T. L. Fries H.N. Elsheimer 1990 A method is described for the determination of total tin in silicate rocks utilizing a graphite furnace atomic absorption spectrometer with a stabilized-temperature platform furnace and Zeeman-effect background correction. The sample is decomposed by lithium metaborate fusion (3 + 1) in graphite crucibles with the melt being dissolved in 7.5% hydrochloric acid. Tin extractions (4 + 1 or 8 + 1) are executed on portions of the acid solutions using a 4% solution of tricotylphosphine oxide in methyl isobutyl ketone (MIBK). Ascorbic acid is added as a reducing agent prior to extraction. A solution of diammonium hydrogenphosphate and magnesium nitrate is used as a matrix modifier in the graphite furnace determination. The limit of detection is > 10 pg, equivalent to > 1 ??g l-1 of tin in the MIBK solution or 0.2-0.3 ??g g-61 in the rock. The concentration range is linear between 2.5 and 500 ??g l-1 tin in solution. The precision, measured as relative standard deviation, is < 20% at the 2.5 ??g l-1 level and < 7% at the 10-30 ??g l-1 level of tin. Excellent agreement with recommended literature values was found when the method was applied to the international silicate rock standards BCR-1, PCC-1, GSP-1, AGV-1, STM-1, JGb-1 and Mica-Fe. Application was made to the determination of tin in geological core samples with total tin concentrations of the order of 1 ??g g-1 or less. application/pdf 10.1016/S0003-2670(00)83845-3 en Determination of total tin in silicate rocks by graphite furnace atomic absorption spectrometry article