Molten metal capillary reactor for the high-temperature pyrolysis of methane

Highlights

• Improvements on gas separation.

• Utilization of low temperature alloys.

• Continuous and stable operation of a molten metal experimental set-up.

• Cooling concept for room temperature operation of the peripherals.

• Hydrodynamic studies on new system GaInSn with N₂.

Abstract

In this work, the further development of the molten metal capillary reactor in slug-flow regime is presented. The preliminary results from the high-temperature pyrolysis of methane at 1300 °C and 9 Nml/min are presented with a calculated conversion of 80%, and, a mean residence time of 1.36 s. Due to carbon deposition, difficult gas separation and unstable slug-flow, it was deemed necessary to redesign the system. For that, several alloys were tested looking for improved wettability and more favorable hydrodynamics. The modified experimental set-up is described, which led to improvements in gas separation, but not enough stability in the slug-flow. Finally, the current experimental set-up is introduced. There, a characterization of the hydrodynamics is performed using a low temperature alloy of gallium, indium and tin, GaInSn, and, a stable regular slug-flow is established for various gas and liquid flows. The presence of a film in the slug-flow remains subject to question and the conclusions on the direction of the project are drawn towards an alternative reactor system or further hydrodynamic studies.

Introduction

Contrary to popular belief, fossil energy consumption is not limited by its availability but rather is constrained by environmental considerations. Over the past decades, the world's proven natural gas reserves have been increasing steadily, from 2592 trillion cubic feet in 1980–6972 trillion cubic feet in 2014 [1], and if non-conventional sources like methane hydrates or shale gas are included, the availability could be greater. Many scenarios for the future production of energy or hydrogen without CO₂-emissions have been investigated [2], [3]. The consensus seems to be directed towards renewable energy; however, a full transition to renewable energy is not foreseeable in the near future [4]. It is thus necessary to search for bridging technologies. Among these, hydrogen presents itself as a very attractive alternative due to the possibility of using existing natural gas infrastructure as well as for its integration with fuel cells and other clean energy technologies.

Hydrogen can be generated with low to zero CO₂-emissions by pyrolysing hydrocarbons at high temperatures. The pyrolysis of methane can be described by the reaction:

$$C{H}_4\rightleftarrows C\left(s\right)+2H_2\Delta H_{R,1000\mathrm{°}C}=91.52kJmo{l}^{-1}\Delta G_{R,1000\mathrm{°}C}=-49.28kJmo{l}^{-1}$$

This reaction is endothermic and becomes thermodynamically spontaneous at 820 K and kinetically feasible at temperatures around 1073 K. The reaction can be catalyzed for lower temperature operation, but the carbon deposition on the catalyst and the limited equilibrium conversions leading to a separation problem and stream recycles are still issues that have to be addressed. High temperature pyrolysis remains an alternative, but does not come without challenges: carbon deposition is still a major hurdle [5], which can lead to the blockage of the reactor. In addition, the endothermicity of the reaction necessitates the introduction of heat at higher temperatures, which is usually challenging due to the sources of energy at such temperatures being few and far between.

The use of liquid media, such as molten metals, has been used to circumvent both these problems. Not only does the liquid medium act as a barrier preventing the carbon deposition at the heat transfer surface, but it also serves as a heat transfer medium due to its excellent transport properties. In 1931, Tyrer [6] initially patented a molten iron based concept for decomposing methane into its elements. Furthermore, the use of molten metal, tin for his case, bubble columns was first performed by Steinberg [7]. A modified version was explored by Serban et al. [8] exhibiting a relatively high conversion (57%) for a low temperature operation (750 °C). Nonetheless, in this publication, carbon deposition occurred at both the reactor surface and in the sparger used to feed the methane. Other molten metal concepts include the works of Martynov et al. [9] and Gulevitch et al. [10] using PbBi as the molten medium. Paxman et al. [11] presented some initial theoretical and experimental investigations on a solar-molten tin reactor, although the experiments with the molten media itself have not been published. Currently, a molten tin bubble column reactor is also being investigated by the group of Prof. Abanades in the IASS in Postdam in cooperation with the Karlsruhe Institute of Technology amongst others [12], [13], [14]. Their first experimental results [12] showed a hydrogen yield of 30% at 1000 °C with little carbon deposition on the reactor surface. Following experiments [13] with the inclusion of a fixed bed inside the bubble column, and operating with temperatures up to 1175 °C, hydrogen yields up to 78% were found. However, the lack of control over the residence times in bubble columns leads inherently to a decrease in conversion. To circumvent this issue, the use of a capillary reactor using molten metals in slug-flow regime was proposed as a solution at the TU Dortmund [15], [16] since the slug-flow regime has a residence time closer to that of plug-flow [17].

The proposed system in Ref. [15] is presented in Fig. 1. The operation mode was that of an alternating pressure reservoir which generates a quasi-continuous flow. Preliminary studies were carried out for a system of nitrogen water, and the initial experiment with methane reported an average conversion of 32% without carbon depositions at temperatures under 1100 °C.