil surface. A collimated radiation detector measures the intensities of the characteristic gamma rays released by elements in the soil that capture the neutrons. Analytical procedures are then used for "de-convolution" of measured gamma ray intensities to yield an estimated contaminant concentration profile in the soil. The effort is divided into three main tasks or investigations: (1) neutron transport in the soil, (2) capture gamma-ray transport in the soil, and (3) inversion or de-convolution of gamma-ray spectra. An important aspect of the research is a comprehensive uncertainty analysis, which applies to all tasks.

  • TASK 1 - NEUTRON TRANSPORT IN SOIL. The Monte Carol MCNP and discrete ordinates TORT and DANTSYS computer codes were obtained from the Radiation Shielding Information Center at Oak Ridge National Laboratory and were installed on UNIX work stations at Kansas State University. Installation and testing of the codes, and preliminary results are described in Reports HSRC-94-02-03, and HSRC-94-02-06 through HSRC-94-02-09.

    A thorough review was made of data on soil compositions, porosities, and water contents. It was possible to characterize soil in terms of key parameters, porosity and water content; and the decision was made to carry out calculations for five representative soils, one of nominal composition, density, and moisture content, and four others ranging from a dry porous soil to a wet dense soil. Composition data are contained in Report HSRC-94-02-06. Based on the examination of results of calculations of neutron intensities in soil and capture gamma-ray creation, it was found necessary to compute only the thermal-neutron flux density (one energy group) and to compute capture gamma-ray generation rates as the product of the thermal-neutron flux density and an effective capture cross section. Data on effective cross sections have been derived and are contained in Report HSRC-94-02-09. During the course of the project it was possible to improve the accuracy of both MCNP and discrete ordinated by installing the major neutron cross section data bases MCNPDAT6 and BUGLE-96.

    During the first year, based on the DLC-140 capture gamma-ray data base, software was developed to calculate the expected fluence of characteristic gamma rays at receptor locations above grade for a specified soil contaminant concentration profile and a specified soil type, normalized to a unit fluence of neutrons incident on the soil or a unit strength point source of neutrons above the soil.

    Completion of the discrete-ordinates neutron transport calculations yield the thermal neutron fluence in the soil for five soil classes and two types of sources. The neutron-fluence data were then incorporated into software used to compute capture gamma-ray source strengths in the soil. A two-dimensional interpolation method was developed for obtaining fluences as a function of depth and off-axis radial position for the point-isotropic source. Computational procedures for thermal-neutron fluence determination were then developed to couple the data to neutron-capture cross-section data and capture gamma-ray yield data. The combined package is to be used not only in software for calculation of contaminated profiles from measured gamma-ray energy spectra, but also in modeling the gamma-ray transport and detection to yield data for testing the main profile-inversion software.

    Rigorous mathematical models for the PGNAA analysis of the soil contamination profile have been developed for both point and parallel beam neutron sources. Codes, incorporating the neutron flux profiles determined in Task 1, have been written to invert these mathematical models and determine contaminant profiles. In this project, inversion codes were developed based on the (1) linear regularization (LR) technique, (2) LR with iterative positivity constraints (CLR), (3) the Backus-Gilbert (BG) method, and (4) the maximum entropy (ME) method. These codes were then used to analyze simulated PGNAA data for multiple test contaminant profiles.

    From simulation tests with idealized contaminant profiles, it was found the CLR and ME methods were the most robust at detecting complex shapes in the contaminant profile. However, the CLR method is exceedingly expensive computationally, and the more computationally efficient ME method is of more practical use for routine soil analyses. Under ideal conditions, i.e., data with negligible measured noise, the ME method was able to detect details in the contaminant profile at depths up to 60 cm.

    In this project, efforts were made on using simplified inversion methods that avoid the under-determination problem that forces the use of regularization procedures in the four inversion methods previously developed. These simplified procedures are based on describing a profile by a robust few-parameter distribution (e.g., lognormal, beta, etc.), and thereby avoid the need to estimate more parameters than there are measurements.

    Also, another iterative inversion procedure, the Algebraic Reconstruction Technique (ART), which has seen much use in tomographic image reconstruction, was also applied to the soil contaminant problem. However, the initial results were exceedingly disappointing, and further application of this method was not pursued.

    Finally, methods for simulating the stochastic nature of the measured capture-photon intensities were developed and incorporated into the inversion analysis codes. The Latin hypercube (LHC) method was used in to estimate the sensitivity of the estimated profiles to noise in the measured data. As expected, the more noise in the measured intensities the less detail in the contaminant profile that can be seen at greater depths.

    The more likely users of the research results are (1) government departments or agencies such as the Environmental Protection Agency, the Department of Defense, or the Department of Energy, (2) government-owned, contractor-operated facilities such as Oak Ridge National Laboratories, (3) industries such as Bechtel, Westinghouse, and Scientific Ecology with undertakings in waste management, and (4) instrumentation service industries such as EGG/ORTEC.

    The results have been communicated to users who have expressed interest in the project, and several presentations have been made at professional meetings and other groups.


  • Journal Articles on this Report : 2 Displayed | Download in RIS Format

    Other subproject views: All 22 publications 18 publications in selected types All 2 journal articles
    Other center views: All 904 publications 230 publications in selected types All 182 journal articles
    Type Citation Sub Project Document Sources
    Journal Article Shue SL, Faw RE, Shultis JK. Shultis, Thermal-neutron intensities in soils irradiated by fast neutrons from point sources. Chemical Geology 1998;144(1-2):47-61. R825549C054 (Final)
    not available
    Journal Article Shultis JK, Khan F, Letellier BC, Faw RE. Determining soil contamination profiles from intensities of capture-gamma rays using above-surface neutron sources. Applied Radiation and Isotopes 2001;54(3):563. R825549C054 (Final)
    not available

    Supplemental Keywords:

    prompt gamma ray neutron activation analysis, remote sensing, characterization, monitoring, sensor technology., RFA, Scientific Discipline, Waste, Geographic Area, Contaminated Sediments, Remediation, Environmental Chemistry, Geochemistry, Chemistry, Analytical Chemistry, Hazardous Waste, Ecology and Ecosystems, Hazardous, EPA Region, hazardous waste disposal, hazardous waste management, hazardous waste treatment, remote sensing, PGNAA, fate and transport , contaminated soil, Region 7, Region 8, hazardous organic compounds, hazardous waste identification

    Relevant Websites:

    http://www.engg.ksu.edu/HSRC Exit

    Progress and Final Reports:

    Original Abstract
  • 1995
  • 1996

  • Main Center Abstract and Reports:

    R825549    HSRC (1989) - Great Plains/Rocky Mountain HSRC

    Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
    R825549C006 Fate of Trichloroethylene (TCE) in Plant/Soil Systems
    R825549C007 Experimental Study of Stabilization/Solidification of Hazardous Wastes
    R825549C008 Modeling Dissolved Oxygen, Nitrate and Pesticide Contamination in the Subsurface Environment
    R825549C009 Vadose Zone Decontamination by Air Venting
    R825549C010 Thermochemical Treatment of Hazardous Wastes
    R825549C011 Development, Characterization and Evaluation of Adsorbent Regeneration Processes for Treament of Hazardous Waste
    R825549C012 Computer Method to Estimate Safe Level Water Quality Concentrations for Organic Chemicals
    R825549C013 Removal of Nitrogenous Pesticides from Rural Well-Water Supplies by Enzymatic Ozonation Process
    R825549C014 The Characterization and Treatment of Hazardous Materials from Metal/Mineral Processing Wastes
    R825549C015 Adsorption of Hazardous Substances onto Soil Constituents
    R825549C016 Reclamation of Metal and Mining Contaminated Superfund Sites using Sewage Sludge/Fly Ash Amendment
    R825549C017 Metal Recovery and Reuse Using an Integrated Vermiculite Ion Exchange - Acid Recovery System
    R825549C018 Removal of Heavy Metals from Hazardous Wastes by Protein Complexation for their Ultimate Recovery and Reuse
    R825549C019 Development of In-situ Biodegradation Technology
    R825549C020 Migration and Biodegradation of Pentachlorophenol in Soil Environment
    R825549C021 Deep-Rooted Poplar Trees as an Innovative Treatment Technology for Pesticide and Toxic Organics Removal from Soil and Groundwater
    R825549C022 In-situ Soil and Aquifer Decontaminaiton using Hydrogen Peroxide and Fenton's Reagent
    R825549C023 Simulation of Three-Dimensional Transport of Hazardous Chemicals in Heterogeneous Soil Cores Using X-ray Computed Tomography
    R825549C024 The Response of Natural Groundwater Bacteria to Groundwater Contamination by Gasoline in a Karst Region
    R825549C025 An Electrochemical Method for Acid Mine Drainage Remediation and Metals Recovery
    R825549C026 Sulfide Size and Morphology Identificaiton for Remediation of Acid Producing Mine Wastes
    R825549C027 Heavy Metals Removal from Dilute Aqueous Solutions using Biopolymers
    R825549C028 Neutron Activation Analysis for Heavy Metal Contaminants in the Environment
    R825549C029 Reducing Heavy Metal Availability to Perennial Grasses and Row-Crops Grown on Contaminated Soils and Mine Spoils
    R825549C030 Alachlor and Atrazine Losses from Runoff and Erosion in the Blue River Basin
    R825549C031 Biodetoxification of Mixed Solid and Hazardous Wastes by Staged Anaerobic Fermentation Conducted at Separate Redox and pH Environments
    R825549C032 Time Dependent Movement of Dioxin and Related Compounds in Soil
    R825549C033 Impact of Soil Microflora on Revegetation Efforts in Southeast Kansas
    R825549C034 Modeling the use of Plants in Remediation of Soil and Groundwater Contaminated by Hazardous Organic Substances
    R825549C035 Development of Electrochemical Processes for Improved Treatment of Lead Wastes
    R825549C036 Innovative Treatment and Bank Stabilization of Metals-Contaminated Soils and Tailings along Whitewood Creek, South Dakota
    R825549C037 Formation and Transformation of Pesticide Degradation Products Under Various Electron Acceptor Conditions
    R825549C038 The Effect of Redox Conditions on Transformations of Carbon Tetrachloride
    R825549C039 Remediation of Soil Contaminated with an Organic Phase
    R825549C040 Intelligent Process Design and Control for the Minimization of Waste Production and Treatment of Hazardous Waste
    R825549C041 Heavy Metals Removal from Contaminated Water Solutions
    R825549C042 Metals Soil Pollution and Vegetative Remediation
    R825549C043 Fate and Transport of Munitions Residues in Contaminated Soil
    R825549C044 The Role of Metallic Iron in the Biotransformation of Chlorinated Xenobiotics
    R825549C045 Use of Vegetation to Enhance Bioremediation of Surface Soils Contaminated with Pesticide Wastes
    R825549C046 Fate and Transport of Heavy Metals and Radionuclides in Soil: The Impacts of Vegetation
    R825549C047 Vegetative Interceptor Zones for Containment of Heavy Metal Pollutants
    R825549C048 Acid-Producing Metalliferous Waste Reclamation by Material Reprocessing and Vegetative Stabilization
    R825549C049 Laboratory and Field Evaluation of Upward Mobilization and Photodegradation of Polychlorinated Dibenzo-P-Dioxins and Furans in Soil
    R825549C050 Evaluation of Biosparging Performance and Process Fundamentals for Site Remediation
    R825549C051 Field Scale Bioremediation: Relationship of Parent Compound Disappearance to Humification, Mineralization, Leaching, Volatilization of Transformaiton Intermediates
    R825549C052 Chelating Extraction of Heavy Metals from Contaminated Soils
    R825549C053 Application of Anaerobic and Multiple-Electron-Acceptor Bioremediation to Chlorinated Aliphatic Subsurface Contamination
    R825549C054 Application of PGNAA Remote Sensing Methods to Real-Time, Non-Intrusive Determination of Contaminant Profiles in Soils
    R825549C055 Design and Development of an Innovative Industrial Scale Process to Economically Treat Waste Zinc Residues
    R825549C056 Remediation of Soils Contaminated with Wood-Treatment Chemicals (PCP and Creosote)
    R825549C057 Effects of Surfactants on the Bioavailability and Biodegradation of Contaminants in Soils
    R825549C058 Contaminant Binding to the Humin Fraction of Soil Organic Matter
    R825549C059 Identifying Ground-Water Threats from Improperly Abandoned Boreholes
    R825549C060 Uptake of BTEX Compounds by Hybrid Poplar Trees in Hazardous Waste Remediation
    R825549C061 Biofilm Barriers for Waste Containment
    R825549C062 Plant Assisted Remediation of Soil and Groundwater Contaminated by Hazardous Organic Substances: Experimental and Modeling Studies
    R825549C063 Extension of Laboratory Validated Treatment and Remediation Technologies to Field Problems in Aquifer Soil and Water Contamination by Organic Waste Chemicals