A comparison using Faraday cups with 1013 Ω amplifiers and a secondary electron multiplier to measure Os isotopes by negative thermal ionization mass spectrometry.

Abstract

According to the Johnson-Nyquist noise equation, the value of electron noise is proportional to the square root of the resistor value. This relationship gives a theoretical improvement of 100 in the signal/noise ratio by going from 1011 Ω to 1013 Ω amplifiers for Faraday detection in thermal ionization mass spectrometry (TIMS). We measured Os isotopes using static Faraday cups with 1013 Ω amplifiers in negative thermal ionization mass spectrometry (NTIMS) and compared the results with those obtained with 1011 Ω amplifiers and by peak-hopping on a single secondary electron multiplier (SEM). We analysed large loads of Os (1 μg) at a range of intensities of 187 OsO3 (0.02-10 mV) in addition to small loads of Os (5-500 pg) to compare the results of the three methods. Using 1013 Ω amplifiers, the long-term reproducibility determined from Merck Os was 187 Os/188 Os = 0.1211 ± 0.0086 and 0.120229 ± 0.000034 at 0.02 mV and 10 mV of 187 OsO3 intensities. Meanwhile, the analysed JMC Os loadings of 5 and 500 pg showed 187 Os/188 Os = 0.10669 ± 0.00036 and 0.106807 ± 0.000023. In comparison, the values measured by the SEM were 187 Os/188 Os = 0.10704 ± 0.00056 and 0.10690 ± 0.00013. All errors are in 2 standard deviation (SD). Both the accuracy and the precision determined using the 1013 Ω amplifiers and the SEM are identical when the Os amounts are within 10-50 pg. However, the former analysis time can be shortened by approximately two-thirds. The SEM measurement is still the most precise method for Os amounts <10 pg, but the analyses using 1013 Ω amplifiers suggest they are significantly better than the SEM for Os amounts >50 pg.