EFFECTS OF PHOSPHORUS ON PARTIAL MELTING OF A MODEL MARTIAN MANTLE: IMPLICATIONS FOR GENERATION OF MARTIAN IGNEOUS CRUST
To constrain the effects of phosphorus on martian mantle melting phase relations and melt compositions, we performed two sets of partial melting experiments using a piston-cylinder at 2 GPa on a martian primitive mantle composition (4), one containing 0 wt.% P2O5 (LF0) and another containing 0.5 wt.% P2O5 (LF0.5). All experiments produced an assemblage of ol + opx ± cpx ± sp ± melt depending on the melting degree. Comparing our experimental samples with those obtained under the same conditions using the same starting composition with a bulk P2O5 content of 0.2 wt.% (LF0.2; 4), an enhancement of residual orthopyroxene over olivine is evidenced in LF0.2 and LF0.5, with a depletion of residual olivine up to twice as much in LF0.5 compared to LF0. This leads to compositional variations in the melt including a depletion of SiO2 content up to 10 wt.% in LF0.2 compared to LF0 at 10 wt.% melting. Our experiments demonstrate that the mantle P2O5 is sufficient to dramatically modify the petrogenesis of rocks produced by partial melting. We combined the difference in major element melt compositions (e.g., SiO2, FeO*, CaO/Al2O3) caused by variation in starting P contents with the previously constrained effects of pressure (0.5-5 GPa) on martian mantle-derived melt compositions. Fractionation-corrected shergottites and Gusev basalts thought to be representative of a primary melt mainly correspond to melt compositions expected when the starting P2O5 is 0.2-0.5 wt.%, suggesting that the martian mantle composition with P2O5 of 0.16-0.17 wt.% (3-4) might not be high enough to produce surface compositions of Mars.
(1) Kushiro.(1996) Earth Process:Reading the Isotopic Code, 109-122. (2) Toplis et al.(1994) GCA,58,2,797-810. (3) Dreibus and Wänke.(1985) Meteoritics, 20,267-381. (4) Lodders & Fegley.(1997) Icarus, 126(2), 373–394. (5) Ding et al.(2020) JGR:Planets, 125(4).