Development of perfluorocarbon (PFC) primary standards for monitoring of emissions from aluminum production

Citation
G. Rhoderick et al., Development of perfluorocarbon (PFC) primary standards for monitoring of emissions from aluminum production, FRESEN J AN, 370(7), 2001, pp. 828-833
Citations number
14
Categorie Soggetti
Spectroscopy /Instrumentation/Analytical Sciences
Journal title
FRESENIUS JOURNAL OF ANALYTICAL CHEMISTRY
ISSN journal
09370633 → ACNP
Volume
370
Issue
7
Year of publication
2001
Pages
828 - 833
Database
ISI
SICI code
0937-0633(200108)370:7<828:DOP(PS>2.0.ZU;2-O
Abstract
An EPA Voluntary Aluminum Industrial Partnership (VAIP) program has been fo rmed to help US primary producers focus on reducing the emissions of two pe rfluorocarbons (PFCs), tetrafluoromethane (CF4) and hexafluoroethane (C2F6) , during the production of aluminum. To ensure comparability of measurement s over space and time, traceability to national sources was desirable. Henc e, the EPA approached the NIST to develop a suite of primary standards to c over a mole fraction (concentration) range of 0.1 to 1400 mu mol mol(-1) fo r CF4 and 0.01 to 150 mu mol mol(-1) of C2F6. A total of eight gravimetric PFC gas standards were prepared with relative expanded uncertainties of les s than or equal to 0.52% (approximate to 95% confidence level). These prima ry standards were ultimately used to assign values to a series of secondary gas standards at three mole-fraction levels with relative expanded uncerta inties ranging from +/- 0.7% to 5.3% (approximate to 95% confidence level). This series of secondary standards was used within the aluminum industry t o calibrate instruments used to make emission measurements. Assignment of v alues to the secondary standards was performed by use of gas chromatography with flame-ionization detection (GC-FID) and Fourier transform infrared sp ectrometry (FTIR). Real time pot-line and stack samples from a local alumin um plant were also obtained and sub-samples sent to each participating faci lity for analysis. The data generated from each facility were sent to NIST for analysis. The maximum difference between the KIST and individual facili ties' values for the same sub-sample was +/- 26%.