Viscous flow and diffusion in highly depolymerized water-rich melts: improvement of experimental methods and implication for natural system

富水强解聚熔体的输运性质仍存在争议。到目前为止，还没有实验技术可以确定高温高压下高度解聚熔体的粘度。这些黏度数据对于理解含水伟晶岩和空纹岩等自然系统的结晶和喷发过程的动力学过程至关重要。落球法（斯托克斯定律）已成功地用于各种条件下粘度的测定。然而，它在低粘度熔体中的应用主要受到可实现的实验时间的限制。在项目的第一部分（前12个月），我们开发了一种新的装置，允许在内部加热的气体压力容器（IHPV）中快速加热和快速淬火样品，并建立了一种新的垂直定向冷密封压力容器（CSPV）系统，用于快速加热/快速淬火。此外，我们还进行了高粘度（微渗透）和低粘度范围（落球）的伟晶岩熔体的粘度测定。初步结果表明，富水伟晶岩熔体的黏度明显低于现有模型的预测值（可达2 ~ 3个log单位）。在项目的第二部分，我们计划使用快速加热和快速淬火装置来更准确地确定熔剂元素如F、Li（最终是B和P）在硅酸盐熔体中的单个影响。该装置还将用于确定晶状熔体中HFSE （Nb和Ta）的扩散率，这对于约束经济HFSE浓度形成的成因模型具有重要意义。最后，我们计划确定富水的phonolite的粘度，这将有助于模拟phonolite熔体的上升和脱气过程。

The transport properties of water-rich strongly depolymerized melts are still under debate. So far no experimental techniques are available to determine the viscosity of highly depolymerized melts at elevated temperatures and pressures. Such viscosity data are crucial to understand kinetic processes during crystallization and eruptive processes of natural systems such as hydrous pegmatites arid phonolites. The falling sphere method (Stokes law) has been successfully used for viscosity determination at various conditions. However, its application to low viscosity melts is mainly limited by the achievable experimental duration. In the first part of the project (first 12 months), we have developed a new set-up allowing rapid heat and rapid quench of samples in internally heated gas pressure vessels (IHPV) and set up a new system of vertically oriented cold seal pressure vessels (CSPV) for rapid heat/rapid quench. Additionally, we have conducted viscosity determinations of pegmatitic melts in the high (micropenetration) and low viscosity range (falling sphere). The preliminary results show that the viscosity of water-rich pegmatite melts is significantly lower (up to 2 - 3 log units) than predicted by available models. In the second part of the project we plan to use the rapid heat and rapid quench devices to determine more accurately the individual effects of fluxing elements such as F, Li (eventually B and P) in silicate melts. The setup will also be used to determine the diffusivity of HFSE (Nb and Ta) in pegmatitic melts, which is of importance to constrain the genetic models proposed for the formation of economic HFSE concentrations. Finally we plan lo delermine the viscosity of water-rich, phonolites which will be useful for modeling ascent and degassing of phonolite melts.