An interlaboratory study to test instrument performance of hydrogen dual-inlet isotope-ratio mass spectrometers

An interlaboratory comparison of forty isotope-ratio mass spectrometers of different ages from several vendors has been performed to test H-2/H-1 performance with hydrogen gases of three different isotopic compositions. The isotope-ratio results (unsufficiently corrected for H-3(+) contribution to the m/z = 3 collector, uncorrected for valve leakage in the change-over valves, etc.) expressed relative to one of these three gases covered a wide range of values: -630 parts per thousand to -79o parts per thousand for the second gas and -368 parts per thousand to -462 parts per thousand for the third gas. After normalizing the isotopic abundances of these test gases (linearly adjusting the delta values so that the gases with the lowest and highest H-2 content were identical for all laboratories), the standard deviation of the 40 measurements of the intermediate gas was a remarkably low 0.85 parts per thousand. It is concluded that the use of scaling factors is mandatory for providing accurate internationally comparable isotope-abundance values. Linear scaling for the isotope-ratio scales of gaseous hydrogen mass spectrometers is completely adequate.