TY - JOUR

T1 - The effect of ƒo2 on the solubility, diffusion, and speciation of tin in haplogranitic melt at 850°C and 2 kbar

AU - Linnen, Robert L.

AU - Pichavant, Michel

AU - Holtz, François

AU - Burgess, Simon

N1 - Funding Information: Acknowledgmenrs-We are grateful to R. Helbig for supplying us with synthetic cassiterite crystals, for discussions with B. Scaillet, and to 0. Rouer for help with the electron microprobe analyses. This study was supported by a Natural Science and Engineering Research Council of Canada postdoctoral fellowship to R. L. L. and a grant to S. B. from the University of St. Andrews, and by the CNRS program “mCtallogCnie”. Reviews by D. Baker, P. Candela, and D. London improved the manuscript, but the content of this paper is solely the responsibility of the authors. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.

PY - 1995/4

Y1 - 1995/4

N2 - Diffusion profiles of tin were produced in hydrous silicate melts adjacent to cassiterite crystals; the method of Harrison and Watson (1983) was adapted to produce Sn diffusion profiles in hydrous silicate melts adjacent to cassiterite crystals at 2 kbar, 850°C, and various redox conditions, from which information on the solubility, diffusion, and speciation of Sn in silicate liquids can be obtained. The use of diffusion profiles and a hydrous, yet slightly H2O-undersaturated melt composition were chosen, in order to avoid the loss of Sn to the noble-element capsule walls. Such losses occurred at reduced conditions in the previous experimental studies on the solubility or partitioning of Sn in silicate liquids (±fluid phase), which probably interfered with the redox or SnO2 activity control of those experiments. The redox conditions investigated in this study were controlled by the intrinsic or an imposed f{hook}H2 in rapid-quench cold-seal and internally heated pressure vessels, and were measured by either the hydrogen sensor or Shaw membrane techniques. Cassiterite solubilities at 850°C and 2 kbar range from 28,000 ppm SnO2 at FMQ-0.84 to approximately 800 ppm at FMQ+3.12, in a haplogranitic melt with a normative (anhydrous) composition of 37.2% quartz, 28.3% orthoclase, 34.1 % albite, 0.4% corundum, and 5.6 wt% H2O. For redox conditions higher than FMQ+1.5, SnO2 solubility is independent of f{hook}O2, indicating that cassiterite dissolved into the melt largely as Sn4+. By contrast at more reduced conditions, log SnO2 concentration vs. log f{hook}o2 define a slope of approximately -0.5, implying that Sn is dominantly in the 2+ valence, if γSnmelt2+. is constant. The solubilities obtained at reduced conditions in this study are an order of magnitude higher than previously published data (at a comparable P-T, composition and f{hook}H2 of the autoclave); the low values in the previous work are attributed to the loss of Sn to the capsule walls. The diffusion of Sn is also apparently related to f{hook}o2, ranging from approximately 10-8 cm2/s at FMQ-0.84 to 10-9 cm2/s at FMQ+3.12. This is consistent with Sn2+ behaving as a network modifier, whereas the diffusion of Sn4+ is slower, similar to other high field strength elements. The strong f{hook}O2 dependence of Sn solubility can be used to explain some magmatic tin deposits and also indicates that tin might prove useful as a marker of redox changes of a variety of magmatic systems.

AB - Diffusion profiles of tin were produced in hydrous silicate melts adjacent to cassiterite crystals; the method of Harrison and Watson (1983) was adapted to produce Sn diffusion profiles in hydrous silicate melts adjacent to cassiterite crystals at 2 kbar, 850°C, and various redox conditions, from which information on the solubility, diffusion, and speciation of Sn in silicate liquids can be obtained. The use of diffusion profiles and a hydrous, yet slightly H2O-undersaturated melt composition were chosen, in order to avoid the loss of Sn to the noble-element capsule walls. Such losses occurred at reduced conditions in the previous experimental studies on the solubility or partitioning of Sn in silicate liquids (±fluid phase), which probably interfered with the redox or SnO2 activity control of those experiments. The redox conditions investigated in this study were controlled by the intrinsic or an imposed f{hook}H2 in rapid-quench cold-seal and internally heated pressure vessels, and were measured by either the hydrogen sensor or Shaw membrane techniques. Cassiterite solubilities at 850°C and 2 kbar range from 28,000 ppm SnO2 at FMQ-0.84 to approximately 800 ppm at FMQ+3.12, in a haplogranitic melt with a normative (anhydrous) composition of 37.2% quartz, 28.3% orthoclase, 34.1 % albite, 0.4% corundum, and 5.6 wt% H2O. For redox conditions higher than FMQ+1.5, SnO2 solubility is independent of f{hook}O2, indicating that cassiterite dissolved into the melt largely as Sn4+. By contrast at more reduced conditions, log SnO2 concentration vs. log f{hook}o2 define a slope of approximately -0.5, implying that Sn is dominantly in the 2+ valence, if γSnmelt2+. is constant. The solubilities obtained at reduced conditions in this study are an order of magnitude higher than previously published data (at a comparable P-T, composition and f{hook}H2 of the autoclave); the low values in the previous work are attributed to the loss of Sn to the capsule walls. The diffusion of Sn is also apparently related to f{hook}o2, ranging from approximately 10-8 cm2/s at FMQ-0.84 to 10-9 cm2/s at FMQ+3.12. This is consistent with Sn2+ behaving as a network modifier, whereas the diffusion of Sn4+ is slower, similar to other high field strength elements. The strong f{hook}O2 dependence of Sn solubility can be used to explain some magmatic tin deposits and also indicates that tin might prove useful as a marker of redox changes of a variety of magmatic systems.

UR - http://www.scopus.com/inward/record.url?scp=0028812969&partnerID=8YFLogxK

U2 - 10.1016/0016-7037(95)00064-7

DO - 10.1016/0016-7037(95)00064-7

M3 - Article

AN - SCOPUS:0028812969

VL - 59

SP - 1579

EP - 1588

JO - Geochimica et Cosmochimica Acta

JF - Geochimica et Cosmochimica Acta

SN - 0016-7037

IS - 8

ER -