Jump to main content
Jump to site search

Issue 5, 1995
Previous Article Next Article

Determination of hydrogen in silicates by secondary ion mass spectrometry


A quantitative procedure for the determination of hydrogen in silicates was developed for incorporation in the general procedure used with the Cameca IMS 4f ion microprobe for light and trace elements. The procedure employs a static 16O primary beam and the ‘energy filtering’ technique. Moreover, it allows the analysis of samples placed on separate mounts with respect to standards and, in particular, of mineral grains directly in thin sections of rock. Factors influencing the hydrogen background were studied in order to ascertain the appropriate analytical conditions and to develop a specific procedure for the determination of low hydrogen contents. A low and reproducible background was obtained with simple heating and de-gassing of the sample. The background value was equivalent to 0.007–0.015% H2O and its fluctuation over a 1 week working session indicated that the detection limit was ≈0.015% H2O. The results obtained confirm that the use of high-energy ions reduces the influence of the matrix composition on the ionization of hydrogen (relative to that of Si); effects related to the structure of the sample appear to be negligible. Residual matrix composition effects are related, to a first approximation, to the Si content of the sample. As such effects could not be completely removed, two separate ion yields were used for samples with SiO2 contents in the range 27–51% and for rhyolites (≈72% SiO2). The reproducibility of the measurements was typically ≈3% and the accuracy for the former group of samples was 15% for H2O contents >0.1% and 30% for H2O contents of ≈0.05%.

Back to tab navigation

Article type: Paper
DOI: 10.1039/AN9952001309
Citation: Analyst, 1995,120, 1309-1313

  •   Request permissions

    Determination of hydrogen in silicates by secondary ion mass spectrometry

    L. Ottolini, P. Bottazzi, A. Zanetti and R. Vannucci, Analyst, 1995, 120, 1309
    DOI: 10.1039/AN9952001309

Search articles by author