Byron Kratochvil

Statistical Discrimination of Flat Glass Fragments by Instrumental Neutron Activation Analysis Methods for Forensic Science Applications

Instrumental neutron activation analysis (INAA) and cyclic INAA (CINAA) techniques were used for the discrimination of flat glass fragments. Employing short irradiation, delay, and count times, sample test portions of 100 µg and 1 mg (INAA) or 1 and 5 mg (CINAA) were considered. A statistical model was used to assign the elemental variation among glasses to different experimental factors; by advanced statistical methods elemental differences between specific glasses can be identified. Analysis and comparison of float glass panes produced by two major Canadian manufacturers were also performed; the two were readily differentiated, primarily on the basis of their aluminum levels.

MODELLING OF RADIUM-226 LEACHING FROM BARIUM-RADIUM SULFATE SLUDGES

Abstract A geochemical computer model was applied to the leaching of (Ba,Ra) SO4 sludge produced in the treatment of uranium mining and milling effluents to remove 226Ra2+. Previous laboratory studies had shown the following leachant characteristics to be important: initial SO42− and Ba2+ concentrations, pH, and ionic strength. The computer simulations confirmed in a general way previous experimental data, but more importantly showed that the most significant variable was the redox state of the sludge-leachant system. Under all oxidizing conditions (Ba,Ra)SO4 remains sparingly soluble. When conditions become reducing, however, BaSO4(s) is not thermodynamically stabe and the entire mass of 226Ra2+ in the sludge can be transformed to the aqueous phase. Substantial release under reducing conditions was demonstrated in experiments involving bacteria. The use of geochemical computer models represents a fundamental approach to leaching investigations and allows identification of potentially critical environmental conditions which should be investigated experimentally.

Sampling Errors in the Determination of Cocaine in Seized Drugs

Experiments in which 28-g (1-oz) seizures of cocaine plus diluent (mannitol, inositol) were ground in a mortar for only a few minutes before removal of 20-mg portions for assay (by gas chromatography using bupivicaine as internal standard) showed the standard deviation of the sampling operation to be several times larger than the standard deviation of the analytical operations. Measurement of the particle size distribution of ground mixtures allowed estimation of the weight of sample required to lower the sampling standard deviation to any specified level. The sample weight required was shown to depend in a predictable way on the percentage of cocaine present in the material.

Analysis of Athabasca oil sand by near-infrared–diffuse reflectance spectroscopy

A near-infrared–diffuse reflectance spectroscopy (NIR–DR) method is described for the measurement of the bitumen and solids content in Athabasca oil sand over a 15 m vertical transect of a 7 cm diameter core at a sampling resolution of 1 cm. Principal component regression and partial least squares statistical techniques were evaluated as tools for optimizing the calibration models. The two techniques were found to be comparable and satisfactory for this application. Surface roughness was found to exhibit a significant effect on bitumen results. Sampling frequency studies showed that to attain a target precision of 5% for bitumen and solids, and to reduce errors arising from heterogeneity, measurements should be made on areas larger than 1 cm2.

Combined Benedetti-Pichler/Stange–Poole sampling equation for two-component particulate mixtures

General equations for the estimation of statistical uncertainty in the sampling of mixtures of two types of particles were developed that take into account variability in particle size and composition. Different expressions were obtained for situations where the component of interest is present in bulk or trace amounts. The general equation for a bulk component was verified experimentally by sampling KCl–NaCl and KCl–sucrose mixtures with various particle sizes and compositions with a spinning riffler and determining the relative standard deviation of the percentage of Cl or KCl under conditions where the analytical uncertainty was small.

Characterization of trace amounts of aluminium in biological reference materials by electrothermal atomic absorption spectrometry

The homogeneity of aluminium at the ppm level in a set of reference materials provided by Agriculture Canada was determined by electrothermal atomic absorption spectrometry. The materials included bran, wheat flour, gluten, whole egg powder, whole milk powder, meat powder, starch, cellulose, corn kernel and corn stalk. Reproducible values were obtained, which indicated that all of the materials studied can be considered to have a homogeneous distribution of aluminium between samples at test portion sizes of 250 mg. All of the materials, except milk, egg and meat, required HF in addition to HNO3 to effect complete dissolution, indicating the presence of mineral matter.

Sampling theory for environmental analysis

able. The benchmarks given in the introductory report2 are for the Whetstone benchmark and the results are given in Table I. Subsequently, they have published comparative data for IEEE single and double precision performance on the ‘LINPACK’ linear algebra algorithms3 and these are presented in Tables II and III, respectively. These boards with software cost under US

Potentiometric titrations of metal ions in acetonitrile with polyamine ligands

3300 and thus produce substantial computational power for much less cost than a standalone workstation, and the PC or AT clone that houses it can be used for all of the standard MS-DOS software. By using a multitasking program like DoubleDOS, the DSI board can be running in a 90 kilobyte partition in the background and the rest of the memory is available for DOS applications like word processing or data entry. Thus, low cost, easily used, high computational power is readily available by using either of these coprocessor boards.

Uncertainty in sample mass for small numbers of particles from granular materials

Ethylenediamine, diethylenetriamine, and triethylenetetramine were investigated as titrants for metal ions in acetonitrile. Copper(II) gave the best titration plots; cobalt(II), manganese(II), nickel(II), iron(III) and magnesium(II) also gave results acceptable for analytical determinations. Platinum, silver-silver(I), mercury-mercury(II), carbon, and a copper ion-selective electrode were studied as indicating electrode systems; of these platinum gave the sharpest and largest inflections. The mechanism by which platinum responds as an indicating electrode in these complexation titrations is unclear.

Coulometrically generated copper(II) in acetonitrile as an analytical oxidant.

The standard deviation of sample mass was quantitatively related to the number of particles in the sample, individual particle masses and the fractions of the numbers of different types of particles in the mixture. The theory was verified by sampling of cereal grain mixtures with a spinning riffler and Monte Carlo computer simulation. The theory is applicable to random sampling of well-mixed populations containing two or more types of particles.