Search

For more than 50 years, nuclear methods have been applied to chemical analysis at the National Institute of Standards and Technology. Radiochemical, instrumental, and prompt-gamma activation analysis are used, as well as neutron depth profiling and other techniques. The history of this group in methods development and the certification of Standard Reference Materials, among other applications, is reviewed.

Prompt gamma neutron activation analysis has become an important part of the analytical toolkit, practiced at several research reactors worldwide. An extensive review of the physics, engineering, and applications of the technique was published in 2004: the present work gives an overview and survey of the literature in the succeeding dozen years.

A new instrument for cold neutron prompt gamma-ray activation analysis is being designed as part of the recent expansion of the cold neutron guide hall at the NIST Center for Neutron Research. This instrument has a 10-fold increase in neutron flux and lower gamma-ray and neutron background compared to the PGAA instrument it will replace at NG7. Monte Carlo based simulation sofware and experimental setups are used in the design phase to mitigate background while preserving high neutron fluence. The new instrument will also provide more space for samples and for experimental setup compared to the old facility.

It has been difficult to characterize the thermal to epithermal neutron flux ratio (f) and the measure of the nonideal epithermal neutron flux distribution (α) for the RT-2 pneumatic rabbit facility at the NIST National Bureau of Standards Reactor (NBSR). In a previous paper, only cadmium-covered irradiations yielded physically reasonable parameters. New measurements were performed using chromium, manganese, cobalt, zinc, zirconium, molybdenum, antimony, gadolinium, lutetium, and gold. The neutron temperature (T n ) in RT-2 measured using bare lutetium and gold foils gave unphysical values. The bare foil methods for measuring f and α gave inconsistent results. The underlying reasons are demonstrated via MCNP simulation results for cumulative reaction rates of selected isotopes. To determine expected intervals for f, α and T, parametric methods were explored. Measured reaction rate probability per target atom (R p ) values for the listed elements were fitted to a modified Westcott curve using an iterative least-squares method to verify consistency of measurements and nuclear data. An advanced parametric approach using a detailed MCNP model of the NBSR was used to calculate neutron flux characterization parameters.

In an attempt to improve the accuracy of neutron activation analysis, gamma-ray spectrometry and least-squares methods developed for other nuclides have been applied to redetermine the decay constants of 90mY and 97Ru. Several irradiated samples of each pure element were counted continually through several half-lives, acquiring up to 100 spectra with good statistics. Photopeaks from each nuclide were evaluated with a region of interest integration routine, and each data point corrected for rate-related losses before fitting to an exponential function by a nonlinear least-squares procedure. The half-lives found are 3.204 ± 0.008 h for 90mY and 2.836 ± 0.014 days for 97Ru, where the uncertainty is an estimate of the 95 % confidence level. Although (by Chi squared criteria) the fits to the decay curves were very good and the uncertainty of the resulting half-lives an order of magnitude smaller than the previously published uncertainty, care is necessary before assigning a best value with a totally defensible uncertainty to the overall data set.

An automated delayed-neutron activation analysis system has been installed at the NIST research reactor. This work involved characterization of the transfer time of the system, evaluation of blanks, and tests of the system’s analytical capabilities through quantitative analysis for uranium in several natural matrix standard reference materials (SRMs). The calibration curve was shown to be linear up to at least 20 μg of uranium, and the well-thermalized reactor irradiation position makes the system insensitive to thorium and oxygen. For SRMs 1646a Estuarine Sediment and 2710a Montana Soil the values determined for this work agree with the reference and certified values, respectively. The mass fraction of uranium in SRM 695 Multi-Nutrient Fertilizer is the first reported for this material. For this system and the irradiation, transfer, and counting times used, the limit of detection for natural uranium is 20 ng, which corresponds to approximately 200 pg of 235U.

Nearly fifty years ago, two landmark papers appeared that should have cured the problem of ambiguous uncertainty statements in published data. Eisenhart’s paper in Science called for statistically meaningful numbers, and Currie’s Analytical Chemistry paper revealed the wide range in common definitions of detection limit. Confusion and worse can result when uncertainties are misinterpreted or ignored. The recent stories of cold fusion, variable radioactive decay, and piezonuclear reactions provide cautionary examples in which prior probability has been neglected. We show examples from our laboratory and others to illustrate the fact that uncertainty depends on both statistical and scientific judgment.

A nondestructive analytical method based on autoradiography and gamma spectrometry was developed to perform activity distribution analysis for particulate samples. This was applied to aerosols collected in Fukushima Japan, 40 km north of the Daiichi nuclear power plant for a 6 week period beginning shortly after the March 2011 tsunami. For an activity distribution of 990 “hot particles” from a small filter area, the hottest particle was nearly one Bq 137+134Cs but most of the activity in the filter was produced by particles having

Analytical chemists are scientific scavengers, using any available physical phenomenon to determine the composition of materials. Accuracy, unlike mere reproducibility, depends on understanding the measurement process, most usefully by stressing it until it bends. The use of neutron capture gamma rays for analysis involves some issues not encountered in conventional neutron activation analysis which, when understood, extend the utility and reinforce the value of nuclear methods of elemental analysis. These methods are being increasingly recognized as tools for SI-traceable metrology.

A new instrument for cold neutron prompt gamma-ray activation analysis (CNPGAA) is being designed and constructed at the NIST Center for Neutron Research (NCNR). The new instrument is expected to have lower gamma-ray and neutron background and better detection limits for most elements than the current cold neutron PGAA instrument. Other advantages over the current facility will include the ability to analyze larger samples and greater overall measurement capability due to the addition of scanning stages, cryostats, and sample changers.

For accurate activation analysis, it is necessary to understand the neutron and gamma-ray environment experienced by a sample during irradiation. A new irradiation position was recently installed in the RT2 thimble of the NIST reactor. Modeling by MCNP was used to select the radial location in the reflector thimble and the size of the D2O shielding tank. After installation, the new rabbit was characterized for thermal, epithermal, and fast fluence rates, spectral parameters f and α, temporal stability, axial and radial flux gradients, and temperature.

The basic assumptions of activation analysis are that the induced radioactivity is proportional to the amount of analyte, and that the quantity of radioactivity can be related simply to the number of counts observed. Quantitative measurement of activity (and of its uncertainty) is not always simple, especially when accuracy better than a few percent is sought. Recent work with 77Ge and 76As has demonstrated that the accuracy of half-lives in the literature is sometimes insufficient. Despite these and other problems, quantitative understanding and documentation of uncertainties can be accomplished, providing demonstrable quality assurance and supporting claims of traceability to the Systeme International.

Two prompt gamma-ray activation analysis (PGAA) facilities at the NIST Center for Neutron Research have been used routinely to perform elemental analyses of a variety of materials. Results from these analyses are usually expressed as mass fraction values with expanded uncertainties. The expanded uncertainty consists of the combined uncertainty multiplied by the appropriate coverage factor (k) required to achieve a 95% confidence interval. The combined uncertainty includes the uncertainties associated with preparation, irradiation, and γ-ray spectrometry of samples and standards, and corrections for γ-rays from the background or blanks where necessary. To determine the combined uncertainty, each component of uncertainty associated with each variable and constant in the basic measurement equation is evaluated. In this paper we present the PGAA measurement equation, a description of the potential sources of uncertainty for each component of the equation, and three examples of uncertainty evaluation. The examples are for determination of H in standard reference material (SRM) 2454, hydrogen in titanium alloy using the cold neutron PGAA facility, Cd in SRM 2702 Inorganics in Marine Sediment using the original thermal neutron PGAA facility, and N in SRM 3244 Ephedra-Containing Protein Powder using the recently designed thermal neutron PGAA facility.

This work provides a systematic experimental approach to the demonstration of viability of instrumental neutron activation analysis (INAA) in chemical metrology. The practical approach was derived from a complete survey of uncertainty components that affect the INAA measurement process. These uncertainty components were classified by their magnitude and origin and subsequently minimized by appropriate steps in the INAA process. The process was tested with the INAA determination of Cr in SRM 1152A stainless steel; the Cr value is certified at 17.76% with an estimated uncertainty of 0.04% (0.23% relative). The INAA results from this procedure are in agreement with these specifications. Similar procedures have been applied to INAA multi-element determinations in a high temperature alloy. Agreement with available consensus values was demonstrated in the alloy. The guidelines on the determination of uncertainty were fully met, providing through INAA a valuable independent non-destructive tool in chemical measurements of metrological value such as required in the CCQM key comparisons.

As a contribution toward improving the accuracy of neutron activation analysis, the half-life of 111mPd was measured by gamma-ray spectrometry. The uncertainty of this parameter is greater than desired for accurate work, largely because 111mPd represents a small part of the total activity of palladium exposed to neutrons. In this work, several irradiated samples of pure palladium were counted continually with high-resolution Ge gamma detectors through about ten half-lives. The 172 keV photopeak was integrated with a region-of-interest procedure, and each datum corrected for rate-related losses before fitting to an exponential function by a nonlinear least-squares procedure. The half-life found is 5.580 ± 0.032 h, where the uncertainty is an estimate of the 95 % confidence level, a substantial improvement over published values.

Prompt-gamma activation analysis (PGAA) differs from conventional neutron activation analysis in the set of elements measured, the irradiation conditions, and the nature of the gamma-ray spectra. Each of these affects the conduct of the measurements and interpretation of the data. In its recent development, the k0 standardization framework provides a solution to many of the problems of this analytical method, and may provide a bridge to greater certainty of analysis.

We have been exploring ways to evaluate the activity of radioactive particles that have been detected by phosphor plate digital autoradiography based on photostimulated luminescence (PSL). A PSL system with 25 μm pixel digitization has been applied to particle analysis problems, both qualitatively and quantitatively. Two data evaluation methods are currently employed: (1) bulk area signal measurement, and (2) discrete event counting which may include spectral evaluation. The first method is conventional, whereas the second method requires high spatial resolution and is presented here for the first time. The counting methods can discriminate between alpha and background counts. The unshielded background signal accumulation rate was determined by the bulk area method. Using the spectral method of evaluation for α-particle events, the mean signal intensity per recorded α-particle was measured, and the detective quantum efficiency (DQE) was found to be nominally 100%. We present a comparison to gamma-spectrometry for sub-Bq 137Cs activities, and demonstrate an application for the qualitative assay of International Atomic Energy Agency swipe samples collected from uranium enrichment facilities.

Prompt gamma-ray activation analysis (PGAA) is an important nuclear analytical technique that complements conventional neutron activation analysis (NAA). When a target is placed in a beam of neutrons, gamma-rays emitted upon neutron capture are measured by a shielded germanium detector, yielding quantitative elemental analysis. The radiation is penetrating and the analysis both nondestructive and independent of the chemical form of the element(s) being measured. The technique is most useful for measurement of light elements (H, B, C, N, Si, P, S, Cl) which can not be easily measured by other methods. Best sensitivity is achieved with neutron beams from research reactors. Although sample preparation is minimal, care must be taken to select proper standards and blanks, and numerous corrections must sometimes be applied to the data from the complex spectra. PGAA has proven useful for multielement analysis of a wide variety of different materials spanning a broad range of scientific disciplines. Of particular importance has been the measurement of hydrogen in materials.

Standard reference material (SRM) 2134 Arsenic Implant in Silicon was produced at the National Institute of Standards and Technology (NIST) as a calibrant for secondary ion mass spectrometry. Instrumental neutron activation analysis was used as a primary method for certification of the arsenic implanted dose. A complete evaluation of all sources of uncertainty yielded an expanded relative uncertainty for the mean value of this SRM to be 0.38% at approximately the 95% level of confidence. No evidence indicating significant heterogeneity among samples was observed.

A simple and accurate method is presented to evaluate the burnup effects involved in the neutron activation of 197Au prior to any neutron flux characterisation, based on the spectrometry of the 198Au and 199Au decay gammas. The obtained burnup factor can be used as input for reactor neutron field characterisation techniques using 197Au(n,γ)198Au as a monitor. This way an iterative procedure is avoided.

Applying thek 0 standardization method to prompt-gamma activation analysis (PGAA) offers similar benefits as in instrumental neutron activation analysis. It has been demonstrated that under constant flux conditionsk 0-factors obtained by normalizing to a titanium comparator, measured separately, yield consistent analytical sensitivity ratios. The ratio method has been generalized by using stoichiometric compounds for the determination ofk 0-factors. Since chlorine forms compounds with essentially everyelement and it also serves as a detector efficiency standard,k 0 values have been determined relative to chlorine as an internal standard for several analytically important elements in two reactor facilities: the thermal guided beam at the BRR in Budapest and the cold-neutron beams at the NBSR at NIST.

In the past decade, the combination of readily accessible, reliable data in electronic form with well-tested spreadsheet programs has changed the approach to experiment planning and computation of results. This has led to a flowering of software applications based on spreadsheets, mostly written by scientists, not by professional programmers trained in numerical methods. Formal quality systems increasingly call for verified computational methods and reference data as part of the analytical process, a demand that is difficult to meet with most spreadsheets. Examples are given of utilities used in our laboratory, with suggestions for verification and quality maintenance.

Commercially available pulse generators for use in γ-ray spectroscopy to correct for dead-time and pile-up losses share the problem that the fall-time of the output pulses does not match the fall-time of the detector pulse. Upon pulse shaping in the main amplifier the pulser's signals result in a considerable undershoot of the amplifier's output signal, which cannot be restored by theP/Z network. As a result, peaks in the γ-ray spectrum are broadened. Moreover, amplitude stability of the generators is often poorer than the overall gain stability of the spectrometer. To overcome these problems a special pulse generator has been developed at the Interfaculty Reactor Institute, which simulates Ge-detector signals. It generates pulses with a constant frequency of 25 Hz using a crystal oscillator controlled mercury relay. The stability of the generator's amplitude and frequency is extremely high and the peak-shapes in the γ-ray spectrum are not distorted.

The cold neutron capture prompt γ-ray activation analysis (CNPGAA) spectrometer located in the Cold Neutron Research Facility (CNRF) at NIST has proven useful for the analysis of hydrogen and other elements in a wide variety of materials. Modifications of the instrument and the CNRF have resulted in improved measurement capabilities for PGAA. The addition of an atmosphere-controlled sample chamber and Compton suppression have reduced γ-ray background and increased signal-to-noise ratio. More recent revisions are expected to yield still further improvement in analytical capabilities. Replacement of the D2O ice cold source with a liquid H2 moderator is expected to yield a 5–10 fold increase in neutron capture rate, and improved neutron and γ-ray shielding will result in further reduction of the background. Other modifications to the instrument allow easier sample mounting and more precise positioning of samples in the neutron beam. Significant improvements in detection limits and analytical accuracy are expected.

Apparatus is being developed for short irradiation and rapid counting at the NIST Reactor, optimized for accurate neutron activation analysis via activation products ranging in half-life from about 500 ms to 500 s. This facility is designed to irradiate a sample either in a well-thermalized neutron flux at 3·1013 n·cm−2·s−1, or in a higher flux with a larger fast and epithermal component. The design transfer time for a 1-ml rabbit is 500 ms, measured to 10 ms precision. Timing information for both irradiation and counting will be transferred automatically to the activation analysis workstation computer. The γ-ray spectrometer system is selected and tuned for accurate measurement at high and varying counting rates, using loss-free counting (LFC) technology. A detailed calibration and characterization of this system has been performed. Accurate measurement requires that attention be paid to the systematic and random errors to which LFC is subject, but this requirement is minor compared to the advantages of undistorted spectral shape, the ability to solve the decay equations exactly, and the wider useful range of counting rates in the spectrometer system.

A neutron lens has been constructed to focus cold neutrons from the exit of a58Ni neutron guide, which delivers a beam to the Prompt-Gamma Activation Analysis (PGAA) station at the NIST Cold Neutron Research Facility. The lens compresses a neutron beam of cross section 50 mm× 45 mm onto a focal spot of diameter 0.53 mm (fwhm) wich an average gain of 80 in neutron current density. PGAA measurements have been performed to demonstrate the enhanced sensitivity and detection limits for various elements and the spatial resolution in one transverse dimension. For the two test particles (a gadolinium glass bead and cadmium metal of sizes less than 0.5 mm), the gain in the γ-count rate with the lens is a factor of 60, and the detection limit is improved by a factor of 20. The system can be used for two-dimensional mapping of samples on a sub-millimeter scale to complement other analytical techniques such as neutron depth profiling (NDP).

Radiochemical neutron activation analysis has proven to be a powerful tool for measuring trace mercury in a variety of materials. In a modification of an established NIST method, selenium interfering with the 203Hg indicator is removed by precipitating mercury as periodate, and the chemical yield is measured gravimetrically. The current method is better suited to multiple samples, less subject to explosion during the combustion step, provides a measurement of the yield for each sample, and involves less radiation exposure to personnel. The procedure has been applied to the certification of mercury in SRMs 1575a Pine Needles, 1632c Bituminous Coal, and 2702 Marine Sediment.

Methods developed to redetermine the 76As half-life were applied to four additional radionuclides. Several gamma-spectrometers were used to follow the decay of pure sources during 4-14 half-lives, comparing with a pulser stable to better than 2 . 10-8 per day. Complementary measurements (also for 76As) employed a pressurized ion chamber. Several least-squares methods were used to derive the best value of each decay constant. We find the half-lives to be 14.959±0.009 hours for 24Na, 12.339±0.011 hours for 42K, and 2.6924±0.0011 days for 198Au, where the uncertainties are the combined standard uncertainty u.

At the National Institute of Standards and Technology, there are two techniques for chemical analysis that use neutron beams from the reactor for target irradiation: neutron depth profiling (NDP) and prompt γ-ray activation analysis (PGAA). There are two facilities for each technique, one equipped with a thermal neutron beam and the other, with a cold neutron beam. In addition, focused beams of cold neutrons will be used to measure the two-dimensional element distributions by PGAA and three-dimensional distributions by NDP. This paper includes a brief description of the facilities, the measurement capabilities of each, some recent applications of NDP and PGAA, and neutron focusing as applied to these techniques.

Determination of53Mn in meteorites by neutron activation analysis requires a thermal neutron flux high enough to ensure adequate production of54Mn from53Mn with a sufficiently low fast neutron component to minimize its production through fast neutron reactions. Thermal and fast neutron fluxes were mapped as a function of sample position within the NIST research reactor in order to determine the optimum position for irradiation of53Mn.

In the course of making high-accuracy measurements of arsenic, we found that the most recently published and compiled half-life of 76As did not agree with our data as well as the earlier accepted value. To redetermine this parameter, 76As sources were measured on four Ge gamma detector systems, and an exponential function was fitted to the decay data by two different nonlinear least-squares methods. We obtained T1/2 = 1.09379 days with a standard uncertainty of 0.00045 days. This result is 1.5% higher than the most recent value, but is in agreement with the older, less precise, consensus value.

An instrument for prompt gamma-ray activation analysis is now in operation at the NIST Cold Neutron Research Facility (CNRF). The cold neutron beam is relatively free of contamination by fast neutrons and reactor gamma rays, and the neutron fluence rate is 1.5·108 cm−2·s−1 (thermal equivalent). As a result of a compact target-detector geometry the sensitivity is better by a factor of as much as seven than that obtained with an existing thermal instrument, and hydrogen background is a factor of 50 lower. We have applied this instrument to multielement analysis of the Allende meteorite and other materials.

By irradiating with cold neutrons and avoiding hydrogenous materials of construction, we have developed a PGAA instrument at the Cold Neutron Research Facility at NIST with hydrogen detection limits in the microgram range in many materials. Quantities of 5–10 μg H/g are presently measurable in gram-sized samples of silicon or quartz, and of order 0.01 wt % can be quantitatively measured in complex silicate rocks.

Analyses for certification have been made for the determination of 30 elements in the National Institute of Standards and Technology (NIST) Tomato Leaves renewal reference material, SRM 1573a. Three of the analytical techniques used were instrumental neutron activation analysis (INAA), radiochemical neutron activation analysis (RNAA), and prompt gamma activation analysis (PGAA). These techniques provided data on 19 elements by INAA, 10 elements by PGAA, and 7 elements by RNAA, with some overlap between techniques. For example, INAA was able to obtain overall analytical uncerainties (at the 95% confidence level) averaging ±2.2% for major and minor constituents (Ca, Mg, K), ±3.3% for constituents from 1 to 1000 μg/g (Na, Fe, Al, Mn, Ba, Zn, Rb, La, Cr), and ±6.4% for elements between 10 and 1000 ng/g (Co, V, Se, Th, Sc, Sb), using sample dry weights of approximately 150 mg. These analyses represent the most extensive use to date of nuclear analytical techniques in the certification of a trace element SRM at NIST.

Both at IRI and at NIST, Local Area Networks (LANs) are being used to acquire and process data from multiple gamma-ray spectrometers. In this paper, differences and similarities between three systems are discussed, resulting in recommendations for new systems to be set up.

An instrument for neutron capture prompt gamma-ray activation analysis (PGAA) has been constructed as part of the Cold Neutron Research Facility at the 20 MW National Institute of Standards and Technology Research Reactor. The neutron fluence rate (thermal equivalent) is 1.5·108 n ·cm−2·s−1, with negligible fast neutrons and gamma-rays. With compact geometry and hydrogen-free construction, the sensitivity is sevenfold better than an existing thermal instrument. Hydrogen background is thirtyfold lower.

Neutron activated reference materials have been analyzed using a standard Ge(Li)-detector with 17% relative efficiency, a very large Ge-detector with 96% relative efficiency, and a well-type Ge detector. Sensitivities are presented, and usefulness of these systems for NAA is compared on the basis of performance, economics and complexity.

I found the recent article by Cho and Park (Analysis of hydrogen concentration in low-alloy steel, Journal of Radioanalytical and Nuclear Chemistry 284 (2010), 533–537) most tantalizing. The authors do not give enough details for me to understand how they are able to accurately measure 0.1 lg of hydrogen (1 ppm in a 100 mg sample) in the presence of a 13 lg background, especially since the sample is ‘‘irradiated while attached to a paper filter.’’ They validated their measurements by comparison with ICP-AES, a claim that my colleagues in that field are unable to understand. Although hydrogen can be detected via the 656 nm line (Spectrochim. Acta 40B (1985), 177–194], since that method in general practice requires the sample to be in water solution, a detailed reference to their procedure is needed. I look forward to reading their future publications.

Guided beams of cold neutrons being installed at a number of research reactors may become increasingly available for analytical research. A guided cold beam will provide higher neutron fluence rates and lower background interferences than in present facilities. In an optimized facility, fluence rates of 109 n·cm−2·s−1 are obtainable. Focusing a large area beam onto a small target will further increase the neutron intensity. In addition, the shift to lower neutron energy increases the effective cross sections. The absence of fast neutrons and gamma rays permits detectors to be placed near the sample without intolerable background, and thus the efficiency for counting prompt gamma rays can be much higher than in present systems. Measurements made at the hydrogen cold source of the FRJ-2 (DIDO) reactor at the KFA provide a numerical evaluation of the improvements in PGAA with respect to signal-to-background ratios of important elements and matrices.

A facility for neutron-capture γ-ray spectroscopy for analytical purposes has been developed and tested at the National Bureau of Standards reactor. The system consists of an internal beam tube with collimators, an external beam tube and irradiation station, a Compton-suppressed Ge(Li) γ-ray detection system, and a minicomputer-based data-collection and-analysis system. Detection limits have been established for many elements and errors arising from neutron self shielding, γ-ray peak overlap, neutron beam variations, and sample matrix evaluated.

Because of the short half life of Al-28, the determination of aluminum by neutron activation is subject to many inaccuracies: variation of irradiation conditions between sample and standard, uncertainties in timing, and the effects of high and varying count rate, in addition to other sources of error that must be controlled even in work with long-lived nuclides. These errors can all be made smaller than the fundamental limit set by counting statistics, even when that limit is below 0.5 percent. The transfer function from the observed number of net counts to the counting rate at the end of irradiation is modeled as a product of three processes: radioactive decay and extending and nonextending dead time.The procedure has been applied to the analysis of NBS SRM 1633a Fly Ash. The mean concentration measured was 14.085% Al, with a standard deviation of the mean 0.023% Al for four determinations. The final results showed no significant imprecision beyond counting statistics. The accuracy of the metho...

Major levels of boron in borosilicate glasses have been determined nondestructively by neutron activation analysis. In contrast to chemical methods for determining boron as a major component, the described nuclear method has few interferences and does not require chemical separation of boron prior to its quantitation. The effects of neutron self-shielding by boron (1 to 8% by weight) are examined, minimized by dilution of powdered samples with high purity graphite, and circumvented by comparative analyses. Results of the analysis of a series of glasses with increasing boron composition are 1.150±.005% and 7.766±.035% for the low and high members of the series. Accuracy of these results depends upon comparative analyses with the certified Standard Reference Material, SRM 93a (3.89±0.02% B). Once analyzed, the glasses are useful as secondary standards for alpha track counting, and also ion and electron microprobe analyses of glasses.