ON THE OCCURRENCE OF APPARENT NON-HENRY LAW BEHAVIOR IN EXPERIMENTAL PARTITIONING STUDIES

被引:49
作者
BEATTIE, P
机构
[1] Department of Earth Sciences, University of Cambridge, Cambridge
关键词
D O I
10.1016/0016-7037(93)90467-B
中图分类号
P3 [地球物理学]; P59 [地球化学];
学科分类号
0708 ; 070902 ;
摘要
The strongest evidence for the occurrence of non-Henry's Law behaviour in magmatic systems is the apparent dependence of the garnet-melt partition coefficients for the REEs on REE concentration. Several of the garnet-melt pairs which showed deviations from Henry's Law when analysed using the optical autoradiography technique were reanalysed using secondary ion mass spectrometry. The SIMS results for these experiments showed that the partition coefficient is independent of concentration. It is concluded that the optical autoradiography technique can produce a spurious doubling of REE partition coefficients at low concentrations. Similar doubling of REE partition coefficients at low concentration have been observed with the optical autoradiography technique in plagioclase-melt systems. These deviations may also be an artefact of the autoradiographic technique. Non-Henry's Law behaviour has also been reported for Ni partitioning in olivine-melt systems. The data which deviate from Henry's Law are shown to be anomalous when compared with the data from all non-autoradiographic techniques, and from studies of phenocryst-matrix pairs. It is shown that for these experiments the partition coefficient is not dependent on the NiO concentration in either phase. It is concluded that these apparent deviations from Henry's Law are due to problems associated with the experimental procedure used and that the experiments do not record equilibrium partitioning. There are therefore good reasons for doubting the analyses of the autoradiographic studies which have found deviations from Henry's Law at low concentration. In the absence of non-autoradiographic data to the contrary, it is concluded that Henry's Law may be satisfied over a wide range of concentrations. There are therefore no data to suggest that partition coefficients measured in percent-level doping experiments cannot be used to model magmatic processes where the components of interest may only be present at trace concentrations.
引用
收藏
页码:47 / 55
页数:9
相关论文
共 55 条
[31]  
LEEMAN WP, 1974, THESIS U OREGON
[32]  
Lindstrom D.J., 1976, THESIS U OREGON US
[33]  
LINDSTROM DJ, 1984, THESIS U P M CURIE
[34]   EXPERIMENTAL-DETERMINATION OF THE CHROMIUM-ALUMINUM MIXING PARAMETER IN GARNET [J].
MATTIOLI, GS ;
BISHOP, FC .
GEOCHIMICA ET COSMOCHIMICA ACTA, 1984, 48 (06) :1367-1371
[36]  
MYSEN BO, 1979, AM MINERAL, V64, P1107
[37]   PRESSURE-DEPENDENCE OF NICKEL PARTITIONING BETWEEN FORSTERITE AND ALUMINOUS SILICATE MELTS [J].
MYSEN, BO ;
KUSHIRO, I .
EARTH AND PLANETARY SCIENCE LETTERS, 1979, 42 (03) :383-388
[38]   TRACE-ELEMENT PARTITIONING DETERMINED BY BETA TRACK MAPPING - EXPERIMENTAL STUDY USING CARBON AND SAMARIUM AS EXAMPLES [J].
MYSEN, BO ;
SEITZ, MG .
JOURNAL OF GEOPHYSICAL RESEARCH, 1975, 80 (17) :2627-2635
[39]   CARBON-DIOXIDE IN SILICATE MELTS AND CRYSTALS .1. SOLUBILITY MEASUREMENTS [J].
MYSEN, BO ;
EGGLER, DH ;
SEITZ, MG ;
HOLLOWAY, JR .
AMERICAN JOURNAL OF SCIENCE, 1976, 276 (04) :455-479