One of common problems in multi-anvil experiments is that temperatures are heterogeneously distributed in sample capsules. As a result, it is extremely difficult to obtain samples in which melt and solid are homogeneously distributed due to high mobility of melt. Additionally, heterogeneous temperature distribution also cause significant errors in temperature estimation.
The prototype of multi-anvil heater is a straight cylinder (Fig. 1a), which produce a steep temperature gradient especially in the axial direction due to the limited length. Therefore, tapered or stepped heaters, in which less heat is generated in the central part of the heater, were developed, but temperature gradients in the radial direction still remain (Fig. 1b), because of the axial heat flow at both end of the furnace. In order to eliminate this axial heat flow, two heater rods are placed in our design (Fig. 1c). This furnace is referred to as the zero-temperature-gradient furnace. As shown in Fig. 2, this furnace prevents the melt-solid segregation.
Fig. 1. A schematic drawing of heat flow and temperature distribution within the furnace of multianvil cell assembly. (a) Straight furnace system, (b) stepped furnace system, and (c) new designed furnace with minimized temperature gradient.
Fig. 2. An example of homogeneous melt-solid distribution in a 10-mm pressure medium. A back-scattered electron image. Starting material: 99 wt.% of olivine + 1% of basalt. Conditions: 12
GPa, 2300 K, 10 min. Although the initial grain sizes were 2-5 μm, the olivine grains grew to 40-100 μm during heating. Olivine grains are darker.
Zarei, A., Li, Y., Fei, H.Z., Katsura, T., 2018. A nearly zero temperature gradient furnace system for high pressure multi-anvil experiments. High Pressure Res. 38, 243-249.