545718 Low Temperature Methane Steam Reforming in a Hydrogen-Selective Membrane Reactor

Wednesday, June 5, 2019
Texas Ballroom Prefunction Area (Grand Hyatt San Antonio)
Francesco Basile1, Gabriele Centi2, Andrea Fasolini1 and Salvatore Abate2, (1)University of Bologna, Bologna, Italy, (2)University of Messina, Messina, Italy

The present work is focused on pure hydrogen production via Low Temperature Steam Reforming (LTSR) of methane carried out in a membrane reactor. This was composed by a Pd-based membrane deposited on a ceramic support. Different active phases and two process were screened in order to identify the suitable conditions for the membrane reactor, which was able to provide pure hydrogen in a higher yield compared to a classical fixed bed reactor.
  1. Scope

Pure hydrogen production from methane is a multi-step process, which is usually run on a large scale for economic reasons. However, hydrogen can be produced in a one-pot continuous process suitable for small scale applications, namely the Low Temperature Steam Reforming of methane. In such a setup the Steam Reforming is carried out in a reactor whose walls are composed by a membrane which is selectively permeable to hydrogen. In particular, Pd is the most used material for such application due to its high permeability and selectivity toward hydrogen. However this material deteriorates at temperatures higher than 500°C and for this reason the operative temperature of the Steam Reforming reaction, which usually runs around 900°C, has to be lowered. This has a bad influence on the equilibrium yields that decrease due to thermodynamics. However, the employment of a membrane reactor allows to produce high yields even at these temperatures thanks to hydrogen removal, which shifts the reaction toward the products. Moreover, pure hydrogen can be produced directly in this way thanks to the high selectivity of Pd toward this molecule. On the opposite reducing the reaction temperature can lead to energy saving and opens the possibility of performing also the Water Gas Shift (WGS) reaction together with the Steam Reforming one. Moreover, if Catalytic Partial Oxidation is also conducted together with LTSR, a more energy-efficient process can be obtained as the heat produced by CPO feeds the endothermic steam reforming. However, the former may produce hot spots on the catalytic bed that can provide membrane breaking. For this reason, this process was screened in a fixed bed reactor and compared with the LTSR, with different active phases. Once suitable catalyst and conditions were found, the membrane reactor was run to produce pure hydrogen. As high membrane permeation is a fundamental factor to provide high hydrogen recoveries and enhance methane conversion and hydrogen production, thin dense Pd membrane deposited over an alumina support through electroless plating deposition was employed. This was characterized by a very thin layer of uniformly distributed Pd.

  1. Results and discussion

The LTSR and oxy-reforming were first conducted in a fixed bed at temperatures between 350 and 500°C, with different pressures, contact times and steam to carbon ratios (S/C). Particular attention was given to the catalytic bed temperature, which could affect membrane performances when too high, methane conversion and hydrogen partial pressure, which is the driving force of the permeation.

A Ce0.5Zr0.5O2 support synthesized by reverse microemulsion technique1 was selected has it was proved to have good properties in carbon removal thanks to its high oxygen storage capacity2 and produced efficient catalysts which could withstand the conditions of low temperature oxy-reforming and steam reforming in which carbon formation over the active phase is favoured. This oxide was impregnated with Ru, Rh, Pt and a mix of Pt and Rh and tested in oxy-reforming conditions. Methane conversion was increased at high contact times and was slightly raised by an increase of S/C. On the opposite, lower conversion were obtained when pressure was enhanced or temperature decreased due to the thermodynamics of the process. Between the active phases Rh and one of Pt:Rh mix were found to be the most active ones, although Ru provided interesting results, given its cheaper price. These catalyst reached the equilibrium and provided high hydrogen partial pressures. However, temperatures as high as 600°C were developed in some cases over the catalytic bed due to the occurrence of CPO and exothermic side reactions. Although such high temperatures may provoke membrane breaking, a small part of the catalytic bed can be placed just before the membrane, in order to consume oxygen toward CPO and reduce the development of hot spots inside it. In fact, oxygen is completely consumed before entering the membrane in this way. However, the catalysts was tested even in the low temperature steam reforming and provided high yields and conversion close to the equilibrium value. The methane conversion was positively influenced by low contact times and high S/C ratios, while lower ones were obtained at high pressures. However, the tests conducted at high pressure were able in some conditions to provide an hydrogen partial pressure higher than 1 atm, which can provide permeation if developed inside a membrane reactor. In fact, a membrane reactor was charged with Rh-based catalyst and provided improved results in methane conversion and hydrogen production compared to the fixed bed reactor, while being able to produce a pure hydrogen stream under different conditions with high recoveries thanks to the high permeability of the thin Pd layer. The best results were obtained at 10 atm, given the higer driving force produced in these cases.

  1. Conclusions

Pure hydrogen can be produced from methane in a one-step process consisting a low temperature steam reforming membrane reactor in which the reaction and hydrogen separation are carried out together. Different catalyst were tested in both low temperature oxy-reforming and low temperature steam reforming providing high methane conversions. The most promising ones were tested inside a membrane reactor, which helped to increase both hydrogen production and methane conversion with high hydrogen recoveries.

  1. References
  2. 1. A. Martínez-Arias, et al., Langmuir 1999, 15, 4796–4802.
  3. 2. F.Basile et al Journal of the European Ceramic Society 2018 Article in press

Extended Abstract: File Not Uploaded