a multi-layer phoswich radioxenon · pdf filedavid m. hamby, oregon state university 1 volume...

David M. Hamby, Oregon State University

1

VOLUME 1OFFER DOCUMENTS

Research Proposalfor

Financial Assistance

submitted in response to solicitation number

DE-SC52-05NA26703

TOPIC NO. 7

A MULTI-LAYER PHOSWICH RADIOXENON DETECTION SYSTEM

Oregon State UniversityCorvallis, Oregon

Principal Investigator (and proposal preparer):David M. Hamby, PhD

Professor, Dept of Nuclear Engineering and Radiation Health PhysicsE120 Radiation Center

Oregon State UniversityCorvallis, OR 97331-5902

Tel: 541-737-8682Fax: 541-737-0480

[email protected]

Contract Administrator:Clem LaCava

Assistant Contract Officer, Contract AdministrationOregon State University

B306 Kerr Administration BuildingCorvallis, OR 97331-2147

Tel: 541-737-2373Fax: 541-737-2069

[email protected]

Total First Year’s Costs: $426,333 First Year’s Equipment Costs: $0Total Proposed Costs: $1,294,478 Total Equipment Costs: $9,500

Date Submitted: July 10, 2005Period of Performance: Jan. 1, 2006 – Dec. 31, 2008

Offer Acceptance Period: July 27, 2005 – July 1, 2006

the use of IRIS PASSCAL seismic equipment is NOT proposed


1

VOLUME 2TECHNICAL PROPOSAL

Solicitation No.: DE-SC52-05NA26703

TOPIC NO. 7

A MULTI-LAYER PHOSWICH RADIOXENON DETECTION SYSTEM

Oregon State UniversityCorvallis, Oregon

Principal Investigator:

David M. Hamby, PhDProfessor, Dept of Nuclear Engineering and Radiation Health Physics

E120 Radiation CenterOregon State University

Corvallis, OR 97331-5902Tel: 541-737-8682Fax: 541-737-0480

[email protected]

the use of IRIS PASSCAL seismic equipment is NOT proposed

Equipment (detector prototype) to be purchased: The researchteam will design, and have constructed, two phoswich detectors aspart of this work, one of which is expected to require funds inexcess of the value defining “equipment”. Therefore, in the thirdproject year, a saddle-design phoswich will be modeled andconstructed that is consistent with the current ARSA configuration(cost estimated at $9,500).


2

1. SUMMARY

Our instrumentation research group at Oregon State is in the latter stages of developing astate-of-the-art radiation detection system for real-time identification and dosimetry of beta-emitting radionuclides in mixed radiation fields. The group has been working on this problemfor the past seven years, with twenty-six related publications and three doctoral students havingcompleted their degrees (the work proposed herein will provide an additional three doctoralstudents with a real-world problem on which to base their doctoral dissertation). Threeinnovative prototypic multi-layer phoswich radiation detectors, developed by the PI's researchgroup, have been studied with an ultimate goal of sophisticated beta spectroscopy and dosimetry.The current proposal is focused on the development of a low-cost, optimized multi-layerphoswich scintillating detector in the saddle geometry and a high-speed digital processing systemthat will be retrofitted to the ARSA structure. Our radioxenon-optimized phoswich(XEPHWICH) and its supporting signal-analysis firmware will reduce the cost of the currentARSA detector by more than 70% and will significantly increase reliability of nuclear-detonationradioxenon detection, even in the presence of high radon concentrations. The digital signalprocessing technique will be capable of spectral discrimination, detailed pulse componentanalysis, neural network nuclide identification, memory-based quantification of radioxenons, andcomputationally enhanced beta spectroscopy. Multi-modal counting schemes (combinations ofelectronic and light coincidence/ anti-coincidence methods) will be employed to digitally analyzescintillator light output, and its multi-component signal shape, from simultaneous electron andphoton interactions in the layered scintillation material. Our ultimate objective for the proposedresearch is to enhance the existing ARSA radioxenon detection technology to be more compact,reliable, and inexpensive, requiring less power and maintenance, and using state-of-the-art digitalpulse processing and phoswich technology.

2. NARRATIVE

PROBLEM STATEMENTThe Comprehensive Test Ban Treaty (CTBT) mandates that the International Monitoring

System (IMS) establish a worldwide network of radiation detection systems for nuclearexplosion monitoring. These detection systems must be capable of detecting fission-productradioxenons with a minimum detectable concentration of 1 mBq/m3 for 135Xe. A vast amount ofresearch has been carried out by various groups to develop the Automated RadioxenonSampler/Analyzer (ARSA), as well as three other similar technologies (SAUNA, ARIX, andSPALAX) around the world (Reeder and Bowyer 1998; Bowyer et al. 1999; Hayes et al. 1999;Heimbigner et al. 2002; Penn et al. 2002; Penn 2003; Ringbom et al. 2003; Ely et al. 2003;McIntyre et al. 2004; Rynes et al. 2004; Reeder et al. 2004). Additionally, a manufacturingprototype of the ARSA system was installed in China, in late 2002 (Penn et al. 2002; Penn 2003;Rynes et al. 2004).

Because of the need for a large array of monitoring systems in the IMS network, furtherresearch is needed to greatly reduce power requirements, size, production costs, and complexityof individual systems and components. Radionuclide detection techniques that are definitive inthe identification and quantification of fission-product radioxenons in the presence of highbackground radiations (including radon) are of great value. The current ARSA detector,however, is large, requires power and electronics for 12 photomultiplier tubes, and is incapableof definitive radioxenon detection and measurement (Reeder and Bowyer 1998; Penn et al. 2002;


3

Penn 2003; Rynes et al. 2004). The research proposed herein will take large strides towardincreasing detection sensitivity, speed, and accuracy by modifying the ARSA design, andthrough the use of digital signal processing (DSP), field programmable gate arrays (FPGA), andneural network techniques.

Approximately seven years ago, our research group began development of a triplephosphor sandwich (“TPS”) detector that we have shown provides information for identifyingand quantifying beta-emitting radionuclides (Bush-Goddard 2000). Based on advancementsmade to the Bush-Goddard design, the PI and his students developed a 2nd prototype scintillator(Kriss and Hamby 2004a) to investigate beta dosimetry in thin detector layers (Kriss and Hamby2003; 2004b; 2004c). Expanding on what we’ve learned (Tavakoli-Farsoni and Hamby 2004a;2004b) and from current work on a 3rd generation design for beta dosimetry (DE-FG07-05ID14704), we are proposing an extension of our research that involves full development andcharacterization of a prototype radioxenon multi-layer phoswich detector (which we will callXEPHWICH) and automated high-speed digital signal analysis for real-time measurement offission-product radioxenons in the presence of radon and other background sources.

We will employ digital pulse-shape discrimination with the XEPHWICH to differentiatebetween the contributions to light output from different radiation types in a mixed beta/gammafield, thus allowing the generation of 3-dimensional histograms of coincident energy deposition,as well as the simultaneous collection of individual energy spectra from gamma and betaradiations. Ultimately, our software-based spectral manipulation routines will be coded ashardware using FPGA digital filtering and digital signal processing to enhance the usefulness ofsignal pulses and to drastically increase spectral processing speeds.

BACKGROUNDAn ideal scintillator is transparent to its own light, possesses fast light-emission times,

has a high degree of scintillation efficiency, has light emission properties that are linear withdeposited energy, and has good optical qualities (Knoll 2000). The electron response functionfor scintillators is dependent on the material selected, its thickness, and the angular incidence ofelectrons. Additionally, backscatter and bremsstrahlung are important elements in anyinstrument used for electron detection, with their significance increasing as the atomic number ofthe scintillator increases. Scintillators with low Z, therefore, are desired for electronspectroscopy, and thin scintillators will have minimal gamma contributions because of lowinteraction probabilities. These contributions are a function of incident photon energy and canbe characterized by experimentation or modeling of photon interactions.

There are several types of scintillators in various physical and chemical states. Theseinclude organics as polymers, crystals, thin films, loaded scintillants, and liquids; inorganics ascrystals (e.g., NaI, CsI, LiI, ZnS, CaF2, CsF); silicate glasses containing lithium or cerium; andnoble gases. Fluorescence in the organic scintillators originates from energy-level transitionswithin the structure of a single molecule, whereas fluorescence in inorganic crystals takes placein the lattice structure. This being the case, the organic scintillators can be in a variety ofdifferent physical forms (e.g., solid, liquid), but the inorganics require the crystalline structure toproduce the characteristic scintillations.

The excitation and de-excitation processes in organic scintillators can be describedadequately by simple exponential rise and decay times. Rise times are typically very short (onthe order of hundreds of picoseconds) and decay times are typically on the order of 1-4nanoseconds, dependent on the molecular energy-state structure of the organic.

The light emission processes of inorganic crystals are based on excitation and de-excitation of energy states in the lattice structure. Excited states essentially are all formed at


4

once and then decay exponentially. Although other processes, like afterglow and quenching dooccur, the light emission timing characteristics are adequately described by a single exponential.The inorganics tend to have greater light yield and scintillation linearity with energy deposition,but have longer decay times relative to the organics, on the order of hundreds to thousands ofnanoseconds. Even though their light output is not as great, organic scintillators, because of theirlower atomic numbers, are most commonly used for electron or beta spectroscopy. Inorganicsgenerally have a higher Z than the organics and, therefore, will be more susceptible to low-energy gamma interference and more likely to cause backscatter of incident betas. Many of theinorganic scintillators are specifically used for gamma spectroscopy because of their higherdensity. Most inorganic scintillators are hygroscopic and, therefore, require encasement toprevent breakdown from atmospheric moisture. Calcium fluoride (CaF2) is an inert, non-hygroscopic scintillating crystal that’s use is ideal where harsh environments may beencountered or where the application requires the scintillator to be in direct contact with themedium being analyzed (e.g., tritium quantification in groundwater).

Even though its light output is relatively low, glass is sometimes used for beta or gammadetection when harsh environmental conditions (e.g., corrosive chemicals or high temps) preventthe use of other forms of scintillator. Decay times, at about 50-70 nanoseconds, are faster thanthe inorganics, but slower than the organics. Glass is not the ideal scintillator for low levelcounting because of the likelihood of it containing naturally radioactive potassium or thorium.

Individual molecules in the noble gases can be excited to higher energy states by thepassage of photons or charged particles. These elevated states then return to the ground state bythe emission of UV photons. Although in the ultraviolet region, wave-shifting molecules can beadded to the gas or specialized photomultipliers can be used for signal collection. Thescintillation efficiency is quite low and de-excitation occurs in a few nanoseconds or less,making gases among the fastest of all radiation detectors. Because of the low stopping power,however, gases are usually used as scintillation detectors only for alpha or heavy chargedparticles (Knoll 2000).

Phoswich Detectors. A phoswich detector is generally thought of as one in which twodifferent scintillators have been optically coupled to each other and to a single photo-collectiondevice. The two scintillators are referred to as a phosphor sandwich, hence the name phoswich.Scintillators in the sandwich are chosen specifically so that their light emission decay times aredifferent. Thus, it can be determined in which phosphor a given interaction takes place. Thisallows rudimentary discrimination by particle type using pulse-shape analysis techniques(Wissink et al. 1997; Lautridou et al. 1996; Frontera et al. 1993). Sodium iodide (NaI:Tl) andcesium iodide (CsI:Na) are often chosen as the two sandwich materials because their decay timesare quite different, and pulses arising from only one scintillation are easily distinguished fromthose with both components, using the pulse shape discrimination method.

Our first triple phoswich (TPS) detector (Fig. 1), designed in the late 1990’s, follows oneapproach to analyzing beta energy spectra (Bush-Goddard 2000). The concept for that detectorinvolves layering three distinctly different scintillators on top of a photomultiplier tube. Low-energy betas (< 100 keV) will stop in the first layer, intermediate betas (100 keV – 1 MeV) in thesecond layer, and high-energy betas (> 1 MeV) in the third layer. Each scintillator has a uniquescintillation decay time. By analyzing the decay time of the photomultiplier’s output,conclusions can be reached as to which layers produced the signal. This provides a quantitativemeasure of the energy range of the betas incident on the detector.

Phoswich detectors have been used for a number of particle discrimination applications(Langenbrunner et al. 1992; Usuda 1992; Wang et al. 1994; Usuda et al. 1994a; Usuda and Abe1994; Nagornaya et al. 1996; Usuda et al. 1994b; Ely et al 2003) as well as gamma telescopes in


5

astronomical investigations (Schindler et al., 1997; Qi et al. 1997a; Lum et al. 1997), mixed beta-gamma dosimeters (Vasil'ev and Volodin 1996), PET and SPECT components (Dahlbom et al.1997), and heavy ion detectors (Fox et al. 1996; Qi et al. 1997b). A few researchers haveexamined unique variations of the phoswich detector, including a gas proportional phoswich(Benchekroun et al. 1993), detectors that utilize both scintillator and solid-state designs (Strausset al. 1990), and well-counter phoswich configurations (Kamae et al. 1993). The use of threescintillators layered together is uncommon, but has been investigated primarily by Usuda et al.(1994a; 1997) at the Japan Atomic Energy Research Institute. These studies and our work haveshown that excellent discrimination between radiation types can be obtained, given the use of theappropriate scintillators and timing electronics.

Fig. 1. First generation triple-layer phoswich (TPS) detector (Bush-Goddard 2000).

Figure 2 shows four plots of captured waveforms from the TPS developed by Bush-Goddard (2000) and further characterized by Tavakoli-Farsoni (2004a). In each plot the uppercurve shows preamplifier output and the lower curve is the clipped-amplifier output. The falltime of the clipped-amp waveform corresponds to the rise time of the preamplifier output. Fastand slow components can be extracted from these digital waveforms, but timing componentdetail is lost if using analog processes.

Beta spectroscopy with plastic scintillators. Beta spectroscopy is performed generallyfor the purpose of nuclide identification or beta dosimetry. Scintillation dosimeters may becategorized by function: those that are used in laboratory or work settings to measureoccupational dose; those that are used in medical settings for measuring patient dose; and thosedeveloped for special research purposes. In the lab or workplace, a common technique used tomeasure beta dose is to first measure the beta spectrum with a scintillator, and then calculate adose from that information. Martz et al. (1986) used a plastic scintillator 2.5 cm diameter by 0.9cm deep to measure beta spectra and convert those spectra to dose. They used a beta energydeposition function, derived from calibrated sources, to convert the measured spectra to dose at adepth of 7 mg/cm2. Thus, calculation of dose relied not only on direct extrapolation ofscintillator light output to dose, but on previously derived calibration curves, in order to isolatethe dose to a thin layer at a specific depth. Gammas were excluded by measuring spectra withand without a beta shield. Shen et al. (1987) used plastic scintillators to measure spectra, fromwhich they subsequently calculated doses using electron transport theory as applied to TLDs.Swinth et al. (1989) constructed a combination proportional counter-plastic scintillation counterfor measuring beta spectra and dose. They used coincidence gating to exclude gamma events.


6

Dose was calculated from spectral information and compared to extrapolation chamber data forcalibration. Horowitz et al. (1993) developed a two-detector telescope device consisting of athin, front silicon detector and a thick, back plastic scintillator. Again, gamma rejection wasaccomplished by electronic coincidence analysis. Dose was calculated by comparison to MonteCarlo depth distributions for the spectra measured. Vapirev et al. (1996) employed a plasticscintillator to measure beta spectra after passage of the betas through absorbers of variousthicknesses. Dose was calculated via specific energy losses, dE/dx, and the collected energyspectra. Results were compared to the calculations of Cross and Marr (1960).

Fig. 2. Digital waveforms captured from the TPS detector with a high-speed oscilloscope.

In the medical setting, much of the effort has gone into dose measurements of high-energy photon beams (Beddar et al. 1992a, 1992b; de Boer et al. 1993; Mainardi et al. 1997;Clift et al. 2000). Though not measuring beta energy, the materials are the same, namely plasticscintillators coupled to a light detector and associated electronics. The complications are alsosimilar, for instance, the need to account for Cerenkov radiation. Not all efforts have beendirected towards photon radiation therapy: Bambynek et al. (2000) developed a dosimetrysystem for cardiovascular brachytherapy beta sources using a plastic scintillator; several authors(Williamson et al. 1999; Kirov et al. 1999; Fluhs et al. 1996) worked on plastic scintillatorresponse to low-energy photons from brachytherapy sources; and de Sousa et al. (2000) studied adosimeter for patients undergoing diagnostic radiology procedures. The primary advantages ofplastic scintillator material in all of these cases are its near-water equivalence, a property usefulwhen dose to tissue is desired, and small backscatter factors.

Several authors have studied thin plastic scintillators for beta dosimetry. Bingo et al.(1980) developed a beta dose survey meter using a 2 mm thick scintillator. The premise was thatthere existed a certain thickness of scintillator that would satisfy a directly proportionalrelationship between count rate and dose rate, for all beta energies, i.e. independent of betaenergy. Two millimeters happened to be the experimentally determined optimum thickness.Johnson et al. (1983) deliberately chose to use a very thin plastic scintillator, backed by a 1 cm


7

thick Lucite light pipe, to measure dose to skin directly. Kriss and Hamby (2003; 2004b) usedthin sections of scintillator to estimate beta dose as a function of tissue depth (Fig. 3).

Fig. 3. Our 2nd generation scintillation detection system, using a large area photodiode.

The spectra of Figs. 4 and 5 were collected using the 2nd generation scintillator and thelarge-area avalanche photodiode developed and characterized by Kriss (2004d). Figure 4 showsthe results of raw measurements and Kriss’ spectral shaping method (Kriss and Hamby 2004c)compared to theoretical emission spectra. Figure 5 shows an example of dosimetric calculationsfrom our measured spectra as compared to Monte Carlo (MCNP) estimates.

Fig. 4. 14C and 36Cl measured (Kriss detector) and theoretical emission spectra, respectively.

Fig. 5. 90Sr/Y shallow dose and 210Bi surface dose spectrum (source distance = 20 mm).

Finally, Watt and Alkharam (1995) proposed using extremely thin (20 µm) plasticscintillators to directly simulate DNA damage, in the sense that the fluor spacing in thescintillator is analogous to the DNA diameter of around 2 nm. So, two scintillation emissionswithin 2 nm can be considered a double strand break, and thus an indication of biological effect.

Radioxenon Detection. A number of researchers have investigated various techniques toidentify radioactive isotopes of xenon (Reeder and Bowyer 1998; Bowyer et al. 1999; Arthur et


8

al. 2001; Heimbigner et al. 2002; Ely et al. 2003; Ringbom et al. 2003; McIntyre et al. 2004;Reeder et al. 2004), primarily for nuclear explosion monitoring. Reeder, Bowyer and others(Reeder and Bowyer 1998; Bowyer et al. 1999; Arthur et al. 2001; Heimbigner et al. 2002)developed the ARSA prototype to collect beta and gamma spectra on a single sample and, withthe use of analog electronic coincidence methods, developed two-dimensional (electron vsphoton energy) pulse-height spectra. Ely et al. (2003) suggest the use of a two-layer phoswichdesign for beta-gamma coincidence. They use rise-time analysis to determine the layer in whichenergy deposition took place, however, their method is unable to identify beta-gamma coincidentpulses and also unable to produce pure beta or pure gamma spectra. Triple coincidence methodshave been investigated by McIntyre et al. (2004). These methods appear to improve on theARSA design, however, power requirements are still excessive and detector complexity remainshigh.

Pulse-Shape Analysis. Some scintillators respond to different types of radiation (i.e.,different rates of energy transfer) by emitting light with different timing characteristics.Likewise, different scintillators respond to the same radiation by emitting light differently.Because of these differences, a pulse-shape analysis (PSA) technique, or pulse-shapediscrimination (PSD), can be performed to identify and selectively analyze the signal from eithera particular radiation type (in two scintillators) or from a particular scintillator (with tworadiation types). The majority of PSA/PSD systems operate on analog signal pulses; currenttechnology, however, allows us to do more sophisticated analyses with the use of digital signalprocessing (DSP).

Traditionally, one of two approaches is used to perform the pulse shape discriminationfor phoswich detectors. On the one hand, the rise time technique is based upon the integration ofthe light pulse (e.g. or the anode pulse of the phototube), followed by the determination of thetime at which this integral reaches a certain fraction of its maximum. On the other hand, thecharge integration method requires the comparison of the charge collected at the anode signalover two different time intervals, one normally encompassing the entire duration of the pulse,and the other limited only to a certain portion. Both methods are achieved using analogelectronic systems.

Recently, with the development of fast ADCs, digital signal processing (DSP) methodshave gained popularity for analyzing signals from radiation detectors. The use of digital systemsoffers several advantages over conventional analog units, including digital pulse chargeintegration, reduced dead time, elimination of distorted pulses, noise analysis and minimization,pulse shape discrimination capabilities, and pulse-shape component analysis.

In our lab, we have developed a digital pulse shape discrimination algorithm specificallyfor the triple-layer phoswich detector. To illustrate this approach, Figs. 6 and 7 show originaland processed digitized waveforms, respectively, that we’ve measured. The abscissa representstime (ns) and the ordinate represents amplitude (volts). The radiation pulses were captured usinga 1 GHz-digital oscilloscope (Tektronix TDS1002).

The first two waveforms across the top in Fig. 6 were constructed using mathematicallymodeled pulses to evaluate the process; the remaining waveforms were collected from our first-generation TPS detector on exposure to 36Cl, a pure beta emitter. The TPS detector has two frontscintillation layers similar to our 3rd generation design. The original waveforms were digitallyprocessed using the following functions:

1. noise minimization using digital filters;2. zero-point determination (the point where the waveform begins);3. baseline determination (by averaging over a finite duration of time prior to the zero

point);


9

4. baseline subtraction from the whole waveform; and5. determination of amplitude for both fast and slow components.

Fig. 6. Original waveforms (36Cl detected by the TPS).


10

Fig. 7. Processed waveforms (36Cl measured by the TPS, digitally processed and shaped).

The last step is carried out by fitting the reference pulses to the captured waveforms; thecurve fitting is only possible by using the digital processing approach; analog systems areincapable of component analysis. The calculated fast, FA, and slow, SA, components ofamplitude (volts) for each waveform are presented in Fig. 7.

Corrections to the waveforms of Fig. 6 are necessary due to the light yield differencesbetween scintillation layers, but eventually, these fast and slow component values will be used togenerate the digitally-processed energy spectra. Waveforms A and B (top of Fig. 7) weregenerated by combining modeled/normalized fast and slow components with differentamplitudes to simulate beta absorption in two successive scintillation layers (with differentenergy deposition in each layer). They have been made with different fast and slowcombinations [FA1 = 0.3 V; SA1 = 0.7 V; FA2 = 0.7 V; SA2 = 0.3 V], each having the same totalamplitude of 1 Volt. In this simple example, the processed results (Fig. 7) show good agreementwith the original amplitudes. We currently have this algorithm under development to fit to our3rd generation design for a simultaneous gamma/beta spectrometer and dosimeter. Modificationsto this algorithm will allow waveform manipulation for the proposed XEPHWICH.

In the XEPHWICH detector, the pulse-component analysis will result in estimates ofthree amplitude determinations relating to the amount of energy deposited in each of the threelayers. This analysis can be carried out using a least-squares minimization technique with,

()20,[][]nNMSnFn==−∑, where

[][][][]abcFnAfnBfnCfn=⋅+⋅+⋅

and where fx[n] represents the normalized response of scintillating layer x, and the coefficients(A, B, and C) represent the contributions to signal amplitude of the three signal components.The method can be hardware-implemented to provide rapid, real-time amplitude estimates (Fig.8). The hardware includes three multipliers, accumulators, and adders and nine scalars, allconfigured in FPGA by VHDL (Bolic and Drndarevic 2002). The recursive analysis results inestimates of the coefficients until the residual error is minimized to an acceptable level. Ourpreliminary studies show that these estimates can be obtained within about 2.5 microseconds(Fig. 9).

Sa Acc.

Sc Acc.

Sb Acc.

S[n]

fa[n]

fb[n]

fc[n]

A[n]

B[n]

C[n]]

X11

X12

X13

X23

X21

X22

X31

X32

X33

Sa[n]

Sb[n]

Sc[n]


11

Fig. 8. Recursive DSP realization for resolving light-pulsecomponents using least-squares fitting.

Fig. 9. Least squares residual error for resolving light-pulse components.

Digital signal processing. Applications of digital techniques for radiation pulse processinghave increased steadily in recent years as a means of replacing the conventional analogelectronics used in radiation detection and measurement. The digital processing approach hasbeen demonstrated for:

1. improving energy resolution in _-ray or x-ray spectroscopy through such methods asCompton continuum suppression (Aspacher 1994) or optimum filtering (Fazzi 1998);

2. obtaining digital pulse shape or pulse height spectra (Simes 1995);3. improving throughput rates by using fast recursive digital algorithms (Jordanov et al.

1994) or reducing dead time (O’dell et al. 1999);4. estimating the occurrence time of events for coincidence or anti-coincidence through

least-mean-squares algorithms or linear algorithms (Geraci 1999); and5. separating different radiation type-induced pulses in a phoswich detector by pulse

shape discrimination (White and Miller 1999).Although most of these applications utilized hardware implementation of digital signal

processing algorithms, a more efficient alternative approach is to employ software algorithms.Software algorithms for pulse height, pulse shape discrimination, and dual parameter analyseshave been developed by several research groups using digital oscilloscopes (White and Miller1999; DeVol et al. 1999). The software implementation can be improved by decreasing processtime with a CPU optimized for fast digital signal processing. For prototyping, softwareimplementation is performed using an appropriate digitizer, but once the algorithm for the signalprocessing is optimized, it can be translated to a hardware implementation form using field-programmable gate arrays (FPGA) (Jordanov and Knoll 1995; Warburton et al. 1999; Bolic andDrndarevic 2002). Processed data by the FPGA then will be fed to a fast digital signal processorto drastically increase system throughput rates.

There are several advantages to digitally processing radiation pulses over theconventional analog approach (DeVol et al. 1999). For example: 1) the pulse processingalgorithm is easy to edit, because changes are made through the software; 2) the algorithm isstable and reliable, since it is not affected by thermal noise or other fluctuations; 3) it is possible

A[n] B[n]

C[n]


12

to make the detection equipment portable by eliminating most of the bulky analog electronics; 4)it is convenient to post-process the pulses; 5) it is more cost-effective; and 6) effects, such aspile-up (Chrien et al. 1986), ballistic deficit (Georgiev and Gast 1993) and charge trapping (Hesset al. 1994) can be corrected or eliminated at the processing level. Additionally, signal captureand processing can be based more easily on coincidence criteria between different detectors ordifferent parts of the same detector (Warburton et al. 1999).

Neural Networks. Neural networks are composed of parallel mathematical elementsdesigned to mimic the biological nervous system (Wunsch et al. 2003). As in nature, thenetwork function is determined largely by the connections between elements. A neural networkcan be trained to perform a particular function by adjusting the significance of the connections(i.e., weights) between elements. Commonly, neural networks are trained, based on acomparison of the true output and the target output, so that a particular input leads to a specifictarget output. Typically, many such true/target output pairs, coupled with the input dataset, areused to train a network in what is referred to as "supervised learning".

Batch training of a network occurs iteratively by making weight and bias changes basedon an entire set (batch) of input vectors. Incremental training changes the weights and biases ofa network as needed after presentation of each individual input vector. Incremental training ofthis sort is referred to as "adaptive" training. Supervised training methods are commonly used,but other networks can be obtained from unsupervised training techniques or from direct-designmethods. Neural networks have been trained to perform complex functions in various fields ofapplication including pattern recognition, identification, classification, speech, vision, andcontrol systems. The PI has utilized neural networks in previous work to aid in environmentaldecision-making (Hamby 1995; Hamby and Famiano 1995).

Neural networks for this proposed application will be trained and parameterized frommeasured response functions for two specific uses, both utilizing the pattern recognitioncapabilities of properly trained networks: 1) to aid the identification of radiation-type andnuclide; and 2) to facilitate determination of a most-probable xenon source by isotopic ratioanalysis. Characteristics particular to given radiation fields and peculiarities of specific radiationtypes will be cataloged and used to train the neural network to recognize these characteristics andidentify the radiation components incident on the detector.

Our Prototype 3rd Generation Phoswich. In a portion of our current research (DE-FG07-05ID14704), we are investigating the use of a 3rd generation triple-phosphor sandwich detector(which we call GEN3) design for real-time, digital beta radiation spectroscopy and dosimetry ina mixed beta/gamma radiation field (Fig. 10). To facilitate pulse shape discrimination, thescintillators are chosen to have sufficiently different decay times (Table 1).

The first two layers of GEN3 were chosen specifically for beta spectroscopy, with thethird layer intended for gamma-ray measurements. The first layer is a very thin inorganic(CaF2:Eu) and the second layer is a plastic (BC-400); both are sensitive to beta radiations andtheir total thickness is enough to stop betas with energies up to 3.18 MeV. The thinness of theCaF2:Eu scintillator minimizes the likelihood of gamma ray interactions and increases theprobability that an incident beta particle will traverse the inorganic and enter the plasticscintillator before stopping.

The design is such that the first layer must be penetrated by the incident betas for a pulseto be recorded as a beta-induced pulse. In other words, all fast pulses originating from the plasticscintillator without having a slow component (from the CaF2:Eu layer) are rejected since they areconsidered to be gamma-induced pulses.


13

Fig, 10. The GEN3 phoswich detector.

The third layer (NaI:Tl) is an inorganic scintillator and is included for gamma-raymeasurements. Since plastic scintillators cannot be fully dried and would ultimately hydrate anddestroy the performance of the crystal, the NaI:Tl is completely isolated by a thin quartz opticallayer. Additionally, an hermetic seal is made around the quartz and the PMT. The quartz acts asa light guide and does not affect the light produced in neither the CaF2:Eu nor the plastic. In amixed beta/gamma field, the energy distribution of gamma-rays detected by the NaI:Tl can bedistorted by: (1) beta particles possessing enough energy to reach this layer; or (2) scatteredgamma rays, due to Compton interactions originating from other layers (mostly from the plastic).Given the thickness of the first two layers and the quartz, only beta particles of very high energy(> 6.7 MeV) can reach the third layer. However, since absorption of scattered gamma-rays in thethird layer produces a pulse with more than one timing component, the second interference canbe eliminated by rejecting that pulse in an anti-coincidence filtering with other layers. Thecriteria for accepting or rejecting a pulse are presented in Table 2. By using this design andutilizing digital pulse shape discrimination, we will be able to measure beta and gamma-rayenergy distributions simultaneously with minimal crosstalk in either spectrum.

Table 1. Scintillation material characteristics for the GEN3 detector.

Scintillator Density(g/cm3)

Wavelength ofMax

Emission(nm)

Ligth Output% of NaI:Tl

Index ofRefraction

PrincipleDecay

Constant (ns)CaF2:Eu 3.19 435 50 1.47 900

BC-400 1.032 423 26 1.58 2.4

NaI:Tl 3.67 415 100 1.85 230

To study seven possible combinations of light components corresponding to possiblegamma and/or electron interactions within each layer of the phoswich detector, an analysis using


14

MCNP has been performed separately for monoenergetic photons and electrons. Table 2, forinstance, shows the results of this study for 1.0 MeV photons and electrons. Columns five andsix of Table 2 indicate total probabilities of the event (pulse detected containing markedcomponents) for that radiation type; events (detector response) with energy deposition less than20 keV were excluded as noise. To better visualize the results of Table 2 and to identifyimportant events in a mixed beta/gamma field, a 3-D histogram (Fig. 11) was generated byresolving the three decay components from each signal pulse and attributing each axis of thisplot to the amount of energy deposited in the corresponding phoswich layer.

Table 2. Total probabilities for seven possible combinations of three signal light components

Event Scintillation Layers Total Probability**

1CaF2:Eu

2BC-400

3NaI(Tl)

Gamma Beta

1 _ 0.022 0.370

2 _ _ 0.0028 0.585

3 _ 0.102 0.0308

4 _ _ 0.0022 0.001

5 _ _ 0.0103 0.0001

6 _ _ _ 0.0003 0.0002

7 _ 0.080 0.00009 [X] Pulse detected containing marked component(s)

** Total probabilities were calculated using MCNP for 1.0 MeV photons and electrons.

If considering the major interaction scenarios, event 1 in Table 2 corresponds to energydeposition of incident electrons or photons in only the first layer (CaF2:Eu). In terms of pulseshape, therefore event 1, an interaction in the first layer only, generates a single-light componentpulse (Td = 900 ns) with probabilities of 0.022 and 0.37 for 1.0 MeV photons or electrons,respectively. Note that event 1 relates to all interactions in which the radiation energy is eitherfully absorbed in the CaF2 or, after releasing a fraction of its energy, with no interaction in otherscintillation layers, escapes the detector.

In event 2, since 1.0 MeV electrons have enough energy to penetrate into the secondlayer, a pulse with double decay components (from CaF2 and BC-400) can be produced. Theprobability for this event due to 1 MeV electrons is 0.585. For 1 MeV photons, the probabilityof an interaction in the first and second layers was calculated to be 0.0028. Compton-photoelectric and Compton-Compton are among successive gamma interactions which can takeplace in the first and second layers and generate event 2, but the probability of such aninteraction sequence is quite low.

Considering the relatively high probabilities of events 1 and 2 for electrons, it is highlyreasonable to attribute these types of pulses to incident electrons. In fact, with this configurationand the short mean-free-path of incident electrons, a passive discrimination mechanism isprovided for the electron-induced pulses to be separated from the gamma-induced pulses in thefirst two layers. This mechanism would be more efficient if we only consider event 2 for betadetection. But, for covering a wide beta energy range, the first layer must be very thin. Current


15

machining technology allows a minimum thickness of about 0.3 mm for CaF2 and thus at thispoint both events 1 and 2 are considered as beta-induced pulses.

Detector output from events 1 or 2 (Table 2) will fall into region 1 (R1) of Fig. 11.Therefore, this plane, including the CaF2 axis will cover almost all pure beta interactions in thefirst two layers. Since the related gamma and electron probabilities in R2 (BC-400 axis in Fig.11and event 3 in Table 2) have similar values, no useful information can be extracted from theseevents and therefore are rejected. Rejection of these pulses efficiently minimizes gammacrosstalk, mostly due to Compton interactions within BC-400, in measured beta spectra. This isone of the important features of the XEPHWICH design over the ARSA detector, in whichCompton interactions in ARSA beta cells due to incident gamma rays introduce some errors inthe 2-D gamma/beta coincidence plot (Rynes et al. 2004; Reeder et al. 2004).

Fig. 11. 3-D histogram to locate the three decay components of each signal pulse.

To eliminate any beta crosstalk in measured gamma spectra, the thickness of the first twolayers and the quartz has been designed in our 3rd generation prototype to stop any beta particlein the field. Pulses due to event 7, therefore, can be considered as gamma-induced pulses, e.g.events falling on the NaI axis (R4 in Fig. 11). Event 5 (R3 in Fig. 11) similar to event 3 (R2),has no functional information for discrimination of beta and gamma radiations and is thereforerejected. All other locations of Fig. 11, including the plane of CaF2/NaI, can be considered asgamma/beta coincidence events, since their probabilities for single-particle interactions are quitesmall (events 4 and 6 of Table 2). It should be noted that the GEN3 detector is not optimized forthe detection of radioxenon isotopes (131mXe, 133mXe, 133Xe and 135Xe) necessary for nuclearexplosion monitoring. In Phase 1 of our proposed research, however, all corresponding photon

CaF2

BC-400

R1

R2

R3

R4

NaI


16

and electron particle energies will be considered to optimize a new multi-layed phoswichdetector (XEPHWICH) to respond to the events of Table 2. Radon Emanation Following Seismic Activity. The OSU research group will proposeherein to develop what we will call “radon correlation analysis”, a novel approach to utilize,rather than discard, the radon signature in air and to correlate the radon/xenon response forpositive identification of subsurface nuclear detonations. A number of researchers (Wakita1996; Bassignani et al. 1999; Planinic et al. 2000; Choubey et al. 2004) have shown that radonconcentrations in soils, spring water, and groundwater vary following seismic activity.Observations by Wakita (1996) demonstrate that long-term variations in groundwater radonconcentrations increase by about 3-10%, dependent on relative seismic activity. Planinic et al.(2000) suggest that temporal variations of radon emanation from soil and water can give earlyevidence of tectonic disturbances, and that radon-emanation detectors might be used forearthquake prediction. Likewise, Choubey et al. (2004) found that radon levels in spring waterwere affected by a particular earthquake in India and they concluded that there is a significantrelationship between groundwater radon concentration and earthquakes. The work of Yakovleva(2003), however, shows that radon flux density from the earth (mBq m-2 s-1) is a more sensitiveindicator of seismic activity than is soil radon concentration. Therefore, measurements ofradioxenons, particularly 135Xe, that are accompanied by an increase in radon concentration mayprove useful for identification of underground nuclear explosions. We plan to investigate thispossibility and utilize the radon signature in our analyses.

Xenon Isotopic Ratio Analysis. The Comprehensive Nuclear-Test-Ban Treaty (CTBT)calls for the development of radioxenon monitoring systems to detect four specific isotopes ofxenon: 131mXe, 133mXe, 133Xe, and 135Xe. Sources of these radionuclides typically include medicaluses, spent fuel reprocessing, reactor operations, and nuclear explosions. Because of thedifferences in production rates, decay rates, process amounts, etc., the numerical values ofvarious concentration ratios of the xenon isotopes can be used to predict the source of detectedradioxenon. Three ratios in particular (131mXe:133Xe, 133mXe:133Xe, and 135Xe:133Xe) have receivedthe most attention. Heimbinger et al. (2002) note that a high 131mXe:133Xe ratio, along with verylow concentrations of 135Xe or 133mXe, is indicative of an aged medical source or a nuclear fuelreprocessing source. Several researchers (Bowyer et al. 1999; Arthur et al. 2001; Heimbigner etal. 2002; Ringbom et al. 2003; Reeder et al. 2004) indicate that the 133mXe:133Xe ratio, when high,suggests nuclear weapons tests, whereas a low ratio indicates reactor operations. Likewise, the135Xe:133Xe ratio, with other indicators, can be used to determine weapons test versus reactoroperations (Arthur et al. 2001; Ringbom et al. 2003). A 135Xe:133Xe ratio resulting from a nucleardetonation is several orders of magnitude greater than for a reactor at equilibrium, whereas ahigh ratio with a lack of 133mXe indicates a nuclear reactor not at equilibrium (Heimbigner et al.2002). Additionally, because of the lack of 135Xe burnup, the 135Xe:133Xe ratio increases quitedramatically, from 0.5 to 10, in the first day following reactor shutdown (Heimbigner et al.2002). In the work proposed herein, we will examine further options for isotopic ratio analysisand merge those findings with neural network routines to aid the source identification task of theradioxenon detection system.

THE PROPOSED RESEARCHResearch Objectives. The research team will develop a task-specific atmospheric

radioxenon analysis system that includes a radioxenon-optimized, multi-layer phoswich detector(XEPHWICH) and a digital, firmware-based pulse-shape processing structure (Figs. 12 and 13).The system will be designed such that it can be retrofitted to the existing ARSA framework inthe four-cell arrangement to allow continuous sampling. Signal management will be entirely


17

digital and will be executed with an infrastructure of high-speed, field-programmable gate arrays(FPGA) and digital signal processor (DSP) circuitry (Fig. 13). The completed system willrequire at least 2/3 less power than ARSA and will provide positive identification of radioxenonsoriginating from nuclear explosions by isotopic ratio analysis, neural-network-assistedidentification routines, and novel radon correlation analysis. The 2-D β/γ electronic-coincidenceplots (Reeder and Bowyer 1998; Bowyer et al. 1999; Ely et al 2003; Rynes et al. 2004) currentlygenerated by ARSA will be replaced with 3-dimensional light/electronic-coincidencexenon/radon histograms, along with pure beta and pure gamma differential pulse-height spectra,to positively identify/quantify radioxenon isotopes. More technical details about theXEPHWICH are given in the Technical Approach section.

Isolation Medium Layer 2 Layer 1

PMTPMT

Optical Isolation

Layer 3 Layer 3

Fig. 12. Prototypic saddle-XEPHWICH design with one sample cell.

Long FIFO

Short FIFO

USB Processor

Host PC

PMT 1

Digital Signal Processor

AnalogDigital

DACAnalog output 1

ADC12 bit-100 MHz

Nyquist Filter

Analog Signal Conditioning

FPGA

Short FIFO

PMT 2

DACAnalog output 2

ADC12 bit-100 MHz

Nyquist Filter


FPGA

Long FIFO

Short FIFO

USB Processor

Host PC

PMT 1

Digital Signal Processor

AnalogDigital

DACAnalog output 1

ADC12 bit-100 MHz

Nyquist Filter


FPGA

Short FIFO

PMT 2

DACAnalog output 2

ADC12 bit-100 MHz

Nyquist Filter


FPGA

Fig. 13. Simplified block diagram of our 2-channel digital pulse processing unit (DPP2).

DPP2, A Two-Channel Digital Pulse Processor Unit: DPP2, consists of two analoginputs, two analog testing outputs and one high-speed USB 2.0 interface (Fig.13). The detectoroutputs, PMT 1 and 2, are fed directly to the DPP2. The signal pulses from each PMT thenundergo analog signal conditioning (ASC) by passing through a variable gain amplifier in orderto ensure that they are within the range of the ADC. After processing through low-pass analogNyquist filters, the signal pulses in each channel are sampled by a 12-bit/100 MHz ADC toprovide sufficient time and amplitude resolution. The sampled pulse may be reconstructed if,and only if, the ADC sampling rate is greater than or equal to twice the highest frequency


18

component contained in the waveform (Nyquist-Shannon Theorem). The Nyquist filter stretchesthe signal pulse (Bardelli et al. 2002), thereby minimizing the need for higher sampling rates tocapture very fast pulses. The digitized pulses in each channel then pass through a FieldProgrammable Gate Array (FPGA) and from there to a circular short FIFO (First-In-First-Out)memory buffer. The FPGA, which consists of gates, flip-flops, distributed and block RAM’s,has been described as a “highly parallel configurable digital signal processor” (Bolic et al. 2002).The FPGA allows calculations to be carried out in parallel, thereby significantly increasing thedigital processing speed. It is fully reprogrammable, thereby allowing different application-specific configurations to be used with the same hardware. For prototyping, our programmedFPGA (Fig. 14) will perform the tasks of noise reduction, pulse detection, time stamping, pile-uprejection, and the resolving of pulse time components (Fig. 8). The FPGA configuration will bedeveloped using the high-level language VHDL (Yalamanchili, 2001). The FPGA will cause anevent trigger to be issued if it detects a valid pulse. The role of the trigger is to recognize whenan event has arrived such that subsequent processing can occur (i.e., a signal pulse determined tobe valid versus pileup pulses, pulses due to cosmic radiations or pulses with amplitudes belowthe threshold level, all of which would be rejected). The trigger causes the short FIFO (Fig. 13)to stop filling its memory with incoming samples and also informs the Digital Signal Processor(DSP) that a valid radiation pulse in one of the short-FIFO buffers is ready to be processed. Oncompletion of reading the short-FIFO buffer, the FPGA will be informed to rearm itself, bysetting a flag bit, and enable the short-FIFO buffer to accept the next in a line of radiation signalpulses. Pre-processed pulses then will be stacked in the long-FIFO buffer (Fig. 13) for furtherprocessing by the DSP. For development purposes, the DPP2 unit will be equipped with tworeconstruction digital-to-analog converters (DAC). The DAC allows the manipulated waveformto be viewed easily on an oscilloscope and compared to the raw signal directly from the detector.

Figure 14. Signal processing schematic for the saddle-geometry XEPHWICH.


19

Relevance. Topic 7 of the RFP (DE-SC52-05NA26703) states that proposals are soughtthat will “enhance the United States’ ability to monitor foreign nuclear tests through improvedinstrumentation used for the detection of radioactive particles and xenon and processes/methodsthat facilitate improved data discrimination.” The request states the necessity of beta detectorsfor improved radioxenon monitoring and simple, innovative, low-cost techniques that achievehigh accuracy and field-condition stability. Methods that eliminate re-calibration and manualintervention for at least 6-month periods are desired, along with “techniques that minimizecomplexity, power usage, and costs.” Additionally, Topic 7 seeks data discrimination techniques“to improve the quantification and identification of radionuclides in the presence of high naturalbackgrounds …” and “… methods to correlate specific radionuclides.”

The PI and his research team propose a radioxenon detection system that will address allof these requirements. The research group has been investigating the problem of gamma-discriminated beta spectroscopy for a number of years using phoswich techniques and signal-component analysis. This work will be expanded to develop a saddle geometry, radioxenon-optimized phoswich, at a power and cost reduction of greater than 70%, with supporting digitalsignal processing for xenon/radon discrimination that will fit the current ARSA infrastructure(Fig. 14). Cost and power reductions come primarily from the reduction in the number of PMTsrequired of the current ARSA (from 12 to 4, by eliminating the 8 beta-cell PMTs) and theelectronics associated with an analog system. The proposed system, being digital, also will havethe advantage of remote auto-calibration and single-pulse analysis for background reduction andradon correlation analysis. Advantages of the proposed XEPHWICH detection system over theARSA method are given in Table 3.

Table 3. Advantages of the XEPHWICH system over the ARSA configuration.

Current ARSA System: Proposed XEPHWICH System:

Pulse Processing

analog pulse-height spectral analysis; gamma-or beta-gating; pulse-height spectra archival;analog electronic noise; pulse pileup; analogpulse amplitude; on-site calibrations

signal-shape (3 component) analysis; light-coincidencemethods; raw signal archival; digital signal processing anddiscrimination using on-board field-programmable gate array(FPGA) and a digital signal processor (DSP); digitalcancellation of jitter; noise elimination; elimination of pulsepileup; remote software edit; remote calibration

Detector Design ElementsCompton scatter interference in beta signal;delayed coincidence gating; phoswichresearchers (Ely et al 2003) unable to resolvebeta-gamma coincidence pulses with 2-layerdesign; spurious counts due to varying lightefficiencies in beta tube

greatly reduced gamma cross-talk due to elimination of signalscreated in the beta detectors from Compton interactions;instantaneous light coincidence; proven concept with the triple-layer design via transport modeling and prototypic analyses;uniform efficiency and reduced MDCs

Firmware Control & Post-Processing

radon signal discriminated and excluded;requires in-field setup/calibration; 2-D pulseheight spectra; 7Be used for QA/QC

radon signal discriminated and utilized for positiveidentification of nuclear detonation; FPGA and DSP allowremote reprogramming; signal autocorrelation and 3-Dhistograms; capability for continuous calibration

Physical/Cost/Power

12 PMTs; independent β/γ scintillators;optically separated; a 2-layer phoswich hasbeen suggested by Eli et al. 2003

4 PMTs (more than 70% power & cost reduction); 3-layerphoswich analysis (3 component); preconcentration notnecessary since radon signal is utilized; optimized specificallyfor radioxenons


20

REFERENCES

Arthur, R.J.; Bowyer, T.W.; Hayes, J.C.; Heimbigner, T.R.; McIntyre, J.I.; Miley, H.S.; Panisko, M.E.Radionuclide measurements for nuclear explosion monitoring. 23rd Seismic Research Review. 57-63;2001.

Aspacher, B.; Rester, A. C. Compton continuum suppression by digital pulse shape analysis Part II.Nuclear Instruments & Methods in Physics Research, Section A: Accelerators, Spectrometers,Detectors and Associated Equipment. 338A: 516-521; 1994.

Bambynek, M.; Flühs, D.; Quast, U.; Wegener, D.; Soares, C.G. A high precision, high-resolution andfast dosimetry system for beta sources applied in cardiovascular brachytherapy. Medical Physics.27(4): 662-667; 2000.

Bardelli, L.; Bini, M.; Poggi, G.; Taccetti, N. Application of digital sampling techniques to particleidentification in scintillation detectors. Nuclear Instruments & Methods in Physics Research, SectionA: Accelerators, Spectrometers, Detectors and Associated Equipment. A491: 244-257; 2002.

Bassignani, A.; Colombo, G.; Degli Esposti, L.; Fresca Fantoni, R.; Giacomelli, G.; Maltoni, G.;Mandrioli, G.; Mascoli, M.; Materazzi, F.; Patrizzi, L.; Sirri, G.; Ugolotti, D. Measurement of theradon concentration in the water from the Gran Sasso fault. LNGS-EXP 20/99. LNGS Annual Report.pg. 209-215; 1999.

Beddar, A.S.; Mackie, T.R.; Attix, F.H. Water-equivalent plastic scintillation detectors for high-energybeam dosimetry: I. Physical characteristics and theoretical considerations. Physics in Medicine andBiology. 37(10): 1883-1900; 1992a.

Benchekroun, D.; Benrachi, F.; Chambon, B.; Cheynis, B.; Drain, D.; Pastor, C.; Vagneron, L.;Desesquelles, P.; Giorni, A.; Heuer, D.; Lleres, A.; Viano, J.B.; Fabris, D.; Nebbia, G.; Viesti, G.Scintillating gas proportional phoswiches. Nuclear Instruments & Methods in Physics Research,Section A: Accelerators, Spectrometers, Detectors and Associated Equipment. 335A(3):503-508;1993.

Bingo, K.; Suga, S.; Kajimoto, Y.; Numakunai, T. Beta-ray survey meter for measuring absorbed doserate independently of beta-ray energy. Health Physics. 39: 21-28; 1980.

Bolic, M.; Drndarevic, V. Digital gamma-ray spectroscopy based on FPGA technology. NuclearInstruments & Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors andAssociated Equipment. A482: 761-766; 2002.

Bowyer, T.W.; McIntyre, J.I.; Reeder, P.L. High-sensitivity detection of xenon isotopes via beta-gammacoincidence counting. 21st Seismic Research Symposium. 231-241; 1999.

Bush-Goddard, S.P., “Beta Spectroscopy Using Deconvolution and Spectral Stripping Techniques with aTriple Layer Phoswich Detector”, University of Michigan, Ann Arbor, MI: Doctoral Dissertation,April 2000.

Choubey, V.M.; Mukherjee, P.K.; Ramola, R.C. Radon variation in spring water before and after Chamoliearthquake, Garhwal, Himalaya, India. 11th International Congress of the International RadiationProtection Association, Madrid, Spain. 23-28; May 2004.

Chrien, R.E.; Sutter, R. Noise and pileup suppression by digital signal processing. Nuclear Instruments &Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and AssociatedEquipment. 249A: 421-425; 1986.

Clift, M.A.; Sutton, R.A.; Webb, D.V. Water equivalence of plastic organic scintillators in megavoltageradiotherapy bremsstrahlung beams. Physics in Medicine and Biology. 45: 1885-1895; 2000.

Cross, W.G.; Marr, J.D. Radiation Dosimetry. Atomic Energy of Canada. Report 802: 58-60; 1960.Dahlbom, M.; MacDonald, L.R.; Eriksson, L.; Paulus, M.; Andreaco, M.; Casey, M.E.; Moyers, C.

Performance of a YSO/LSO phoswich detector for use in a PET/SPECT system. 1996 NuclearScience Symposium and Medical Imaging Conference (NSS/MIC). 3-9 Nov. 1996; Anaheim, CA.IEEE Transactions on Nuclear Science. 44(3):1114-1119; 1997.

de Boer, S.F.; Beddar, A.S.; Rawlinson, J.A. Optical filtering and spectral measurements of radiation-induced light in plastic scintillation dosimetry. Physics in Medicine and Biology. 38: 945-958; 1993.


21

de Sousa, M.C.; Aubert, B.; Ricard, M. Evaluation of physical performance of a scintillation dosemeterfor patient dosimetry in diagnostic radiology. British Journal of Radiology. 73: 1297-1305; 2000.

DeVol, T. A.; Tan, H.; Fjeld, R. A. Dual parameter analysis of CsI:TIJPMT with a digital oscilloscope.Nuclear Instruments & Methods in Physics Research, Section A: Accelerators, Spectrometers,Detectors and Associated Equipment. 435A: 433-436; 1999.

Ely, J.H.; Aalseth, C.E.; Hayes, J.C.; Heimbigner, T.R.; McIntyre, J.I.; Miley, H.S.; Panisko, M.E.;Ripplinger, M. Novel beta-gamma coincidence measurements using phoswich detectors. 25th SeismicResearch Review. 533-541; 2003.

Fazzi, A.; Varoli, V. A digital spectrometer for ‘optimum’ pulse processing. IEEE Transactions onNuclear Science. 45(3): 843-848; 1998.

Flühs, D.; Heintz, M.; Indenkämpen, F.; Wieczorek, C. Direct reading measurement of absorbed dosewith plastic scintillators- the general concept and applications to opthalmic plaque dosimetry. MedicalPhysics. 23(3): 427-434; 1996.

Fox, D.; Bowman, D.R.; Ball, G.C.; Galindo-Uribarri, A.; Hagberg, E.; Horn, D.; Beaulieu, L.;Larochelle, Y. Calibration of plastic phoswich detectors for charged particle detection. NuclearInstruments & Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors andAssociated Equipment. 374(1):63-69; 1996.

Frontera, F.; Fiume, D.; Poulsen, J. M.; Taiocchi, G. F.; Basili, A.; Silvestri, S. Pulse shape analyzer forthe multiple phoswich detector on the LAPEX experiment. Nuclear Instruments & Methods inPhysics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment.324A(3):589-597; 1993.

Georgiev, A.; Gast, W. Digital pulse processing in high resolution, high throughput gamma-rayspectroscopy. IEEE Transactions on Nuclear Science. 40(4): 770; 1993.

Geraci, A.; Ripamonti, G. A new on-line digital solution for event timing setups. Nuclear Instruments &Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and AssociatedEquipment. 422A:337-340; 1999.

Hamby, D.M. The use of neural networks in environmental decision making. Michigan Section of theAmerican Nuclear Society. Ann Arbor, MI; September 28, 1995.

Hamby, D.M.; Famiano, M.A. Neural network applications in environmental decision making. GreatLakes Chapter of the Health Physics Society Spring Symposium. Ann Arbor, MI; March 29, 1995.

Hayes, J.C.; Abel, K.H.; Bowyer, T.W.; Heimbigner, T.R.; Panisko, M.E.; Reeder, P.L.; McIntyre J.I.;Thompson, R.C.; Todd, L.C.; Warner, A. Operations of the Automated RadioxenonSampler/Analyzer – ARSA. 21st Seismic Research Symposium. pg. 249-260; 1999.

Heimbigner, T.R.; McIntyre, J.I.; Bowyer, T.W.; Hayes, J.C.; Panisko, M.E. Environmental monitoring ofradioxenon in support of the radionuclide measurement system of the international monitoringsystem. 24th Seismic Research Review. 694-700; 2002.

Hess, R.; De Antonis, P.; Morton, E. J.; Gilboy, W. B. Nuclear Instruments & Methods in PhysicsResearch, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment. 353A: 76;1994.

Horowitz, Y.S.; Horowitz, A.; Hirning, C.R.; Yuen, P.; Cross, W.; Aikens, M. The cog-beta rayspectrometer for mixed field beta/photon dosimetry. Radiation Protection Dosimetry. 47(1/4): 415-418; 1993.

Johnson, L.O.; Alvarez, L.; Hoggan, J.M.; Dickson, R.L. Pulse shape discrimination in a portable beta-gamma dose rate meter. IEEE Transactions on Nuclear Science. NS-30(1): 543-546; 1983.

Jordanov, V. T.; Knoll, G. F.; Huber, A. C.; Pantazis, J. A. Digital techniques for real-time pulse shapingin radiation measurements. Nuclear Instruments & Methods in Physics Research, Section A:Accelerators, Spectrometers, Detectors and Associated Equipment. 353A: 261-264;1994.

Jordanov, V.T.; Knoll, G.F. Digital pulse-shape analyzer based on fast sampling of an integrated chargepulse. IEEE Transactions on Nuclear Science. 42(4): 683-687; 1995.

Kamae, T.; Gunji, S.; Hirayama, S.; Miyazaki, S.; Nagato, T.; Nakao, A.; Sekimoto, Y.; Suzuki, K.;Takahashi, T.; Tamura, T.; Tanaka, M.; Yamaoka, N.; Yamagami, T.; Nomachi, M.; Murakami, H.


22

Well-type phoswich counter for low-flux x-ray/ gamma -ray detection. IEEE Transactions on NuclearScience. 40(2):204-207; 1993.

Kirov, A.S.; Hurlbut, C.; Dempsey, J.F.; Shrinivas, S.B.; Epstein, J.W.; Binns, W.R.; Dowkontt, P.F.;Williamson, J.F. Towards two-dimensional brachytherapy dosimetry using plastic scintillator: Newhighly efficient water equivalent plastic scintillator materials. Medical Physics. 26(8): 1515-1523;1999.

Knoll, G.F. Radiation Detection and Measurement. 3rd Edition. John Wiley and Sons, Inc. New York.2000.

Kriss, A.; Hamby, D.M. Scintillation beta dosimetry and spectroscopy utilizing a large area avalanchephotodiode. Proceedings of the Forty-eighth Annual Meeting of the Health Physics Society. SanDiego, CA. Health Physics. 84(6): S168; 2003.

Kriss, A.A.; Hamby, D.M. Beta spectroscopy with a large-area avalanche photodiode module and aplastic scintillator. Nuclear Instruments and Methods in Physics Research - Section A. 525(3): 553-559; June 2004a.

Kriss, A.A.; Hamby, D.M. Measurement and modeling of dose from external beta emitters utilizingplastic scintillator volumes. Health Physics. submitted March 2004b.

Kriss, A.A.; Hamby, D.M. Monte Carlo based enhancement of beta spectra by removal of fractionalenergy deposition events in a plastic scintillator. Nuclear Instruments and Methods in PhysicsResearch – Section A.. submitted April 2004c.

Langenbrunner, J. L.; Morris, C. L.; Whitton, R. M. CsI-phoswich detector for charged-particleidentification. Nuclear Instruments and Methods in Physics Research, Section A: Accelerators,Spectrometers, Detectors and Associated Equipment. 316(2-3):450-451; 1992.

Lautridou, P.; Eudes, P.; Germain, M.; Guilbault, F.; Laville, J.L.; Rahmani, A.; Reposeur, T.; Roy, D.Extended pulse shape discrimination capabilities using a CsI-BGO phoswich. Nuclear Instruments &Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and AssociatedEquipment. 373(1):135-138; 1996.

Lum, K.S.K.; Mohr, J.J.; Barret, D.; Grindlay, J.E.; Manandhar, R.P. Simulations and measurements ofthe background encountered by a high-altitude balloon-borne experiment for hard X-ray astronomy.Nuclear Instruments & Methods in Physics Research, Section A: Accelerators, Spectrometers,Detectors and Associated Equipment. 396(3):350-359; 1997.

Mainardi, R.T.; Bonzi, E.V.; Martinez, V.D. Design of tissue equivalent scintillators for precisedosimetry purposes. Radiological Physics Chemistry. 50(20): 159-163; 1997.

Martz, D.E .; Rich, B.L.; Johnson, L.O. A portable beta spectrometer for tissue dose measurement.Radiation Protection Dosimetry. 14(2): 183-186; 1986.

McIntyre, J.I.; Aalseth, C.E., Bowyer, T.W.; Hayes, J.C.; Heimbigner, T.R.; Morris, S.; Reeder, P.L.Triple coincidence radioxenon detector. 26th Seismic Research Review. 581-587; 2004.

Nagornaya, L.L.; Zelenskaya, O.V.; Budakovsky, S.V.; Chichikalyuk, Yu. A. Highly efficient alpha, beta,gamma-spectrometer with low background. IEEE Transactions on Nuclear Science. 43(3):1284-1286;1996.

O’dell, D.M.C.; Bushart, B. S.; Harpring, L. J.; Moore, F. S.; Riley, T. N. Zero dead time spectroscopywithout full charge collection. . Nuclear Instruments & Methods in Physics Research, Section A:Accelerators, Spectrometers, Detectors and Associated Equipment. 422A: 363-367; 1999.

Penn, D.; Biegalski, S.; Biegalski, K. Status of the automated radioxenon sampler analyzer, and theinternational monitoring system international noble gas experiment. 24th Seismic Research Review.pg. 701-710; 2002.

Penn, D. Status of the United States international monitoring system radionuclide monitoring stations.25th Seismic Research Review. pg. 552-567; 2003.

Planinic, J.; Radolic, V.; Culo, D. Searching for an earthquake precursor: temporal variations of radon insoil and water. FIZIKA B (Zagreb). 9(2), 75-82; 2000.

Qi, Z.; Guo, Z.; Zhan, W.; Zhou, J.; Liu, G.; Zhang, W.; Wang, J.; Lin-Y.; Zhu, Y.; Xu, H.; Xie, Y. Aphoswich scintillation detector telescope. Nuclear Techniques. 20(6):352-355 1997a.


23

Qi, Z.; Guo, Z.; Zhan, W.; Liu, G.; Zhou, J.; Zhang, W.; Wang, J.; Lin, Y.; Zhu, Y.; Xu, H.; Xie, Y. Ascintillant multidetector for measurement in heavy ion reactions at intermediate energy. High EnergyPhysics and Nuclear Physics. 21(7): 577-582; 1997b.

Reeder, P.L.; Bowyer, T.W. Xe isotope detection and discrimination using beta spectroscopy withcoincident gamma spectroscopy. Nuclear Instruments & Methods in Physics Research, Section A:Accelerators, Spectrometers, Detectors and Associated Equipment. 408: 582-590; 1998.

Reeder, P.L.; Bowyer, T.W.; McIntyre, J.I.; Pitts, W.K.; Ringbom, A.; Johansson, C. Gain calibration of ab/g coincidence spectrometer for automated radioxenon analysis. Nuclear Instruments & Methods inPhysics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment. 521:586-599; 2004.

Ringbom, A.; Larons, T.; Axelsson, A.; Elmgren, K.; Johansson, C. SAUNA – a system for automaticsampling, processing, and analysis of radioactive xenon. Nuclear Instruments & Methods in PhysicsResearch, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment. 508: 542-553; 2003.

Rynes, J.C.; Penn, D.; Donohoe, P. Research and development of radioxenon monitoring systems. 26th

Seismic Research Review. 588-597; 2004.Schindler, S.M.; Cook, W.R.; Hammond, J.; Harrison, F.A.; Prince, T.A.; Wang, S.; Corbel, S.; Heindl,

W.A. GRIP-2: a sensitive balloon-borne imaging gamma-ray telescope. Nuclear Instruments &Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and AssociatedEquipment. 384(2-3):425-434; 1997.

Shen, L.; Catchen, G.L.; Levine, S.H. Experimental and computational techniques for beta-particledosimetry. Health Physics. 53(1): 37-47. 135; 1987.

Simes, J. B.; Simöes, P. C. P. S.; Correia, C.M.B.A. Nuclear spectroscopy pulse height analysis based ondigital signal processing techniques. IEEE Transactions on Nuclear Science. 42(4): 700-704; 1995.

Strauss, M. G.; Keane, A. T.; Reinke, S. A.; Pehl, R. H. L X-ray spectrometry in vivo with a Si(Li)-NaI(Tl) detector. IEEE Transactions on Nuclear Science. 37(1):859-867; 1990.

Swinth, K.L.; Sisk, D.R.; Simons, G.G. A proportional scintillation counter beta spectrometer. IEEETransactions on Nuclear Science. 36(1): 1166-1171. 1989.

Tavakoli-Farsoni, A.; Hamby, D.M.; Bush-Goddard, S.P. A performance study on a triple-layer phoswichdetector for beta spectroscopy. Proceedings of the Forty-ninth Annual Meeting of the Health PhysicsSociety. Washington, DC. Health Physics. 86(6): S144; 2004a.

Tavakoli-Farsoni, A.; Hamby, D.M. Optimization modeling of two multi-layer phoswich detector designsusing MCNP. Health Physics. submitted. Dec. 2004b.

Usuda, S. Development of ZnS(Ag)/NE102A and ZnS(Ag)/stilbene phoswich detectors for simultaneousalpha and beta (gamma) counting. Journal of Nuclear Science and Technology. 29(9):927-929; 1992.

Usuda, S.; Abe, H. Phoswich detectors for flow monitoring of actinide solutions with simultaneous alphaand beta (gamma) counting. Journal of Nuclear Science and Technology. 31(1):73-79; 1994.

Usuda, S.; Abe, H.; Mihara, A. Simultaneous counting of alpha , beta and gamma rays with phoswichdetectors. Journal of Alloys and Compounds. 213:437-439; 1994a.

Usuda, S.; Abe, H.; Mihara, A. Phoswich detectors combing doubly or triply ZnS(Ag), NE102A, BGOand/or NaI(Tl) scintillators for simultaneous counting of alpha, beta and gamma rays. NuclearInstruments & Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors andAssociated Equipment 340A: 540-545; 1994b.

Usuda, S.; Sakurai, S.; Yasuda, K. Phoswich detectors for simultaneous counting of alpha -, beta (gamma)-rays and neutrons. Nuclear Instruments and Methods in Physics Research, Section A:Accelerators,Spectrometers, Detectors and Associated Equipment. 388(1-2):193-198; 1997.

Vapirev, E.I.; Jordanov, T.; Amin, S.; Stoilov, N.; Georgieva, K. Scintillation spectroscopy for beta raydose measurements. Radiation Protection Dosimetry. 64(4): 303-308; 1996.

Vasil'ev I.O.; Volodin A.V. Use of a scintillation spectrometer-dosimeter for separately determiningdoses under conditions with mixed beta and gamma radiation. Measurement Techniques. 39(7):778-781; 1996.


24

Wakita, H. Geochemical challenge to earthquake prediction. Proceedings of the National Academy ofScience. 93: 3781-3786; April 1996.

Wang, C.F.; Lee, J.H.; Chiou, H.J. Rapid determination of Sr-89-90 in radwaste by low-level backgroundbeta counting system. Applied Radiation and Isotopes. 45(2):251-256; 1994.

Warburton, W. K.; Momayezi, M.; Hubbard-Nelson, B.; Skulski, W. Digital pulse processing: newpossibilities in nuclear spectroscopy. Conference on Industrial Radiation and RadioisotopeMeasurement Applications; IRRMA-1999.

Watt, D.E.; Alkharam, A.S. A feasibility study of scintillator microdosemeters for measurement of thebiological effectiveness of ionizing radiations. Radiation Protection Dosimetry. 61(1-3): 211-214;1995.

White, T. L.; Miller, W. H. A triple-crystal phoswich detector with digital pulse shape discrimination foralpha/beta/gamma spectroscopy. Nuclear Instruments & Methods in Physics Research, Section A:Accelerators, Spectrometers, Detectors and Associated Equipment. 422A: 144- 147; 1999.

Williamson, J.F.; Dempsey, J.F.; Kirov, A.S.; Monroe, J.I.; Binns, W.R.; Hedtjärn, H. Plastic scintillatorresponse to low-energy photons. Physics in Medicine and Biology. 44: 857- 871; 1999.

Wissink, L.; Hoefman, M.; Seip, M.; Wilschut, H.W. Particle identification in phoswich detectors bysignal width measurement. Nuclear Instruments & Methods in Physics Research, Section A:Accelerators, Spectrometers, Detectors and Associated Equipment. 397(2-3): 472-474; 1997.

Wunsch, D.C.; Hasselmo, M.; Venayagamoorthy, K.; Wang, D. (eds) Advances in Neural NetworkResearch. Elsevier. New York. 2003.

Yakovleva, V.S. The radon flux density from the Earth’s surface as an indicator of a seismic activity.Proceedings of ICGG7. pg. 28-30; 2003.

Yalamanchili, S. Introductory VHDL from simulation to synthesis. Xilinx Design Series; Prentice Hall,2001.

3. TECHNICAL APPROACH

The proposed work will involve designing, modeling, constructing, and characterizing afourth-generation prototypic radioxenon phoswich detector which can discriminate radiation typeand energy by digital component analysis of signals from multiple scintillating layers andcharacterization of measured spectra for nuclide identification using neural network methods.State-of-the-art, high-speed digital techniques will be applied in this work for enhancedbeta/gamma spectroscopy. A three-phased approach is proposed that will explore twogeometries of phoswich and development of supporting electronics.

Phase 1 - XEPHWICH Proof of ConceptIn Phase 1, our GEN3 design, in a planar geometry, will be optimized for detection of

specific radioxenon isotopes. Optimization will be carried out by MCNP modeling of energiesspecific to xenon and radon while varying the thickness of various scintillating layers. Theoptimized detector, XEPHWICH, will provide the required technical knowledge for designingthe 4π-geometry/saddle-type XEPHWICH detector for the ARSA infrastructure (Phase 3).Phase 1 is comprised of the following tasks:

A. MCNP modeling/optimization of radioxenon phoswich: Monte Carlo N-Particle(MCNP) software will be used to model and optimize the GEN3 detector design for detection offour radioxenon isotopes (131mXe, 133mXe, 133Xe and 135Xe). The corresponding gamma-rays,conversion electrons and beta particles will be considered to optimize the detector response tothe significant events in Table 2.

B. XEPHWICH prototype design and construction: Based on examining differentscintillation materials and configurations via MCNP modeling, a planar-geometry XEPHWICH


25

prototype detector will be designed and constructed. As with the GEN3 design, the first two andthe third scintillation layers will be assigned for electron and photon detection, respectively.

C. Characterization of XEPHWICH response with existing DPP1: Before anyspectroscopy measurements can be performed, the detector must be calibrated. This taskincludes energy and efficiency calibrations for both electron and photon radiations using betaand gamma-ray laboratory sources. The planar XEPHWICH will be exposed to differentradiation fields; pure beta, pure gamma (by blocking beta particles) and mixed gamma/beta.Separate beta and gamma energy spectra will be collected from a mixed gamma/beta sourceusing a 1-channel Digital Pulse Processor (DPP1) developed in our laboratory. Themeasurement will be repeated by placing a filter (of sufficient thickness to stop incident betaparticles) between the source and detector. These measurements will demonstrate the ability ofthe system to discriminate between gamma- and beta-induced pulses. The results of the firstmeasurement will be compared to literature results of those recorded by separate beta andgamma spectrometers. Also, by analyzing the spectral differences between the measured spectra(e.g., γ1:γ2 and β1:β2), the degree of radiation mischaracterization, either from gamma radiationsinto the beta spectrum or from beta particles into the gamma spectrum, will be quantified.

D. Experimental measurements with radioxenon sources: In this step, the planarXEPHWICH will be exposed to a radioxenon gas source. To contain the gas source near thedetector’s window, a cylindrical chamber suitable for these measurements will be designed andinstalled. The resolving algorithm which has been developed for GEN3 will be enhanced toseparate the three signal components. By extracting three values from each signal pulse, a 3-dimensional histogram will provide the framework to identify and quantify radioxenon isotopesin the corresponding gamma/beta coincidence regions.

E. Neural network design/development for 3-D histogram interpretation: This task,being iterative, is devoted to neural network development and optimization. We will use theMATLAB software package to assist in building and testing various neural networks. MATLABallows for large numbers of inputs, multiple layers, and various functional nodes. The intentionis to produce the foundation for developing a system of neural networks that will identifyradioxenon isotopes from the 3-D histogram. To provide the required data for training the neuralnetwork, a library of responses from four radioxenon isotopes (131mXe, 133mXe, 133Xe and 135Xe)will be created experimentally. These training datasets will provide the knowledge base for theadaptive training and enhancement of the neutral networks.

F. Radioxenon identification routine development/application: Planar XEPHWICH,DPP1 and the trained neural network will be joined to identify and quantify radioxenon isotopesfrom: 1) a pure xenon source with a gamma-shielded chamber; 2) a pure xenon source with agamma background; and 3) a xenon source combined with a source of radon and its progeny.

G. Support product integration and final testing: The final task of Phases 1 and 2 is toprovide support for the final task of Phase 3 in which the entire product (XEPHWICH and digitalsignal processing) will be integrated and readied for implementation into the existing ARSAstructure.

Phase 2 - Development of Two-Channel Pulse Processor and 3-D VisualizationBased on the DPP1 architecture, a 2-channel, DSP-enhanced, digital pulse processor

(DPP2) will be designed and constructed (Fig. 13). In DPP2, analog signal pulses from twoPMT’s of the saddle-geometry detector (XEPHWICH-S) will be individually digitized andcaptured. Digitized pulses then will be preprocessed in the FPGA module. A digital signalprocessor (DSP) then is responsible for management of light/electronic-coincidence andforming/updating the 3-D histogram.


26

A. Design and construct DPP2: Real-time digital pulse shape discrimination, processingand analysis for the saddle XEPHWICH detector will be performed in a 2-channel Digital PulseProcessor unit (DPP2). A one-channel version of the pulse processor unit (DPP1), developed forthe GEN3 detector, and used for proof-of-concept with the planar XEPHWICH, has beendesigned and constructed in our laboratory. In this task, DPP2 will be designed and constructedwith a high speed USB2 interface dedicated to analyze the two signal outputs of the saddleXEPHWICH. In the new design, noise reduction, pulse detection, time stamping, pile-uprejection, and the resolving of light components will be implemented in the field programmablegate array (FPGA) for each of the two lines. A digital signal processor (DSP) then is intendedfor the final signal processing, by: 1) reading digitized pulses from each pre-processed line(FPGA outputs); 2) determining electronic-based coincidence (by comparing the time stamp oneach digitized pulse); 3) constructing and updating the 3-D histogram; 4) organizing digitizedpulses to be stored in the PC memory; and 5) performing a specialized auto-calibration routine.The design will include a battery-powered DPP2 with a fail-safe mechanism by which the DSPcan temporarily continue essential PC tasks in the event of a power failure. On regaining power,the PC can be updated by the battery-powered DPP2.

B. Program DSP and FPGA firmware for DPP2: In this step, the programmingrequired for the DSP will be developed using C++. The FPGA will be configured by the high-level language VHDL. VHDL (Very-high-speed integrated circuit Hardware DescriptiveLanguage) is a firmware code language for programming integrated circuits, the FPGA in thiscase. Coding also will be included to transmit state-of-health data for the XEPHWICH andDPP2 units.

C. Write dedicated data acquisition software for DPP2: To allow the DPP2 tocommunicate with the host PC, customized data acquisition software for DPP2 will be developedin C++. Constructing the required USB2 driver and its on-board processor (8051) firmware aretwo important goals to be completed in this task.

D. Deliver preliminary DPP2 (w/o data transfer capability): At this point, the wholepreliminary DPP2 system, hardware and software, is ready to be utilized in processing signalpulses from the saddle XEPHWICH. Experimental measurements in Phase 3, task (f), start withcompletion of the DPP2.

E. Generate algorithm to construct real-time 3-D histogram: The signatures for 131mXeand 133mXe are based on the coincidence between their conversion electrons and 30 keV xenon X-rays. The unique signatures for 133Xe and 135Xe are also based on the coincidence between theirbeta/gamma or (beta+CE)/X-ray observations. In our XEPHWICH detector, depending onwhich side of the detector each radiation type is detected, the radiation coincidence is determinedby two different mechanism (Fig. 15), electronic-based or light–based coincidence. When thecoincidence electron (beta or CE) and photon (gamma or X-ray) are absorbed in different sidesof the detector (Figs. 15(a) and 15(b)), two coincident single-component pulses can be detectedfrom the two PMT’s. This is called electronic-based coincidence and can be identified from theirtime stamps by the digital signal processor. If the electron and photon are absorbed in the sameside of the detector (Figs. 15(c) and 15(d)), a multi-component pulse will be detected from thecorresponding PMT. This is called light-based coincidence. In this case, however, the multi-component signal pulse will carry the required information to determine if a coincidence eventhas occurred. Any valid pulse then increments the count of an energy bin in the 3-D histogram(each of the three axes corresponding to the energy deposition in a particular scintillation layer).Note that from either electronic- or light-based coincidence only one energy bin is incremented.


27

The location of the bin to be incremented is determined by the digital signal processor (DSP). Inthis task, the analysis algorithms for both the DSP and the data acquisition software will bedeveloped.

F. Match CTBT data-transfer protocol to DPP2: Data collected by the phoswich andprocessed by DPP2, as well as state-of-health (SOH) information, must be transferredperiodically to the International Data Center (IDC). This task will focus on matching the data-transfer formats with protocols established by the IDC.

G. Deliver final product (DPP2): The complete digital pulse processor system,including the DPP2 board and its related software, firmware and drivers will be delivered for fullintegration into the radioxenon detection system.

H. Support product integration and final testing: The final task of Phases 1 and 2 is toprovide support for the final task of Phase 3 in which the entire product (XEPHWICH and digitalsignal processing) will be integrated and readied for implementation into the existing ARSAstructure.

Fig. 15. Electronic- and light-based coincidence in the saddle XEPHWICH detector.

Phase 3 - Development and Characterization of Saddle-XEPHWICH systemPhase 3 is intended to develop a saddle-geometry XEPHWICH detector. Technical

aspects of this phase will be supported by the work performed in the other two phases. Withinthe first two years of the proposed research, while Phase 1 provides the required technicalresponse of the XEPHWICH in the planar geometry using a single-channel digital pulseprocessor (DPP1) developed in our laboratory, Phase 2 will focus on development and

Gamma/X-ray Beta/CE

PMT

PMT

PMT

PMT

PMT

PMT

PMT

PMT

(a)

(b)

(c)

(d)

Gamma/BetaElectronic-based Coincidence

Gamma/BetaLight-based Coincidence

Beta/CE

Gamma/X-ray

Gamma/X-ray Gamma/X-ray

Beta/CE

Beta/CE


28

construction of a 2-channel digital pulse processor (DPP2) unit. This unit and its relatedsoftware, firmware and drivers will be ready to use in task (f) of this phase, when the preliminarytests of the saddle-type XEPHWICH are finished and ready for experimental measurements.Some of tasks of Phase 1 and 3 are similar in name, but will be performed on two differentgeometric designs. To be useful, the tasks in Phase 1 are scheduled (see Table 4) to becompleted before the start of similar tasks in Phase 3.

A. Comprehensive study of Radon signature: Seismic activity following anunderground nuclear explosion provides the mechanism for the increased release of radon gas tothe atmosphere. This task, therefore, will explore the possibility of using the radon signature inair and to correlate the radon/xenon response for positive identification of an undergroundnuclear explosion.

B. MCNP modeling/optimization of saddle geometry: Monte Carlo N-Particle (MCNP)software will be used to model and optimize the XEPHWICH detector in a saddle-type geometry(Fig. 12) for a nearly 4π-detection of four radioxenon isotopes (131mXe, 133mXe, 133Xe and 135Xe),considering all related gamma-rays, X-rays, conversion electrons and beta particles.

C. Saddle XEPHWICH prototype design and construction: The saddle XEPHWICHprototype (Fig. 12) consists of a single cylindrical gas cell. The gas cell is surrounded by twocylindrical scintillation layers. These layers are intended for beta detection. The cylindricallayers are optically divided into two halves, each surrounded by a planar scintillator for gammadetection. To collect the scintillation light produced by all scintillators, in a nearly 4π-geometry,each planar scintillator is viewed by one sideways looking PMT. Unlike the ARSA detector, thebeta cell in the saddle XEPHWICH design is not optically isolated from the planar scintillator.For a complete optical isolation between two sides, the internal wall of the gas cell is covered byan aluminized Mylar layer. Lessons learned from Phases 1(a) and (b) in modeling and designingthe planar XEPHWICH detector will be applied here to design and construct the saddle-geometry XEPHWICH. This prototype will provide the framework to be extended to the currentfour-cell ARSA design, which provides continuous sampling capability.

D. Consider issues of planar knowledge with saddle geometry design: Potential issuesidentified in Phase 1 for the planar XEPHWICH design will be considered in this task to enhancethe performance of the saddle XEPHWICH. From the pulse processing aspect, changes might bewarranted in the pre-processing stages of the FPGA to promote the efficacy of digital signalprocessing in DPP2.

E. Characterization of saddle XEPHWICH response: With some considerabledifferences, similar characterization routines performed for the planar XEPHWICH in Phase 1,task(c), will be repeated in this step for the saddle-type detector. Two noteworthy differences arethat all calibrations and evaluation routines will be performed individually for each side of thesaddle XEPHWICH and, in addition to the light-based coincidence (Fig. 15), the electronic-based coincidence will be considered and characterized. To consider the latter, the DSP in theDPP2 unit will be programmed, first to detect these events and then, to extract the appropriateenergy components from each coincident pulse.

F. Experimental measurements with DPP2: To evaluate the radioxenon detectionsystem, various gaseous sources (including sources of purchased and/or activated xenon) will beintroduced into the sample cell of the saddle XEPHWICH detector. The system response in theenergy regions covered by the four specific CTBT radioxenon isotopes (131mXe, 133mXe, 133Xe and135Xe), and the daughter nuclides of 222Rn, will be determined. To study system capabilities, forexample, in discrimination of 131mXe and 133mXe from the 30 keV photon events due to 133Xedecay, different concentration ratios of the various calibration nuclides will be mixed andstudied.


29

G. Development/application of Rn discrimination/correlation routine: Evenelectron/photon coincidence spectroscopy can not reject the background events from radondaughters which can also appeared in the sample cell. The most dominant backgroundradioisotopes due to radon decay are 214Bi and 214Pb. The contribution of these radioisotopes inthe radioxenon measurements can be determined by detecting the 315.9 keV gamma rayfollowing beta emission from 214Pb (Bower et al. 1999). This task will develop the appropriatealgorithm to locate the background coincidence events due to radon progeny in the 3-Dhistogram and make the required corrections. Regarding the results from the comprehensivestudy performed in task (a) of this phase, the radon concentration history in air, rather than beingdiscarded as background, might be recorded and used as an extra signature for undergroundnuclear explosion monitoring. For this reason, in addition to making the radon backgroundcorrections for radioxenon measurements, the discrimination routine will provide the requireddata for radon concentration measurement.

H. Product integration and final testing: In the final task of Phase 3, product integrationoccurs such that the saddle XEPHWICH, its digital pulse processing circuits, the supportingneural network, visualization routines, etc. are brought together for final unified testing foreventual implementation into the existing ARSA structure.

4. PROPOSED SCHEDULE

A detailed completion schedule is provided in Table 4. In the first project year, the teamwill begin with development and analysis of a prototype XEPHWICH detector and theincorporation of digital processing methods for true separation of beta and gamma components.Phases 1, 2, and 3 will run concurrently, with each phase completed (generally) by each of thethree faculty and their doctoral students. The students, their advisors, and the research associatewill work together as a team, with regularly scheduled meetings. In the final project year,unification of the work will demand intensive collaboration within the research group. Thestudents will receive invaluable teaming experience.

Deliverables include the prototypic planar XEPHWICH and its digital processing system,a two-channel digital pulse processing unit with associated firmware and CTBT data-transferprotocols, and a prototypic saddle-geometry XEPHWICH (and digital system) for retrofitting theARSA structure. Additionally, as his track record shows, the PI will require that all threedoctoral students publish their work in peer-reviewed journals at a minimum of one paper peryear.

5. KEY PERSONNEL

The PI, David M. Hamby, graduated in 1989 from the University of North Carolina atChapel Hill with an MS and PhD in health physics. After six years as a Senior ResearchScientist at the Savannah River Laboratory in Aiken, SC, he shifted to academia and went to theUniversity of Michigan as an Assistant Professor in their radiological health program. In 1999,Dr. Hamby was recruited by Oregon State University as a tenured Associate Professor, and waspromoted to Professor in 2004. He has authored over 90 peer-reviewed publications andpresentations, and more than 30 government reports.

The citations listed below represent a subset of Dr. Hamby’s work and show that hisinterests in radiation detection span the previous decade. Originally, the PI’s work inenvironmental health physics led to investigations with neural networks, microdosimetry,


30

environmental assessment, and radiation detection instrumentation for in situ soil analyses.Additionally, his field work in Kyrgyzstan was significant in directing future detectordevelopment. In late 1997, Dr. Hamby and his students began looking specifically at theprospects of beta spectroscopy as it relates to the dosimetry of mixed beta/gamma sources. Thefirst prototype phoswich detector was developed by his research group in 1998 and utilized forinitial studies into the efficacy of deconvolution and spectral stripping techniques for beta emitteridentification. This effort was lead by two of Dr. Hamby’s doctoral students (Bush-Goddard andMiklos) and resulted in three (3) publications. Dr. Bush-Goddard is now a program managerwith the NRC and Dr. Miklos is the Occupational Safety and Environmental Health coordinatorat the University of Michigan.

Table 4. Schedule of work for development of the XEPHWICH radioxenon identification system.

Task Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4

1. XEPHWICH Proof of Concept a.) MCNP modeling/optimization of radioxenon phoswich b.) XEPHWICH prototype design and construction c.) characterization of XEPHWICH response with existing DPP1 d.) experimental measurements with radioxenon source e.) neural network design/development for 3-D histogram interpretation f.) radioxenon identification routine development/application g.) support product integration and final testing

2. Development of 2-Channel Pulse Processor and 3-D Visualization a.) design and construct DPP2 b.) program DSP and FPGA firmware for DPP2 c.) write dedicated data acquisition software for DPP2 d.) deliver preliminary DPP2 (w/o data transfer capability) e.) generate algorithm to construct real-time 3-D histogram f.) match CTBT data-transfer protocol to DPP2 g.) deliver final product (DPP2) h.) support product integration and final testing

3. Development and Characterization of Saddle-XEPHWICH System a.) comprehensive study of radon signature b.) MCNP modeling/optimization of saddle geometry c.) XEPHWICH-S prototype design and construction d.) consider issues of planar knowledge with saddle geometry design e.) characterization of saddle XEPHWICH response f.) experimental measurements with DPP2 g.) development/application of Rn discrimination/correlation routine h.) product integration and final testing

Year 1 Year 2 Year 3

In early 2002, the PI was awarded a 3-yr Nuclear Engineering Education Research(NEER/DOE) grant and another of his doctoral students (Kriss) extended the original work into asecond-generation prototype of plastic scintillator and large-area photodiode components. In late2003, another doctoral student (Tavakoli-Farsoni) joined the group to work on improvements tothe beta spectroscopy system. Nine (9) publications to date will come from the work by Krissand the preliminary work of Tavakoli-Farsoni. Dr. Kriss is now employed at PNNL as aprincipal research scientist in the Radiological Sciences and Engineering group. Mr. Tavakoli-Farsoni has a BS in applied physics and an MS in nuclear engineering. He has been involved inradiation instrumentation and analysis for more than 8 years and has experience with researchreactors, radioanalytical techniques (INAA, ENAA, and PGNAA), various radiation detectioninstruments, and trace element analysis in environmental media. Dr. Hamby was recentlyawarded a second 3-yr NEER grant, on which Mr. Tavakoli-Farsoni is funded. The PI isconfident that, with NNSA funding, he and his researchers can extend their instrumentation


31

development work significantly, pushing the state-of-the-art in definitive radioxenon detection,digital pulse analysis, and simultaneous beta/gamma spectroscopy.

PRINCIPAL INVESTIGATOR – DAVID M. HAMBY

EducationPh.D. in Health Physics, University of North Carolina, Chapel Hill, NC, 1989

Dissertation: "Measurement of Radial Ionization Probability in Microscopic Volumes for2.90 and 3.79 MeV Alpha Particles"; D.J. Crawford-Brown, Advisor

M.S. in Health Physics, University of North Carolina, Chapel Hill, NC, 1986Thesis: "A Microdosimetric System for use in the Measurement of Specific EnergyDistributions for 15 MeV Electrons in Water"; D.J. Crawford-Brown, Advisor

B.S. in Physics (with Honors), Mercer University, Macon, GA, 1984

Professional Experience06/04 - present Professor in Radiation Health Physics, Department of Nuclear Engineering and

Radiation Health Physics, College of Engineering, Oregon State University,Corvallis.

5/95 - present Faculty Appointee, Environmental Assessment Division, Argonne NationalLaboratory, Argonne, Illinois.

12/99 - 05/04 Associate Professor in Radiation Health Physics, Department of NuclearEngineering and Radiation Health Physics, College of Engineering, OregonState University, Corvallis.

12/99 - 08/03 Adjunct Associate Professor of Environmental Health, Department ofEnvironmental Health Sciences, School of Public Health, University ofMichigan, Ann Arbor.

7/94 - 12/99 Assistant Professor of Radiological Health, Department of Environmental andIndustrial Health, School of Public Health, University of Michigan, Ann Arbor.

1/89 - 7/94 Principal Research Scientist, Waste Management and EnvironmentalTechnology Department, Savannah River Technology Center, WestinghouseSavannah River Company, Aiken, SC.

Selected PublicationsTavakoli-Farsoni, A.; Hamby, D.M. MCNP analysis of a multilayer phoswich detector

for beta particle dosimetry and spectroscopy. Nuclear Instruments and Methods inPhysics Research - Section A. submitted July 2005.

Harvey, R.P.; Hamby, D.M.; Palmer, T.S. Uncertainty of the Thyroid Dose ConversionFactor for Inhalation Intakes of 131I and Its Parametric Uncertainty. RadiationProtection Dosimetry. submitted June 2005.

Kriss, A.A. A beta dosimeter and spectrometer utilizing plastic scintillators and a large-area avalanche photodiode”, PhD Dissertation, Oregon State University. June 2004.

Tavakoli-Farsoni, A.; Hamby, D.M.; Bush-Goddard, S.P. A performance study on atriple-layer phoswich detector for beta spectroscopy. Proceedings of the Forty-ninthAnnual Meeting of the Health Physics Society. Washington, DC. Health Physics.86(6): S144; 2004.

Kriss, A.; Hamby, D.M. Beta spectroscopy and dosimetry with a large area avalanchephotodiode module and plastic scintillators. Proceedings of the Forty-ninth Annual


32

Meeting of the Health Physics Society. Washington, DC. Health Physics. 86(6):S193; 2004.

Kriss, A.A.; Hamby, D.M. Beta spectroscopy with a large-area avalanche photodiodemodule and a plastic scintillator. Nuclear Instruments and Methods in PhysicsResearch - Section A. 525(3): 553-559; June 2004.

Kriss, A.; Hamby, D.M. Scintillation beta dosimetry and spectroscopy utilizing a largearea avalanche photodiode. Proceedings of the Forty-eighth Annual Meeting of theHealth Physics Society. San Diego, CA. Health Physics. 84(6): S168; 2003.

Kriss, A.; Hamby, D.M. A phoswich detector for beta spectroscopy. Cascades Chapter ofthe Health Physics Society. Corvallis, OR; March 14, 2003.

Miklos, J.A. Unique specification of beta-particle sources, PhD Dissertation, Universityof Michigan, 2002.

Bush-Goddard, S.P. Beta spectroscopy using deconvolution and spectral strippingtechniques with a triple layer phoswich detector, PhD Dissertation, University ofMichigan, 2000.

Bush, S.P.; Hamby, D.M. Techniques for beta spectroscopy using Monte Carlo methodsand spectral deconvolution. Health Physics Society Annual Meeting. Philadelphia,PA; June 28, 1999. Health Physics. 76:S109; 1999.

Hamby, D.M.; Tynybekov, A.K. A screening assessment of external radiation levels onthe shore of Lake Issyk-Kyol in the Kyrghyz Republic. Health Physics. 77(4):427-430; 1999.

Tynybekov, A.K.; Hamby, D.M. Radiological description of the southern coast of LakeIssyk-Kul. In: A Collection of Works. Ministry of Labor and Social Protection.Bishkek, Kyrgyz Republic. Vol. 2: pg. 9-17; April 1999. (in Russian)

Hamby, D.M.; Zometsky, J.R. A method for in situ depth profiles of alpha and betacontaminants in soil using scintillators and fiber optic light guides. RadiationProtection Management. 15(5):26-32; 1998.

Bush, S.P.; Hamby, D.M. Initial investigations into developing a wall-less proportionalcounter for use in radiologically contaminated soils. Radiation ProtectionManagement. 15(2):43-47; 1998.

Famiano, M.A.; Hamby, D.M. Demonstration of a time-integrating microdosimeter.Nuclear Instruments and Methods in Physics Research - Section A. 389(3):479-490;1997.

Bush, S.P.; Hamby, D.M. Development of a screened cathode gas flow proportionalcounter for in-situ field determination of alpha contamination in soil. Health PhysicsSociety Annual Meeting. San Antonio, TX; July 1, 1997. Health Physics. 72:S53;1997.

Famiano, M.A.; Hamby, D.M. Time-specific measurements of energy deposition fromradiation fields in simulated sub-micron tissue volumes. Health Physics. 70:S16;1996.

Bush, S.P.; Hamby, D.M. In situ field determination of radioactive contamination in soilusing a wall-less gas flow proportional counter. Health Physics Society AnnualMeeting. Seattle, WA; July 23, 1996. Health Physics. 70:S39; 1996.

Hamby, D.M. The use of neural networks in environmental decision making. MichiganSection of the American Nuclear Society. Ann Arbor, MI; September 28, 1995.

Hamby, D.M.; Famiano, M.A. Neural network applications in environmental decisionmaking. Great Lakes Chapter of the Health Physics Society Spring Symposium. AnnArbor, MI; March 29, 1995.


33

Hamby, D.M. A comparison of sensitivity analysis techniques. Health Physics . 68:195-204; 1995.

Hamby, D.M. A review of techniques for parameter sensitivity analysis of environmentalmodels. Environmental Monitoring and Assessment . 32:135-154; 1994.

Hamby, D.M. A probabilistic estimation of atmospheric tritium dose. Health Physics .65:33-40; 1993.

Hamby, D.M. A methodology for estimating the radiological consequence of an acuteaqueous release. Health Physics . 62:567-570; 1992.

Bauer, L.R.; Hamby, D.M. Relative sensitivities of existing and novel model parametersin atmospheric tritium dose estimates. Radiation Protection Dosimetry . 37:253-260;1991.

Selected Scientific Committees, Affiliations, Awards2000 - present Reviewer, U.S. Civilian Research and Development Foundation, Proposals for

awards to International Science and Technology Centers.2000 - present Reviewer, American Institute of Biological Sciences, U.S. Army Medical

Research and Material Command - and - Military Operational Medicine to theOffice of Naval Research

1999 - present Editorial Advisory Board for Environmental Monitoring and Assessment1996 - present Member, National Council on Radiation Protection and Measurement (NCRP)

scientific subcommittee on “Cesium in the Environment” (SC#64-23).1996 - present Associate Editor for Health Physics (journal of the Health Physics Society)1985 - present Member, National Health Physics Society (1985-present); North Carolina

Chapter HPS (1985-1989); Savannah River Chapter HPS (1989-1994); GreatLakes Chapter HPS (1994-2000); Cascade Chapter HPS (2000-present).

2000 - 2002 Member, ATSDR Expert Panel for Environmental Tritium. Atlanta, GA.1998 Expert in Radiological Health for the International Atomic Energy Agency

(IAEA). Lithuania.1997 Fulbright Scholar Award, Environmental Health. Bishkek, Kyrgyzstan.1996 Consultant, National Academy of Science, Environmental Management

Technology subcommittee on the DOE/EM decision-making process.1994 Panelist, Centers for Disease Control and Prevention (CDC). Environmental

Radiological Dose Reconstruction in the US and the former Soviet Union,Atlanta, GA.

1992 Invited Panelist, EPA/ORP Workshop on Mathematical Modeling -Meteorological Models, Bethesda, MD.

Didactic Courses Taught2005 - present NE/RHP 536: Advanced Radiation Detection and Measurement (Oregon State)2003 - present NE/RHP 236: Radiation Detection and Instrumentation (Oregon State)2001 - present NE/RHP 490/590: Radiation Dosimetry (Oregon State)2001 - present NE/RHP 235: Nuclear and Radiation Physics II (Oregon State)2001 - 2003 NE/RHP 585: Environmental Aspects of Nuclear Systems (Oregon State)2000 - 2003 RHP 483/583: Radiation Biology (Oregon State)1999 - 2002 RHP 592: Radiological Risk Evaluation (Oregon State)1999 EHS 581: Principles of Radiological Health (Univ of Michigan)1997 - 1999 EHS 693: Health Physics Instrumentation: Theory and Practice (Univ of

Michigan)


34

1994 - 1999 EHS 679: Radiological Risk Evaluation (Univ of Michigan)1994 - 1999 EHS 672: Environmental Radiological Assessment (Univ of Michigan)


35

CO-INVESTIGATOR – TODD S. PALMER

EducationPh.D., Nuclear Engineering, University of Michigan, 1993M.S., Nuclear Engineering, University of Michigan, 1988B.S., Nuclear Engineering, Oregon State University, 1987

Professional ExperienceAssociate Professor, Oregon State University, 2001 – presentAssistant Professor, Oregon State University, 1995-2001Physicist, A-Division, Lawrence Livermore National Lab., 1991-1994Research Assistant, University of Michigan, 1987-1991

Selected PublicationsI.M. Davis, T.S. Palmer and E.W. Larsen, “A Comparison of Binary Stochastic Media

Transport Models in ‘Solid-Void’ Mixtures”, Proceedings of the American NuclearSociety Topical Meeting on Reactor Physics (PHYSOR 2004), April, 2004, Chicago,IL.

R. Nes and T.S. Palmer, “An Advanced Nodal Discretization for the Quasi-DiffusionLow-Order Equations”, Proceedings of the American Nuclear Society TopicalMeeting on Reactor Physics (PHYSOR 2002), October, 2002, Seoul, South Korea.

B. S. Ching and T.S. Palmer, “An Acceleration Scheme for Binary Stochastic MixtureDeterministic Transport in Slab Geometry”, Proceedings of the American NuclearSociety International Meeting on Mathematical Methods for Nuclear Applications,September, 2001, Salt Lake City, UT.

C. M. Marianno, K. A. Higley, and T. S. Palmer, “Theoretical Efficiencies for a FIDLERScanning Hot Particle Contamination”, Rad. Protect. Mgmt. 17, 3, pp. 31-34, 2000.

T. S. Palmer, “Discretizing the Diffusion Equation on Unstructured Polygonal Meshes inTwo Dimensions,” Ann. Nucl. Energy, 28, pp. 1851-1880, 2001.

C. M. Marianno, K. A. Higley, and T. S. Palmer, “A Comparison Between Default EGS4and EGS4 with Bound Compton Cross Sections when Scattering Occurs in Bone andFat,” Health Physics , 78, vol. 6, pp. 716-720, June 2000.

B. Laubsch and T. S. Palmer, “Chapman-Enskog Analysis of Discretized TransportEquations”, Proceedings of the International Topical Meeting on Advances inReactor Physics and Mathematics and Computation into the Next Millennium,Session III.E, Paper No. 4, pgs. 1-15, Pittsburgh, PA, May 7-11, 2000.

T.S. Palmer, "A Point-Centered Diffusion Discretization for Unstructured Meshes in 3-D," Proceedings of the American Nuclear Society's International Conference onMathematics and Computations, Reactor Physics, and Environmental Analyses,Portland, OR, Vol. 1, pgs. 897-905, April 30-May 4, 1995.

T.S. Palmer and M.L. Adams, "Curvilinear Geometry Transport Discretizations in ThickDiffusive Regions," Proceedings of the Joint International Conference onMathematical Methods and Supercomputing in Nuclear Applications, Vol. 1, p. 3-14,Karlsruhe, Germany, 1993.

T.S. Palmer and M.L. Adams, "Analysis of Spherical Geometry Finite Element TransportSolutions in the Thick Diffusion Limit," Proceedings of the International TopicalMeeting of the American Nuclear Society--Advances in Mathematics, Computations,and Reactor Physics," Vol. 5, p. 21.1 4-1 to 21.1 4-11, Pittsburgh, PA, 1991.


36

CO-INVESTIGATOR - LUCA LUCCHESE

EducationPh.D., Electrical Engineering, University of Padua, Italy, July 1997.M.S., Electrical Engineering, University of Padua, Italy, March 1993.

Professional ExperienceAssistant Professor, Electrical Engineering, Oregon State University, 3/02 - presentVisiting Assistant Professor, Electrical and Computer Engineering, UCSB, 7/00 – 2/02Associate Researcher, Electrical and Computer Engineering, UCSB, 4/97 – 6/00

Selected PublicationsL. Lucchese, “Frequency Domain Classification of Cyclic and Dihedral Symmetries of Finite 2-

D Patterns,” Pattern Recognition, Vol. 37, No. 12, Dec. 2004, pp. 2263-2280.L. Lucchese and S.K. Mitra, “A New Class of Nonlinear Filters for Color Image Processing.

Theory and Applications,” IEEE Transactions on Image Processing, Vol. 13, No. 4, Apr.2004, pp. 534-548.

L. Lucchese, “Geometric Calibration of Digital Cameras. Part I: Camera Model, SubpixelFeature Extraction, and Algorithm Initialization,” Proc. of the Third IASTED Int’lConference on Visualization, Imaging, and Image Processing, Benalmadena, Spain, Sept. 8-10, 2003, Vol III, pp. 1061-1065.

L. Lucchese, “Geometric Calibration of Digital Cameras. Part II: Internal and External Geometryfrom Multi-view Alignment,” Proc. of the Third IASTED Int’l Conference on Visualization,Imaging, and Image Processing, Benalmadena, Spain, Sept. 8-10, 2003, Vol III, pp. 1066-1070.

L. Lucchese and S.K. Mitra, “A New Method for Denoising Color Images,” Proc. of 2002 Int’lConference on Image Processing, Rochester, NY, Sept. 2002, Vol. II, pp. 373-376.

L. Lucchese, “A Frequency Domain Algorithm for Detection and Classification of Cyclic andDihedral Symmetries in Two-Dimensional Patterns,” Proc. of 2002 Int’l Conference onImage Processing, Rochester, NY, Sept. 2002, Vol. II, pp. 793-796.

L. Lucchese, “A Hybrid Frequency-Space Domain Algorithm for Estimating ProjectiveTransformations of Color Images,” Proc. of 2001 Int’l Conference on Image Processing(ICIP 2001), Thessaloniki, Greece, Sept. 2001, Vol. II, pp. 913-916.

M. Andreetto, R. Bernardini, G.M. Cortelazzo, and L. Lucchese, “Towards Automatic Modelingof 3-D Cultural Heritage,” Proc of. 2001 Int’l Conference on Image Processing (ICIP 2001),Thessaloniki, Greece, Sept. 2001, Vol. I, pp. 574-577.

L. Lucchese, “A Frequency Domain Technique Based on Energy Radial Projections for RobustEstimation of Global 2D Affine Transformations,” Computer Vision and ImageUnderstanding, Vol. 81, pp. 72-116, Feb. 2001.

L. Lucchese and G.M. Cortelazzo, “A Noise-robust Frequency Domain Technique forEstimating Planar Roto-Translations,” IEEE Transactions on Signal Processing, Vol. 48, No.6, pp. 1769-1786, June 2000.


1

VOLUME 3COST PROPOSAL

I. Salaries and WagesFaculty. The Principal Investigator (Dr. Hamby) will commit 30% of his time to the

proposed work, directing the research of two doctoral students, participating in the research ofone other doctoral student, and maintaining responsibility for all aspects of the project. Drs.Palmer and Lucchese (Co-Investigators) will each commit 15% of their time to the project, withDr. Palmer supporting all three doctoral students with computation methods and Dr. Lucchesewith direct supervision of one doctoral student. Dr. Palmer is expert in radiation transportmodeling and will provide guidance with the use of MCNP. Dr. Lucchese is expert in electricalengineering and will provide guidance with design/development of the DSP and FPGA modules.Partial salary support is requested for faculty involvement, commensurate with projectresponsibilities and time commitments. Faculty will normally receive nine months support viathe University. Requested support provides enhanced opportunities for faculty interaction in thesummer and throughout the year on this project. An annual increase of 4% has been applied toFaculty salaries for the 2nd and 3rd project years.

Professional Faculty Research Associate. Funding for 1.0 FTE (12-month appointment)post-doctoral Faculty Research Associate (FRA) is requested for each project year. The FRAwill receive a salary of $4,600 per month, increased by 4% in each of the subsequent years. TheFRA will be expected to manage the day-to-day research functions. Major assignments willinclude a lead role in experimental design, advancement of calculational methods, generation ofrequired reports, development of new areas of research related to overall project objectives, andco-authoring peer-reviewed publications. Additionally, the FRA will conduct regular researchmeetings and provide first-line supervision of project-funded graduate researchers.

Graduate Research Assistants. Funding for three doctoral student graduate researchassistants (GRA) is requested for each project year. The GRAs will be paid a stipend ofapproximately $1,550 per month (increased by 4% in each of the subsequent years), plus anadditional $500 per year to compensate for health insurance (increased by 11% per year). Allresearch assistants will be expected to work a total of 16 hours per week (0.40 FTE), year-round,on this project. Major assignments to tasks of the proposed project will include establishment ofcalibration techniques, development of laboratory systems, characterization of detectorprototypes, development of enhanced spectral analysis methods, and incorporation of neuralnetworks. The GRA doctoral students will be expected to publish multiple manuscripts duringtheir tenure on this project.

II. Fringe BenefitsFringe benefits, using standard Oregon State University rates, are listed on the budget

detail and calculated at a rate of 44% for faculty, 46% for the FRA, and $383 per term perstudent for GRAs (increased by 11% per year).

III. SuppliesSupplies are defined as any item or system less than $5,000.00. Supply funds are

requested in the first year to construct a planar model of the XEPHWICH multi-layer phoswichdetector optimized for radioxenon ($3,500), and general supply funds for radioactive calibration


2

and check sources ($3,500 each year), three computer systems and software ($9,000 first year),and miscellaneous laboratory supplies ($2,500 each year). Two high-speed digital oscilloscopes(Tektronics) are also necessary for signal analysis, detector characterization, and trouble-shooting ($4,000 first year). These supplies are extremely important to the furtherance of ourwork. More detail on the usage of these supplies is provided in the Technical Approach.

IV. Domestic TravelThe proposed travel budget provides airfare (estimated at $750 per round-trip), per diem

for economy subsistence and lodging ($175 per day), auto rental ($75 per day), and conferencefee when appropriate ($300). Trip durations of 4 days are assumed. Travel funds are requestedfor six trips per year (w/ conference fee) for the PI, Co-Investigators, FRA, or students to presentupdates/findings at profession meetings (e.g. American Nuclear Society, Health PhysicsSociety). Requested funds also include resources for six additional trips annually for the PI tomeet twice with the Product Integrator and for the PI and FRA to attend the two NNSA researchmeetings.

V. EquipmentWe are requesting equipment funds ($9,500) to construct the saddle XEPHWICH in the

third project year. This purchase will be greater than five-thousand dollars, thus considered“equipment”. The saddle XEPHWICH is key to demonstrating the utility of our design and theefficacy of placing the XEPHWICH in the current ARSA infrastructure.

VI. Other Direct CostsTuition and fees for each of the three Graduate Research Assistants will cost $2,713 per

quarter in the 2005-2006 academic year. The GRAs will be on 12-month appointments andenrolled for 3 quarters, therefore, the total cost of tuition for one student is $8,139 in the firstyear. Tuition is projected to increase by 8% in each of project years 2 and 3.

VII. MiscellaneousFunds are requested ($2,500 each year) for purchasing miscellaneous office and computer

materials, postal fees, books, publication charges, photocopying, communication expenses, etc.

VIII. Indirect rateAn indirect rate of 41.5% is being applied to all costs covered under this work, except

graduate student tuition and equipment. This rate is the negotiated federal modified indirectreturn rate.

No subcontracts are required.

BUDGET FORMS DOE F 4620.1(attached in file: FA-FORMS_Hamby)


3

BUDGET SUMMARY BY PHASE (All 3 Project Years Combined)

Phase 1 Phase 2 Phase 3

Senior Researcher 64,534 32,267 64,534Senior Researcher 35,987 0 35,987Senior Researcher 0 62,931 0Faculty Res. Assoc. 83,858 83,858 83,858Research Assistant 64,627 64,627 64,627Travel 22,800 22,800 22,800Planar XEPHWICH 3,500 0 0Saddle XEPHWICH 0 0 9,500Computer/software 3,000 3,000 3,000Lab supplies 2,500 2,500 2,500Tuition/Fees 26,422 26,422 26,422Xerox, publishing, etc. 2,500 2,500 2,500Digital Oscilloscope 0 2,000 2,000Calibration sources 5,250 0 5,250---------------------------------------------------------------------------------------------------SUBTOTAL 314,978 302,905 322,978INDIRECT 119,750 114,740 119,128===========================================================TOTAL (by Phase) 434,728 417,645 442,106


1

VOLUME 4BUSINESS MANAGEMENT PROPOSAL

1. Application for Federal Assistance (SF-424)

attached (in file: FA-FORMS_Hamby)

2. Negotiated Indirect Rate Information

Department of Health and Human ServicesWallace Chan(415) 437-7820

3. Assurances of Compliance (DOE F 1600.5)

attached (in file: FA-FORMS_Hamby)

4. FA-CERTS

attached (in file: FA-CERTS_Hamby)

5. Financial Assistance List

Current

a. U.S. Department of Energy. Nuclear Engineering Education Research.“Enhancing State-of-the-Art Beta Detection and Dosimetry”. (PI: David M.Hamby) Solicitation # DE-PS07-04ID14632. 7/05 – 6/08. $358,170. The researchis focused on developing a detection system for simultaneous beta and gammaspectroscopy for advancement of beta dosimetry methods.

Past

b. U.S. Department of Energy. Nuclear Engineering Education Research. “AdvancedBeta Dosimetry Techniques”. (PI: David M. Hamby) Solicitation # DE-PS07-02ID14200. 6/02 – 5/05. $337,557. The research investigated the use of large-area photodiodes for beta scintillation spectroscopy and beta dosimetry.

c. U.S. Department of Health and Human Services. Centers for Disease Control andPrevention. “Atmospheric I-131 Dose Estimates: Comparative Uncertainties”.(PI: David M. Hamby) Solicitation # 99020. 8/99 – 7/02. $372,482. The researchexamined a lung deposition model, an atmospheric transport model, anduncertainty related to estimates of downwind radiological concentrationsfollowing release from a nuclear facility.


2

d. U.S. Civilian Research and Development Foundation. National ScienceFoundation. “Radiological Characterization in the Vicinity of Lake Issyk-Kul,Republic of Kyrgyzstan”. (PI: David M. Hamby) 2/97 – 1/01. $50,000. Theresearch focused on the characterization of naturally occurring radionuclides inthe beach sands of a large mountain lake in eastern Kyrgyzstan.

e. NATO Assistant Secretary General for Scientific and Environmental Affairs.“Radiation Levels in the Vicinity of Lake Issyk-Kul, Kyrgyzstan”. (PI: David M.Hamby) 10/96 – 9/99. $18,600. This research complimented that above bymeasurements of ambient radiation levels above beach sands with high activities.

f. Department of Energy. Office of Environmental Management. “Study of FieldMethods and Worker Risks for Processing Alternatives to Support GuidingPrinciples for FUSRAP Waste Materials”. (co-PI with James E. Martin).Unsolicited. 4/96 – 3/98. $130,000. This research examined waste processingalternatives in the FUSRAP program that could result in reduced humanradiological dose.

6. Performance Information

Current

a. “Enhancing State-of-the-Art Beta Detection and Dosimetry”. Nancy Elizondo,USDOE Idaho Falls, 208-526-4169, [email protected], GrantNo. DE-FG07-05ID14704.

The intent of this research is to develop a state-of-the-art radiation detectionsystem for real-time identification and dosimetry of beta-emitting radionuclides ina mixed radiation field. A 3rd generation phoswich detector has been designedthat builds on experience and lessons-learned from the first two systems. Thisprototype will utilize high-speed digital signal processing capable of spectraldiscrimination, detailed pulse component analysis, neural network identification,memory-based quantification of mixed beta emitters, and computationallyenhanced beta dosimetry. Coincidence and anti-coincidence counting schemeswill be employed to digitally analyze scintillator light output from simultaneouselectron and photon interactions in the layered scintillation material. The betadosimetry system developed will have broad-ranging applications in radioactivewaste management, worker safety, systems reliability, dose assessment, and riskanalysis. The detector, in small dimensions, would also have a medicalapplication as an intraoperative probe where preferential detection of betaemissions over photon interactions is needed.

Past

b. “Advanced Beta Dosimetry Techniques”. Nancy Elizondo, USDOE Idaho Falls,208-526-4169, [email protected], Grant No. DE-FG07-02ID14331.


3

(7 Publications) This research resulted in the development of a state-of-the-artradiation detection system using scintillating plastics and large-area photodiodesfor quantification and dosimetry of beta-emitting radionuclides in a β/γ field.Identification of beta-emitting nuclides is facilitated by a de-convolution andoptimization technique developed specifically for the unfolding of beta spectrameasured from our scintillation detector. Generation of response functions ofvarious radiation types was accomplished such that the detector, coupled with anenhanced deconvolution methodology, is useful for complex beta dosimetry undermixed-field exposure conditions.

Tavakoli-Farsoni, A.; Hamby, D.M. MCNP analysis of a multilayer phoswich detector for betaparticle dosimetry and spectroscopy. Nuclear Instruments and Methods in Physics Research -Section A. submitted July 2005.

Kriss, A.A. “A Beta Dosimeter and Spectrometer Utilizing Plastic Scintillator Volumes and aLarge Area Avalanche Photodiode". Doctoral Dissertation. Oregon State University Collegeof Engineering. June 2004.

Kriss, A.A.; Hamby, D.M. Beta spectroscopy with a large-area avalanche photodiode module anda plastic scintillator. Nuclear Instruments and Methods in Physics Research - Section A.525(3): 553-559; June 2004.

Tavakoli-Farsoni, A.; Hamby, D.M.; Bush-Goddard, S. A performance study on a triple-layerphoswich detector for beta spectroscopy. Proceedings of the Forty-ninth Annual Meeting ofthe Health Physics Society. Washington, DC. Health Physics. 86(6): S144; 2004.

Kriss, A.; Hamby, D.M. Beta spectroscopy and dosimetry with a large area avalanche photodiodemodule and plastic scintillators. Proceedings of the Forty-ninth Annual Meeting of the HealthPhysics Society. Washington, DC. Health Physics. 86(6): S193; 2004.

Kriss, A.; Hamby, D.M. Scintillation beta dosimetry and spectroscopy utilizing a large areaavalanche photodiode. Proceedings of the Forty-eighth Annual Meeting of the Health PhysicsSociety. San Diego, CA. Health Physics. 84(6): S168; 2003.

Kriss, A.; Hamby, D.M. A Phoswich Detector for Beta Spectroscopy. Cascades Chapter of theHealth Physics Society. Corvallis, OR; March 14, 2003.

c. “Atmospheric I-131 Dose Estimates: Comparative Uncertainties”. CM Wood,Centers for Disease Control, Radiation Studies Branch, Atlanta, 404-498-1826,[email protected], Grant No. R32/CCR-018377.

(15 Publications) This work resulted in a number of products, including: thedevelopment of parameter sensitivity rankings and estimates of uncertainty in theingestion and inhalation dose models for chronic atmospheric releases of 131I atthe Savannah River Site (SRS); production of a reconstructed historical (1955-1961) meteorology dataset for the SRS; a determination of the sensitivity of theGaussian model to JFD inputs; the generation of distributions for parameters inthe DCF calculation; estimates of the uncertainty of the 131I age-specific ingestionand inhalation dose factors; the generation of age-dependent distributions ofconsumption/usage for average individuals and defined-maximum individuals;and the determination of parameters to which the unified transport/dosimetrymodel is most sensitive under conditions of data partitioning. The project


4

resulted in estimates of model-segment uncertainties (e.g., usage factors, dosefactors, environmental concentrations) as well as total uncertainties (e.g., aunified transport/dose model) in dose estimates for age-specific averageindividuals and defined-maximum individuals within an 80 km region around theSavannah River Site. Major tasks included estimating historical environmentalconcentrations, determining the uncertainty of the iodine age-specific ingestionand inhalation dose conversion factors, estimating the average and maximumintake of radioiodine, and conducting a sensitivity and uncertainty analysis of aunified environmental dosimetry model.

Harvey, R.P.; Hamby, D.M.; Palmer, T.S. Uncertainty of the Thyroid Dose Conversion Factor forInhalation Intakes of 131I and Its Parametric Uncertainty. Radiation Protection Dosimetry.submitted June 2005.

Harvey, R.P.; Hamby, D.M.; Palmer, T.S. A Modified ICRP 66 Iodine Gas Uptake Model and ItsParametric Uncertainty. Health Physics. 87(5): 490-506; 2004.

Harvey, R.P., "The ICRP 66 Lung Model and the Behavior of Gases Iodine", DoctoralDissertation. University of Michigan School of Public Health. April 2003.

Simpkins, A.A.; Hamby, D.M. Uncertainty in transport factors used to calculate historical dosefrom 131I releases at the SRS. Health Physics. 85(2); 194-203; 2003.

Harvey, R.P.; Hamby, D.M.; Benke, R.R. Age-specific uncertainty of the 131I dose conversionfactor. Health Physics. 84(3): 334-343; 2003.

Weber, A.H.; Buckley, R.L.; Parker, M.J.; Harvey, R.P.; Hamby, D.M. The creation of anhistorical meteorological database for environmental dose assessment. EnvironmentalMonitoring and Assessment. 83(3): 255-281. 2003.

Harvey, R.P.; Hamby, D.M. Age-specific uncertainty in particulate deposition for 1 µm AMADparticles using the ICRP 66 lung model. Health Physics. 82(6): 807-816; 2002.

Hamby, D.M. The Gaussian atmospheric transport model and its sensitivity to the joint frequencydistribution and parametric variability. Health Physics. 82(1): 64-73; 2002.

Harvey, R.; Hamby, D. Age-Specific uncertainty of the 131I ingestion dose conversion factor.Proceedings of the Forty-seventh Annual Meeting of the Health Physics Society. Tampa, FL.Health Physics. 82(6): S150; 2002.

Simpkins, A.A.; Hamby, D.M. Uncertainty estimates for parameters used for dose calculationsfrom amospheric releases of I-131 at SRS from 1955-1961. In: Proceedings of the 35thMidyear Health Physics Topical Symposium. January, 2002.

Harvey, R.P.; Hamby, D.M. Uncertainty in particulate deposition for 1 µm AMAD particles in anadult lung model. Radiation Protection Dosimetry. 95(3): 239-247; 2001.

Weber, A.H.; Buckley, R.L.; Parker, M.J.; Harvey, R.P.; Hamby, D.M. The creation of anhistorical meteorological database for dose reconstruction. Westinghouse Savannah RiverCompany, Aiken, SC: WSRC-TR-2001-00275. June 2001.

Harvey, R.P.; Hamby, D.M.; Uncertainty of the age-specific inhalation model-deposition. In:Proceedings of Radiation Protection for our National Priorities: Medicine, the Environment,and the Legacy. American Nuclear Society. Spokane, WA; pg. 393-400; September 17-21,2000.


5

Harvey, R.P.; Hamby, D.M. Uncertainty of the inhalation model. Proceedings of the Forty-fifthAnnual Meeting of the Health Physics Society. Denver, CO. Health Physics. 78(6): S116;2000.

Hamby, D.M. Sensitivity of the Gaussian atmospheric transport model to parametric variability.Proceedings of the Forty-fifth Annual Meeting of the Health Physics Society. Denver, CO.Health Physics. 78(6): S142; 2000.

d. “Radiological Characterization in the Vicinity of Lake Issyk-Kul, Republic ofKyrgyzstan”. David Lindeman, CRDF, Arlington, VA, 703-526-9720,[email protected], Grant No. YB1-121.

(3 Publications) This research identified areas of high radiation exposurearound the shoreline of Lake Issyk-Kul, in eastern Kyrgyzstan, and determine theextent and source of contamination in lake-water, biota, sediments, and soils ofoutlying areas. The team determined concentrations of natural thorium, uranium,and potassium and correlated those concentrations with radiation levelsdetermined as part of the NATO project.

Hamby, D.M.; Tynybekov, A.K. Uranium, Thorium, and Potassium in Soils along the Shore ofLake Issyk-Kyol in the Kyrghyz Republic. Environmental Monitoring. G.Bruce Wiersma(ed). CRC Press. New York. pg. 371-378; March 2004.

Hamby, D.M.; Tynybekov, A.K. Uranium, thorium, and potassium in soils along the shore ofLake Issyk-Kyol in the Kyrghyz Republic. Environmental Monitoring and Assessment. 73(2):101-108; 2002.

Hamby, D.M.; Tynybekov, A.K. Environmental radioactivity in the Kyrgyz Republic.Presentation to the Cascade Chapter of the Health Physics Society. Kelso, WA: February 11,2000.

e. “Radiation Levels in the Vicinity of Lake Issyk-Kul, Kyrgyzstan”. L. Veiga daCunha, NATO, Scientific and Environmental Affairs Division, Brussels, 02-728-41-11, Grant No. LG.960619.

(3 Publications) The team developed and executed a systematic program tocharacterize the radiological environment of the Lake Issyk-Kul region, and todetermine the major sources and routes of contamination. This researchcomponent initiated the program design and conducted screening measurementsof ambient exposure rates near the shoreline of Lake Issyk-Kul. The screeningphase served as an immediate indicator of the levels of radiation in the area andallowed a focused investigation of those areas of high contamination (the CRDFproject).

Hamby, D.M.; Tynybekov, A.K. A screening assessment of external radiation levels on the shoreof Lake Issyk-Kyol in the Kyrghyz Republic. Health Physics. 77(4):427-430; 1999.

Tynybekov, A.K.; Hamby, D.M.; Doronova, A.K. Radiological description of the coastal zone ofLake Issyk-Kul. In: Environmental Surroundings and People's Health. Ministry of PublicHealth. Bishkek, Kyrgyz Republic. Vol. 7: pg. 100-108; June 1999 (in Russian).


6

Tynybekov, A.K.; Hamby, D.M. Radiological description of the southern coast of Lake Issyk-Kul.In: A Collection of Works. Ministry of Labor and Social Protection. Bishkek, KyrgyzRepublic. Vol. 2: pg. 9-17; April 1999 (in Russian).

f. “Study of Field Methods and Worker Risks for Processing Alternatives toSupport Guiding Principles for FUSRAP Waste Materials”. (co-PI with James E.Martin at the Univ. of Michigan). Shirley Vogel, Contracting Officer, DOE OakRidge, 423-576-0648, Grant No. DE-FG05-96EW00001.

(4 Publications) This work consisted of an investigation of worker risksassociated with various ore processing methods, a determination of theeffectiveness of alternative processes to modify FUSRAP waste materials fordisposal, the effect of these various processes on future radiation risks to personexposed to such wastes, and the potential impact of EPA standards on these risks.The study included a development of field assessment techniques for determiningthe concentrations of thorium-230 in FUSRAP waste and a recommendation of afield protocol that balances reliability and costs of field equipment versuslaboratory measurements.

Porter, R.D.; Hamby, D.M.; Martin, J.E. Treatment methods and comparative risks of thoriumremoval from waste residues. Office of Environmental Management. Department of Energy.Grant No. DE-FG05-96EW00001. University of Michigan. Ann Arbor, MI: July 1997.

Bush, S.P.; Hamby, D.M.; Martin, J.E. Preliminary development of a wall-less gas-filledproportional counter for in-situ field analysis of nuclear contamination in soil. Office ofEnvironmental Management. Department of Energy. Grant No. DE-FG05-96EW00001.University of Michigan. Ann Arbor, MI: March 1997.

Hamby, D.M. Remediation techniques supporting environmental restoration activities. TheScience of the Total Environment. 191(3):203-224; 1996.

Porter, R.D. "A Literature and Current Procedure Review of Thorium Soil Extractions.” Master’sThesis. University of Michigan School of Public Health. 1996.

7. Foreign Nationals Supported on this Project

Abi Tavakoli-Farsoni (qualified US permanent resident, granted indefinite asylum; Iraniancitizenship) is currently a PhD Candidate in the Department of Nuclear Engineering andRadiation Health Physics, under the direction of Dr. Hamby. It is expected that he will graduateduring the 2006 calendar year. Upon being awarded the PhD, Mr. Tavakoli-Farsoni will be theprimary candidate for the position of Professional Faculty Research Associate, funded by thisproject. Mr. Tavakoli-Farsoni’s doctoral degree, in its entirety, is being funded under Dr.Hamby’s NEER grants (DE-FG07-02ID14331 and DE-FG07-05ID14704).

8. NEPA Environmental Check List

attached (next page)


7

ID-EC98.1 APPLICANT ENVIRONMENTAL CHECKLIST

The following information must be provided to and approved by the Department of Energy (DOE) before acontractual document can be awarded. Complete and correct information expedites the review process.

SECTION A:Project Title: A Multi-Layer Phoswich Radioxenon Detection SystemApplicant Organization: Oregon State UniversityApplicant Organization Contact: David M. HambyTelephone Number and Email Address 541-737-8682 [email protected]

SECTION B: Attach a complete and concise description of the project or activity. Include purpose and needand enough information so that a verification of the impacts can be performed. This allows DOE to make theproper NEPA determination. SECTION C: SOURCES OF IMPACTS: WOULD THE PROPOSAL INVOLVE OR GENERATE ANY OF THEFOLLOWING? (If yes, please provide brief explanation. For example, if yes is checked for question 15,indicate how much waste will be generated and the office or procedure in place to handle disposal.)

YES NO YES NO

1. Air Emissions X 10. Contaminated Soil X2. Asbestos Emissions or Waste X 11. Industrial Waste Generation X3. Biological Hazards X 12. PCBs X4. Discharge of Wastewater X 13. Hazardous Waste Generation X5. Cultural/Historical Resources X 14. Radioactive Waste Generation X6. Soil Disturbance X 15. Mixed Waste Generation X7. Radioactive Material Use* X 16. Chemical Waste Disposal X8. Water/Well Use X 17. Interaction with Wildlife/Habitat X9. Work Within a Floodplain X 18. Chemical Use/Storage X*only radiation check/calibration sources to be used in this project.

SECTION D: CATEGORY EVALUATION CRITERIA, WOULD THE ACTION:YES NO

1. Require cultural, historical, or biological clearances? X2. Impact sensitive resources identified in Item 1 above? Describe the mitigation plan. X3. Require or modify federal, state, or local permits, approvals, etc.? X4. Create hazardous, radioactive, PCB, or mixed waste for which no disposal is available? X5. Require siting, construction, or modification of a RCRA or TSCA regulated facility? X6. Is the activity included in an Environmental Impact Statement or EnvironmentalAssessment?

X

SECTION E: CERTIFICATION. To the best of the applicant' s knowledge at the time of signing, theresponses given above are complete and accurate, and should new issues or concerns arise or changes occuranytime after award and during the course of performance, the applicant will alert DOE immediately.

_____e-signature 07/08/05 14:18:10 PDT______________________ ____07/08/2005__________APPLICANT SIGNATURE & TITLE DATE

E USE ONLY

NEPA Doc Number: Solicitation #:

NEPA CX Applied: Contract Specialist:Approved:

Signature/Date: Project Manager:

a multi-layer phoswich radioxenon · pdf filedavid m. hamby, oregon state university 1 volume...

Documents