Nieuwenkamp, G; Veen, A; Wessel, RM; Qiao, H; Oh, SH; Kim, BM; Kim, KS; Castorena, AP; Nambo, CR; Delgado, JK; Caballero, VMS; Murillo, FR; Salas, M; Dias, F; Baptista, G; Konopelko, LA; Kustikov, YA; Pankratov, VV; Selyukov, DN; Balandovich, VS; Vishnyakov, IM; Pavlov, MV; Maltsev, MA; Botha, A; Valkova, M; Stovcik, V; Musil, S; Milton, MJT; Uprichard, IJ; Vargha, GM; Guenther, F; Gameson, L; Cunha, Vd
Industrial stack gas emission measurements are important for process control, control of air pollution, and for implementing legislation regarding the carbon dioxide emission rights. Measurements are typically performed using a range of process analysers for carbon monoxide (CO), carbon dioxide (CO2), nitrogen oxides (NOx), sulphur dioxide (SO2) and miscellaneous hydrocarbons. The calibration of these analysers is often performed using a series of binary mixtures of each component in nitrogen. For reasons of efficiency as well as a better match with true stack gas, the use of multi component mixtures for this purpose would be preferred.
The data from gravimetric preparation have been used as the basis for establishing the key comparison reference value.
Most of the participating NMIs mastered the challenges set in this key comparison well. IPQ reported results for CO, CO2 and SO2 only. INMETRO took part in this comparison as a pilot study.
The results for carbon monoxide showed serious deviations for INMETRO (-26.5% relative), VNIIM (+12.7% relative), and IPQ (-3.9% relative), with none of the cases demonstrating equivalence with the KCRV. Two other results inconsistent with the KCRV were reported by NMISA and KRISS.
For carbon dioxide, all laboratories except IPQ reported results consistent with the KCRV.
For propane, INMETRO reported a seriously deviating result (-6.0% relative). NMISA and VNIIM also reported deviating results (in absolute sense more than 1% relative). None of these three results were consistent with the KCRV.
Three results for nitrogen monoxide were not consistent with the KCRV: KRISS, VNIIM and NMISA. With the exception of CENAM, none of the results differed more than 1% relative from the KCRV.
For sulphur dioxide, two results were not consistent with the KCRV: NIM and VNIIM. Only one result deviated by more than 1% relative from the reference value (NIM).