A review of the procedures and efforts at the National Bureau of Standards (NBS) to provide for meaningful measurements through the use of Standard Reference Materials (SRM's) is presented.

The examples of NBS Standardization efforts for x-ray analysis range from basic metrology to applied environmental measurements. These examples include a determination of x-ray wavelength by a method which in part utilizes simultaneous x-ray and optical interferometry measurements of the atomic planes of near perfect silicon. In addition, Standard Reference Materials (SRM's) are being developed and applied to trace element analysis using x-ray fluorescence techniques. These efforts include development of SRM's for trace element analysis of air particulates. In another area, work is proceeding on the development of a silicon powder Standard Reference Material intended for x-ray diffractometer calibration. An effort to develop a suitable x-ray diffraction technique to determine the amount of quartz in mine dust is also underway. NBS efforts to provide SRM's for the calibration of electron microprobes and the validating of correction factor calculations are also described.

The purpose and the value of standard reference materials (S.R.M.) are defined, and their increasing world importance is emphasized. The desirable features of an S.R.M. are outlined and the procedures necessary to meet, as far as possible, the necessary requirements in S.R.M. production are discussed. The method and organization of certification for homogeneity and composition are detailed, as well as the statistical proceduces to evaluate the certification analysis. The present Status and future plans for the production of S.R.M.'s by the Canadian Standard Reference Materials Project, Mines, Branch, Department of Energy, Mines and Resources, are outlined.

The identification of crystalline phases by x-ray diffraction, either by powder or single crystal techniques requires a dependable body of reference data. It is not only necessary to have data on each phase which are accurate and complete, it also is desirable to have data on as wide a range of compounds as possible, and to have the data organized in such a manner as to be readily usable. The outstanding compilations which approach these goals are the Powder Diffraction File and Crystal Data.

The Powder Diffraction File, published by the Joint Committee on Powder Diffraction Standards has data covering about 22,500 phases, both organic and inorganic. These data are of various degrees of accuracy as is indicated by symbols. The File is continuously being improved by the addition of evaluated data from the general literature and by data produced by supporting projects, the principal one being the Joint Committee Associateship at the National Bureau of Standards.

To be noted in the File with a star, and to be truly considered standard data a powder pattern must be complete in the sense of including all reflections above the minimum “d” spacing covered, both weak lines and those with large “d” spacings. Since the best test of a pattern is its own internal consistency, the reflections must all have hkl's assigned and must show a good agreement between the spacings observed and those calculated from a refined cell, and they must be consistent with the known space group. This agreement can be best obtained by the use of an internal standard and a computer program. The intensities should be measured by a method which minimizes the effect of crystal orientation.

The PDF is provided with search procedure manuals arranged on a scheme of the strongest lines to help in locating data matching that from an unknovm. A computer program for rapid searching is available. A recent development is the inclusion of a “reference intensity” to aid in estimating the quantitative analysis of mixtures.

Crystal Data is a compilation now in the third edition made at the National Bureau of Standards and published by the Joint Committee on Powder Diffraction Standards. It contains data on the unit cell parameters of over 24,000 phases. These data are arranged by crystal system and axial ratios to simplify identification of phases from unit cell data obtained from Single crystal cameras.

Both of these large compilations are also important reference sources for crystallographic information giving structural information and literature references.

Standards required for quantitative powder diffractometry are essentially of three types as shown in Table 1 - external instrument standards for the initial and long term check of instrument alignment (i.e., an “instrument” standard), external analytical standards, which are similar in composition to the specimens to be analysed, for the preparation of calibration curves for “in-type“ analysis; and internal analytical standards for the minimization of problems due to variations in specimen absorption.

The ideal equipment alignment standard should be free from preferred orientation, stable and available in large quantities, since it should serve the dual role of providing a continuing indication of the performance of a given diffractometer, as well as allowing absolute intensity comparisons to be made between different diffractometers.

Sampling is a necessary part of the chemical analysis of particulate matter where the objective is to characterize bulk properties since it is usually undesirable or impossible to test an entire lot. The sample must be a miniature replica of the bulk material at least in respect to those features being tested. In other words, the sample must be representative to permit extrapolation from the sample to the bulk.

There is error associated with sampling particulate material, but this sampling error is only one component of the total error associated with an analytical result. Sample preparation and determination of the property being tested are two other major sources of error, These three errors combine as the squares of their standard deviations, i.e., their variances, to produce the total analytical error. Very often, sampling error is the largest of the three and, therefore, contributes a disproportionately large share to the total error.

The requirements imposed on the standards and samples used for any analytical technique are determined by the sample utilization characteristics of each technique as well as the purpose of the analysis. A practical comparison of said sample utilization characteristics for several analytical techniques will be examined to define limitations imposed on the standards/sampling aspects of analysis.

Quantitative analysis of six clay minerals commonly found in drilling mud solids is studied by x-ray diffraction using the method of internal standards. Standard clay samples are used with three internal standards to derive calibration curves from which four synthetic mud solids are analyzed. If careful sample preparation and handling are employed, reasonably accurate and reproducible results are obtained.

In the quantitative x-ray diffraction analysis of clays, “standard“ clays are compared with “unknowns” assuming both Standards and unknowns have similar diffraction characteristics. This assumption is valid in certain cases involving kaolinites, chlorites, vermiculites, and where standards have been obtained from the unknown clays. In other cases; especially those involving smectites, illites, and some chlorites and vermiculites; lack of hornogeneity within mineral species and inability to measure this inhomogeneity in mixed mineral suites limits the accuracy obtainable by diffraction methods.

This paper reviews the factors in diffractometer geometry most likely to cause large errors in intensity measurements due to imperfect specimen preparation, particularly preferred orientation. It is recommended the same specimen be transferred to two or more diffractometer geometries in which the specimen has different orientations to obtain sets of complementary data. Three geometries are illustrated: θ-2θ scanning with reflection and transmission specimens, and Seeman-Bohlin with reflection specimen.

A unified matrix-flushing theory and its practical applications for quantitative multicomponent analysis by X-ray diffraction are presented.

In this method, a fundamental “matrix-flushing” concept is introduced; the calibration curve procedure is shunted; the matrix (absorption) effect is totally eliminated; all components, crystalline or amorphous, can be determined.

X-ray diffracfion analysis has been used to determine the nature and relative concentration of the phases present in diffusion welded joints of fairly high purity beryllium. The analyses were done in conjunction with the major objective of the work which was to correlate the types and amounts of phases in the weld joint with the weld processing variables.

The beryllium used contained less than one percent by weight of alloying elements. Those elements present in major amounts were aluminum, less than 0.10 weight percent (less than 0.033 atom percent)and iron, less than 0.15 weight percent (less than 0.023 atom percent).

A phase concentration technique was used to obtain the quantities of phases required for positive identification. X-ray powder patterns obtained from the residues provided detailed diffraction data of which those for the phase Al Fe Be4, are of particular interest. These data are compared with the findings of other research investigations.

Experimental procedures are discussed in detail and a summary of the test results in included.

A time saving method has been applied for the determination of retained austenite.

The method involved is based on the approach of energy dispersive X-ray diffraction analysis. With this approach, polychromatic radiation from the X-ray tube is used and diffraction maxima will occur at a fixed angle 2θ in as many wavelengths or energies as “d“ values are present.

Giessen and Gordon published the first application of this method to powder diffraction analysis in 1968 for the identification of crystal structures. As the determination of retained austenite is a quantitative type of analysis, based upon identification of the crystal structure, the new approach should also be applicable in principle.

With almost 100 samples of unknown austenite content, experiments have been carried out both with the conventional and the energy dispersive X-ray diffraction technique. The results obtained are closely comparable and the retained austenite values together with the errors are shown.

For these measurements, experiments have been carried out with the energy dispersive technique to determine the relation between the known concentration of retained austenite in a number of standards and the intensity correction factors (R). The results obtained from these experiments have shown good reproducibility of the intensity correction factors.

Using this technique, a five-fold reduction in analysis time is possible over the conventional technique with no reduction in accuracy.

A direct comparison x-ray method using crystal structure Information has been developed for rapid quantitative analysis of the oxide phases present in oxidized steels. This procedure is similar to that used for the analysis of retained austenite in hardened steels and requires no external or internal standards for accurate oxide determinations. Successful application of this technique has been achieved for a variety of steel oxide mixtures and excellent correlation with microstructural analysis results has been found. Accurate determinations have been made of the volume fractions of Wustite (FeO), Magnetite (Fe3O4) and Hematite (Fe2O3) for a wide range in steel carbon content and transformation treatments. Many oxide mixtures that differ over a wide range in the amounts of each of the three oxides phases have been measured successfully. The procedure is simple, the method direct and the analysis accurate.

Quantitative analysis by x-ray fluorescence and photoelectron and Auger electron analysis can be effectively extended through a precise knowledge of the total aad subshell photoionization cross sections. Light element and intermediate element analysis, as based upon K and L series fluorescence respectively, involve x-ray interactions in the low energy region, Optimized analysis for essentially all the elements by x-ray induced photoelectron and Auger electron spectroscopy involves both x-ray and electron interactions in the low energy region. Unfortunately, theory and measurement for interaction cross sections in this 100-1500 eV region are difficult, particularly for the heavier elements. Nevertheless, recent advances in experimental and computerized-theoretical techniques for the determination of low energy interaction coefficients do permit establishing appreciatly more complete tabulations of cross sections than are currently available in this energy region.

In this paper, the types of interaction cross section data that are needed for quantitative x-ray and electron analysis are defined. Such data that are available from experiment and from theory are reviewed and compared. Some newer techniques for the measurement of cross sections are discussed. And finally, new “state of the art” tables are presented for the mass absorption coefficients of all of the elements and of some special laboratory materials. These are tabulated specifically for twenty-six of the most commonly applied characteristic wavelengths in the 8-110 A region and are based upon the best currently available theoretical and experimental data.

Correction factors, termed α-factors, similar to those defined by LaChance and Traill have been generated by the addition of variable, known amounts of individual oxides, or other compounds, to a base cement sample and measuring the x-ray intensities of the elements of interest. The effects of all common constituents of cement on the determination of CaO, SiO2 and Al2O3 were found. Factors for rhodium and chromium primary radiation were determined and, in general, showed small but significant differences. The factors for rhodium at 50kV and 30kV were substantia11y identical. The correction factors were tested through the use of the NBS 1011- 1016 cements as reference standards to analyze the new proposed NBS cement series. The correction factors not only furnished improved calibration curves, but also allowed the determination of CaO, Al2O3 and SiO2 with an average deviation of less than 0.2% (absolute) from the provisional values furnished with the standard samples.

Specimen standards have been prepared for elemental quantitative analysis of atmospheric aerosols by means of x-ray spectrometry. Efforts were made so that the standards were similar to the samples actually collected in the atmosphere. Soluable salts were placed in a completely enclosed, self-contained bubbler system so as to ensure uniform rate of flow through the filters on which elements were collected. The standards were deposited for various time intervals in the bubbler system and then weighed on a microbalance. Aerodynamic flow ensured that the specimens were uniformly distributed throughout the filter. Very satisfactory calibration curves were constructed of intensity versus mass of element. For low concentrations of the soluable salts and relatively small amounts of total material collected straight line relationships can be obtained for calibration curves. These amounts are mostly well within those actually collected in the field. The problem of impurity levels in collection filters which can far exceed the levels detected in the aerosols has been given careful consideration in the selection of filters used. Absorption-enhancement effects appear to be unimportant at the low concentration levels delt with. Interference problems with lead, arsenic, and bromine are discussed and appropriate calibration curves shown. A set of atmospheric specimens are analysed for several elements placing atmospheric aerosol analysis on a quantitative basis.

Nonproportionality of X-ray fluorescence intensity vs. concentration of sulfur occurs in geologic materials prepared as ground samples. Samples containing sulfur as sulfate yield higher intensities than an equivalent amount of sulfur as the Sulfide. Although the intensity of free sulfur decreases markedly during a short exposure to X-rays in a vacuum, and sulfide intensities decrease over a much longer time period, sulfate intensities show no deterioration with time.

A fusion procedure utilizing a LiBO2-Ce(NH4)2(NO3)5 flux has been developed for geologic materials to oxidize all sulfur in a multiplicity of oxidation states to a single oxidation state, namely, the sulfate. The procedure is applicable over a wide range of concentrations (0.5 - 28 wt. % total sulfur) with a detection limit of 0.01%. A suite of chemically analyzed rocks containing both sulfides and Sulfates is used for standardization.

A number of procedures for fluorescent X-ray analysis have been introduced to accommodate the samples that are produced from research on the recovery of values from secondary metal sources. For some applications, standards are conveniently available such as those that can be purchased from the National Bureau of Standards. For other applications, secondary standards must be prepared and analyzed by independent methods. The sample preparation procedures vary considerably. For monitoring process efficiency, sample preparation is often kept at a minimum such as simply pouring loose powders into disposable cups. For the most accurate analyses, sample preparatton requires casting the alloys and finishing the surfaces. Matrix correction procedures are employed where concentrations of major constituents vary over wide ranges.