A novel method for the determination of platinum, palladium and rhodium in metal carrier catalyst for exhaust gas purification was established by flame atomic absorption spectrometrometry(FAAS). The sample decomposition method, coprecipitation condition, interference factor and elimination method were studied. The metal carrier catalyst sample for exhaust gas purification was treated by hydrochloric acidultrasonic wave, and filtrated. The insoluble substance was decomposed with sodium peroxide and acidized with hydrochloric acid, and then it was mixed with sample solution. 10 mg of tellurium oxide and 10 mL of 200 g/L SnCl2 solution were added to sample solution in the medium of 23 mol/L hydrochloric acid for the coprecipitation, enrichment and separation of palladium, platinum and rhodium from matrix elements such as Fe, Ni, Al, Cr and Na. After the coprecipitate was dissolved with aqua regia, the content of palladium, platinum and rhodium was determined by flame atomic absorption spectrometry. The detection limits of method for Pt, Pd and Rh are 4.72,1.13 and 1.06,respectively. The proposed method was applied to the determination of actual samples, and the results were consistent with those obtained by inductively coupled plasma atomic emission spectrometry. The relative standard deviations (RSD, n=11) for palladium, platinum and rhodium were 3.0%, 1.9% and 4.2%, respectively.

The condensation compound after phosphoric acid dehydration had strong acidity and coordination properties. After the samples were dissolved in phosphoric acid by heating at high temperature for 5 min, the titanium and iron was reduced by aluminum sheet and stannous chloride under acid conditions. Then, the content of titanium and iron in ilmenite was respectively  determined by potassium dichromate standard solution with neutral red and sodium diphenylamine sulfonate as indicator. The effect of sample dissolution condition, sample size, dissolution time and reducing acidity on determination was investigated to obtain the optimal experimental conditions. The determination results of titanium and iron in actual sample by this method were consistent with those obtained by classical ammonium ferric sulfatemercury salt method, and relative standard deviations (n=5) was  smaller  than 0.2 % and 0.5 %. This method has been applied into the determination of titanium and iron in ilmenite certified reference materials YSBC1970476 for five times. The results were consistent with the certified values, and relative standard deviations were 0.12 % and 0.62 %, respectively.