Volume 107, Number 6, November–December 2002Journal of Research of the National Institute of Standards and Technology[J. Res. Natl. Inst. Stand. Technol. 107, 687–691 (2002)]NIST Standards for Microanalysis and theCertification ProcessVolume 107 Number 6 November–December 2002R. B. MarinenkoNational Institute of Standards andTechnology,Gaithersburg, MD 10899-8371The National Institute of Standards andTechnology (NIST) has been involved inthe development of standards for microanal-ysis since the middle of the 1960s. Certi-fication of “traceable” standards that can besold to other laboratories is time-consum-ing and costly, especially when the extentof microheterogeneity within each speci-men becomes part of the uncertainty as-signed to the certified values. The pro-cess of certification of microanalysisstandards and the improvements that havefacilitated the process with the developmentof automation and computerization arereviewed.Key words: certification; electron mi-croprobe; microanalysis standards; micro-homogeneity; microheterogeneity; standardreference materials for microanalysis.Accepted: August 22, 2002Available online: http://www.nist.gov/jres1. IntroductionWith the availability of the first commercially pro-duced electron microprobe instruments in the early1960s, x-ray microanalysis became a significant analyt-ical technique in the Analytical Chemistry Division atNIST (then called the National Bureau of Standards).The person responsible for the development of electronprobe microanalysis at NBS was Kurt. F. J. Heinrichunder whose leadership the Microanalysis Section madenumerous contributions to improvements in instrumen-tation, to the determination of fundamental constants, tothe development of matrix correction procedures, and tothe development of standard reference materials for mi-croanalysis. NBS had been involved in the developmentof metrological and analytical standards from its begin-ning, but microanalysis standards presented the addi-tional requirement of determining the extent of hetero-geneity of a research material on the micrometer scale.Only materials that exhibited minimum microhetero-geneity could be certified as NBS SRMs for microanal-ysis.2. Early SRMsDuring the mid-1960s, SRMs being sold as bulk stan-dards were also evaluated for use as microanalysis stan-dards. Those that were found sufficiently homogeneousto be used as microanalysis standards were CartridgeBrass (SRM 478) and Low-Alloy Steels (SRMs 461 and463) [1]. In the early 1970s several binary and ternaryalloys were issued as SRMs for microanalysis—aW-20%Mo Alloy (SRM 480) [2], Fe-3Si (SRM 483)[3], the Au-Ag alloys (SRM 481) and Cu-Au Alloys(SRM 482) [4], the Fe-Cr-Ni alloy (SRM 479 and 479a)[5,6], and a group of four different steels, (SRMs 661–664). These materials were chosen in part because theywere useful standards for quantitative microanalysis andbecause they were useful in the determination of basicx-ray parameters. In addition, these materials could bemade with little heterogeneity both in the bulk material(from specimen to specimen) and on the micrometerscale (within each specimen). These materials were not687Volume 107, Number 6, November–December 2002Journal of Research of the National Institute of Standards and Technologyreadily available commercially and if so, they certainlywould not have been analyzed on the micrometer scaleor certified at that level for microhomogeneity. Oftenspecial preparation procedures were required, such asrepeated annealing, to achieve the desired level of ho-mogeneity in metal alloys. This could be done withsmall batches of materials (like a few hundred grams)from which numerous microanalysis standards could beobtained.Table 1 is a list of NBS/NIST SRMs for microanalysisthat were certified between 1965 and the present. Manyare no longer in stock (gray background), although forsome (darker gray background) there is more materialfor a reissue if needed, but not without a considerableamount of work. Those with the complete white back-ground are still in stock.3. The Certification Process and MoreRecent SRMsThe question often asked is, “Why doesn’t NISTprovide more microanalysis standards?” There are sev-eral reasons. Firstly, most pure elements and many stoi-chiometric compounds are available commercially, andmany naturally occurring minerals are available. Thesecan be easily purchased and subsequently evaluated bythe user for microheterogeneity. In addition, there are afew commercial suppliers who purchase these commer-cially available materials, mount, polish, and evaluatethem for resale as prepared microanalysis standards.NIST does not compete with such providers. Secondly,the fabrication and evaluation of research materials forcertification as microanalysis standards is expensive andtime-consuming. Details of the process will be de-scribed later. For these reasons, NBS/NIST scientistshave concentrated their efforts on the development ofmore complex materials that were not available com-mercially and that might be more useful in quantitativeelectron probe microanalysis (EPMA), i.e., such asthose that could be used in testing matrix correctionprocedures or in determining basic parameters.During the latter part of the 1970s, glasses becamepopular as standards, throughout the microanalysis com-munity. Because they are vitreous solids, many types ofglasses can be made homogeneous on the micrometerscale. In addition, trace to minor amounts of elementscan be added to glasses dur ing the manufacturing pro-cess without changing the microhomogeneity. This factprovides the possibility of preparing standards withcomplex compositions. There are several limitations,though, with the use of glasses as standards. Not alloxides or phosphates or combination of oxides or phos-phates readily form glasses, therefore limiting the num-Table 1. NIST standard reference materials for microanalysisSRM no. Name Form Nominal composition(% Mass fract.)461 & 463 Low Alloy Steel Rods, ≅ 6 mm dia. Fe with plus 25 other elements at or⫻ 10 cm long near trace level concentrations470 Mineral Glasses for Slices K-411, MgO,SiO2,CaO,FeOMicroanalysis (2 ⫻ 2 ⫻ 12) mm3K-412, MgO,Al2O3SiO2,CaO,FeO478 Cartridge Brass Cube and Cylinder Cu-73, Zn-27479a Fe-Cr-Ni Alloy Wafer Fe-71,Cr-18,Ni-11480 Tungsten Wafer W-78,Mo-2220 % Molybdenum481 Gold-Silver Alloys Six wires Au 100;80;60;40;20;0Ag 0;20;40;60;80;100482 Gold-Copper Alloys Six wires Au 100;80;60;40;20;0Cu
View Full Document
Unlocking...