Is Grain Boundary Sliding the Dominant Deformation Mechanism in the Hydrous Upper Mantle? Experimental Constraints on the Lithosphere-Asthenosphere Boundary.

水是地球上最重要的化学成分，对气候、矿物和岩石性质、地幔对流和板块构造都有影响。最近对地震产生的弹性波在海洋和大陆地区传播的观测表明，这些地震波的速度发生了很大的变化。这种速度的变化发生在从相对坚硬的岩石圈板块向位于下面的相对流体的软流层过渡的时候，这需要上地幔强度的伴随变化。这些变化被认为部分是由于组成上地幔的岩石的水分含量的变化。然而，地震证据也表明，变形机制的改变必须对颗粒尺寸变得敏感。以前在这种晶粒度敏感或晶界滑动(GBS)区域中探索湿条件下变形的实验都产生了模棱两可的结果。这在一定程度上是由于最近在PI的实验室克服了实验困难。PI之前资助的NSF对干燥条件下富橄榄石岩中GBS的研究结果扩大了高应力岩石圈条件下GBS主导变形的区域，将低应力软流圈环境包括在内，这种扩张影响了对地震波的解释和对地球上地幔动态过程的模拟。必须解决岩石在潮湿条件下的力学行为的不确定性，以便将实验室结果外推到地球上水起关键作用的条件。本项目强调对实验室实验的变革性方法，旨在研究GBS在潮湿条件下橄榄石变形中的作用。为了获得稳定的微观结构和稳定的流量，研究人员在他们的变形实验中引入了三种重要的技术。首先，他们制作了被水饱和的样品，其颗粒尺寸大于稳定状态的颗粒尺寸。其次，他们将这些样品扭转变形为大应变，这样动态再结晶就会产生稳定的晶粒度。第三，扭力组件中加入了水源，以保持橄榄石中的饱和度。初步结果表明，在这些实验中达到了稳定状态。之所以提出这些实验，是因为最近的实验结果和野外观测表明，地幔中橄榄石的动态重结晶可能会充分减小颗粒尺寸，使GBS成为主要的变形机制。在GBS体制的变形实验中，即使是由于颗粒长大或水分变化引起的微小的颗粒尺寸变化，也可能导致地幔粘度与应力或颗粒大小以及温度的关系的测定出现重大误差。粘度对应力、颗粒大小、温度和水含量的依赖关系目前还不够清楚，无法将实验室实验结果外推到地球地幔中发生的地球动力学过程。因此，这种方法旨在克服这些困难，并为地震学家和建模人员提供必要的参数，以调查岩石圈-软流圈边界的来源和性质。

Water is the most important chemical constituent of our planet having impact on climate, mineral and rock properties, mantle convection, and plate tectonics. Recent observations of the propagation of elastic waves generated by earthquakes through both oceanic and continental regions have revealed large changes in the velocity of these seismic waves. This change in velocity occurs at the transition from the relatively rigid lithospheric plate to the relatively fluid asthenosphere that lies below and requires a concomitant change in the strength of the upper mantle. These changes are thought to be due in part to a change in the water content of the rocks that comprise the upper mantle. However, the seismic evidence also indicates that a change in the mechanism of deformation must become sensitive to grain-size. Previous experiments conducted to explore deformation under wet conditions in this grain-size sensitive or grain-boundary sliding (GBS) regime have produced ambiguous results. This is due in part to experimental difficulties that have recently been overcome in the PI's lab. The results of the PI's previously funded NSF research on GBS in olivine-rich rocks under dry conditions expanded the region where GBS dominates deformation from high-stress, lithospheric conditions to include low-stress, asthenospheric environments, an expansion that impacts the interpretation of seismic waves and modeling of dynamic processes in the upper mantle of Earth. Uncertainties about the mechanical behavior of rocks under wet conditions must be resolved to allow extrapolation of laboratory results to conditions in the Earth where water plays a vital role.The present project emphasizes a transformative approach to laboratory experiments designed to investigate the role of GBS in the deformation of olivine under wet conditions. To achieve steady-state microstructures and thus steady state flow rates, the investigators have introduced three important techniques to their deformation experiments. First, they fabricate samples saturated with water with grain sizes larger than the steady-state grain size. Second, they deform these samples in torsion to large strains, such that dynamic recrystallization produces a steady-state grain size. Third, a source of water is incorporated into the torsion assembly to maintain saturation levels in the olivine. Initial results demonstrate that steady state is attained in these experiments. These experiments were proposed because recent experimental results and field observations indicate that dynamic recrystallization of olivine in Earth's mantle might reduce the grain size sufficiently to make GBS the dominant deformation mechanism. Even small changes in grain size due to grain growth or changes in water content during a deformation experiment in the GBS regime may lead to significant errors in determinations of the dependence of mantle viscosity on stress or grain size as well as on temperature. The dependence of viscosity on stress, grain size, temperature, and water content are not currently known well enough to allow extrapolation of results obtained from laboratory experiments to geodynamical processes occurring in Earth's mantle. Consequently, this approach is designed to overcome these difficulties and provide the necessary parameters for seismologists and modelers to investigate the origin and nature of the lithosphere-asthenosphere boundary.