Journal of Conference Abstracts

Computational Modelling of Al/Si Ordering in Feldspar

Eva Myers (erm1001@cam.ac.uk)

Cavendish Laboratory, Madingley Road, Cambridge CB3 0HE, UK.

The behaviour of the feldspar structure at different temperatures T and concentrations x of Al was investigated using a Monte Carlo simulation of the nearest-neighbour Ising model. This is useful because the Al/Si order-disorder phase transition can only be observed experimentally for a narrow range of temperature, but the simulation can be observed over a wider range. The simulation worked by swapping randomly chosen pairs of atoms so as to keep the concentration constant.The mean values of the order parameter Q and the enthalpy H were plotted over a grid of values of x and T, with 0.25 ¾ x ¾ 0.5 and 0 ¾ T ¾ 3/4 J.

It was found that the order-disorder transition temperature T_c declined rapidly as x was reduced from 0.5. For concentrations less than x_c = 0.31, the system became disordered even at absolute zero. This result is very different from the prediction of the Bragg-Williams model that T_c (alpha) 4x(1-x).

The plot of T_c (x) was compared with experimental results for T_c (x) to estimate a value for J. The estimated value was rather low at 0.24 eV, indicating that the feldspar structure can easily accommodate the local strain caused by the size difference between Al and Si tetrahedra, leading to a low strain contribution to J . The plots of Q and H were compared with experiment in a similar way.

A Crustal Magma Chamber Beneath the Slow-Spreading Mid-Atlantic Ridge

D. A. Navin¹, L. M. MacGregor², S. Constable³, C. Pierce¹, A. White⁴, G. Heinson⁴, M. A. Inglis¹ & M. Sinha² (sinha@esc.cam.ac.uk)

¹ Dept. Geological Sciences, University of Durham.

² Dept. of Earth Sciences, University of Cambridge.

³ Scripps Institute of Oceanography, University of California, San Diego.

⁴ School of Earth Sciences, Flinders University of South Australia.

In 1993 we carried out a major, integrated geophysical experiment on an "axial volcanic ridge" (AVR) segment of the Mid-Atlantic Ridge, at 57^o 45' N. The experiment included three major components: wide angle and reflection seismic profiles along and across the ridge; controlled source electromagnetic sounding profiles along and across the axis; and magneto-telluric sounding. The experiment was carefully targeted on a segment of the ridge that shows clear signs of very recent volcanic activity at the sea floor.

Both the seismic and electromagnetic studies provide clear evidence for an anomalous body at a depth of approximately 2 km beneath the axis of the AVR. The anomalous region is characterised by very low seismic P-wave velocity and very low electrical resistivity. Its top surface coincides with a clear seismic reflector. Taken together the geophysical data indicate clear evidence for a substantial crustal melt body beneath the axis.

Much debate has been expended in recent years as to the mechanism of crustal construction at slow spreading ridges, and in particular the question of whether or not significant crustal magma chambers can exist at low spreading rates. We shall examine the implications of our new evidence, and argue that the balance of evidence is now in favour of significant but short-lived crustal magma bodies, and of magmatic cycles affecting individual ridge segments and having periods in excess of 10,000 years. We estimate that the melt body at 57^o 45' N contains sufficient magma to create new crust equivalent to between 10,000 and 20,000 years' worth of spreading; but that its lifetime as a molten body must be very much shorter than this.