An ICP-MS instrument with a quadrupole filter is hampered by its limited resolution, in particular polyatomic ions are a major source of interference, With a double focusing magnetic sector mass analyser, operated at high resolution, most of these interferences are eliminated, In addition, these mass analysers are characterized by a low instrumental background, yielding superior detection limits at low resolution. In this work a new type of double focusing ICP-MS instrument, the 'Element' from Finnigan MAT (Bremen, Germany), was tested, The instrument can be used with low resolution (m/Delta m=300, 10% valley definition) and at high resolution settings (3000 and 7500), At low resolution, a linear response versus mass curve was observed [response of 180 x 10(6) counts s(-1) per pg ml(-1) (ppm) for U and 6 x 10(6) counts s(-1) for Ee]. The sensitivity decreases by roughly a factor of 14 when going from resolution 300 to resolution 3000 and by a factor of 10 from 3000 to 7500, An instrumental detection limit at low resolution of about 8 fg ml(-1) (8 ppq) was measured at m/z 240, The detection limit for most elements is not determined by the instrument but by the blank level and laboratory procedures will need to be improved in order to make optimal use of the detection power of the instrument, The stability and the relative abundance of oxide and doubly charged ions are comparable to the values found in quadrupole ICP-MS, Matrix effects were compared for measurements with the 'Element' and with a Perkin-Elmer Elan 5000 quadrupole instrument of In-115 in the presence of 500 mu g ml(-1) of Cs and of 2.5% of ethanol. Though in the 'Element' ions are accelerated over 8 kV immediately after the skimmer, matrix effects were found to be similar to those observed for the 'Elan 5000', The instrument was used to determine Pt in an environmental material (grass)

No values of U in serum have been reported up till now.Paper 7/01652B Measurements were made at low resolution since no spectral

Results of a systematic study concerning non-spectral interferences observed with a commercially available high resolution ICP-mass spectrometer are reported and compared to observations made with a quadrupole-based instrument. In general, matrix effects were observed to be to a large extent comparable for both instruments used. In all cases, the matrix-induced signal suppression or enhancement was seen to depend in a regular way on the mass number of the nuclides monitored. In most cases, the ionization potential of the nuclides has little or no influence on the extent of suppression or enhancement. For As, Se and Te, the introduction of 2.5% ethanol, 0.5 mol/l H(2)SO(4), or to a lesser extent 0.5 mol/l H(3)PO(4), leads to an exceptional increase in the signal intensity for both instruments. Registration of signal behaviour plots (signal intensity as a function of the nebulizer gas flow rate) in different matrices revealed that both the height of the plot and the optimum nebulizer gas flow rate are a function of the matrix composition. Finally, no indication was found that the acceleration of the extracted ions over 8000 V with the high resolution instrument would lead to an alleviation of space charge effects when compared to a quadrupole-based ICP-mass spectrometer.

No abstract

In the framework of an international certification campaign, sector-field inductively coupled plasma mass spectrometry (sector-field ICP-MS) was used for the accurate determination of the rare earth elements in five candidate reference materials: aquatic plant, calcareous soil, mussel tissue, river sediment, and tuna muscle. All samples were taken into solution by use of microwave-assisted or mixed microwave-assisted / open beaker acid digestion. Subsequently, the samples were appropriately diluted and subjected to ICP-MS analysis. Except for Sc, all the elements involved were determined at low mass resolution (R = 300). For Sc, application of a higher resolution setting (R = 3,000) was required to separate the analyte signal from those of several molecular ions which gave rise to spectral overlap at low mass resolution. Some of the heavier REE can also suffer from spectral overlap attributed to the occurrence of oxide ions (MO+) of the lighter REE and Ba. This spectral overlap could be successfully overcome by mathematical correction. Matrix effects were overcome by use of two carefully selected internal standards, such that external calibration could be used. On each occasion, a geological reference material was analyzed as a quality-control sample and the reliability of all results obtained was additionally checked by means of chondrite normalization. For tuna muscle the content of all REE was below the limit of detection. For calcareous soil and river sediment, low to sub microg g(-1) values were observed, whereas the REE content of aquatic plant and mussel tissue was considerably lower (low to sub ng g(-1)). Overall, the results obtained were in excellent agreement with the average values, calculated on the basis of all "accepted" values, obtained in different laboratories using different techniques.