Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Bio synthetic natural gas (Bio-SNG) is being proposed as an alternative to the natural gas of fossil origin (NG) [1]. Bio-SNG consists mainly of CH4 (at least 95%mol), has a high heating value (e.g. 10 kWh/m3N), and respects the quality standards of the NG industry [1]. Bio-SNG can be injected in the existing NG distribution grids and used in the current applications of NG, such as gas turbines, heating, transportation, etc. It can be manufactured from dry biomass such as wood and grass via low pressure gasification (0 to 10 barg), which converts the raw material into the raw producer gas. The raw producer gas is cleaned (i.e. removal of impurities), converted into CH4 and finally upgraded into the Bio-SNG (e.g. removal of CO2 and H2O from the product stream). In the gas cleaning step, a hot catalytic reactor operated between 400◦C and 600◦C could be used, named hydrogenolysis reactor. It should promote (i) hydrodesulfurization of organic sulfur compounds (HDS), (ii) hydrogenation (HYD) and/or steam reforming (SteamRef) of hydrocarbons , (iii) sulfur resistant methanation (S-RM) and (iv) sour water-gas-shift (WGS). In this PhD thesis, catalysts were investigated, which could promote the reactions mentioned above under the conditions projected for the hydrogenolysis reactor. Additionally, a detailed characterization of sulfur-containing organic compounds in the raw producer gas was conducted, since it is necessary for the correct design of the gas cleaning step. The gas streams of biomass gasification plants, such as the raw producer gas, can be characterized using a sampling system and analytical equipments. The use of an in-house constructed sampling system based on a liquid quench of the sampled gas was evaluated in detail. It is shown that 70 to 99% of the molar fraction of condensable compounds (e.g. tars) can be captured into the quenching liquid, depending on how the sampling system is operated. For non-condensable compounds (e.g. H2, CO, CO2, etc.), when 1-methoxy-2-propanol is used to quench the sampled gas, 80% to nearly 100% of the molar fractions of the compounds in the sampled gas can be recovered, depending on the compound considered and how the equipment is operated. Moreover, an analysis of the uncertainties related to this sampling technique was conducted. For this purpose, the systematic measurement uncertainties of this sampling technique were determined and compared with the random uncertainties of the concentrations measured. The results showed that for certain compounds (e.g. H2, CO, CH4, and tars) the concentration measurements are reliable. For the detailed speciation of condensable organic sulfur compounds found in the raw producer gas of biomass gasification, the liquid quench sampling system and a gas chromatograph coupled with a sulfur chemiluminescence detector (GC/SCD) were used. Up to 41 different sulfur-containing compounds could be detected. It is considerably more than what is commonly reported in the literature, which is around three compounds [2]. When the concentrations of all sulfur compounds detected are added up, excluding these three common compounds, the result is above the acceptable concentration of sulfur of the proposed uses of biomass gasification gas, such as Fischer-Tropsch synthesis, production of Bio-SNG and fuel cell. This reveals the importance of a complete speciation. Moreover, considering catalytic conversion of organic sulfur compounds, it has been shown that their reactivity might differ considerably [3]. A complete speciation contributes to the development of catalysts involved in their conversion. Concerning the hydrogenolysis reactor, literature review indicates that in the presence of sulfur compounds in the feed gas, catalysts based on transition metal sulfides (TMSs) can potentially promote the reactions of interest. Traditionally (e.g. in oil refining), catalysts based no molybdenum sulfide have been employed, due to their high activity and low cost [4]. Activity studies comparing several TMSs showed that ruthenium sulfide is also highly active for HDS and HYD [5, 6]. The market price of ruthenium is about two orders of magnitude higher than the one of molybdenum [7]. The conditions of the hydrogenolysis reactor differ from the ones of the traditional application of these catalysts, what might affect their performance. In this PhD thesis four different catalysts were investigated using synthetic gas mixtures representative of the raw producer gas of biomass gasifiers. They were commercial CoMoS/Al2O3 and NiMoS/Al2O3 catalysts and two RuS2/Al2O3 catalysts made in-house: RuS2 type I, previously reported to have a high HYD activity, and RuS2 type II, reported with a high HDS activity [8]. Apparent kinetic constants of pseudo first-order reaction rates were estimated, and used to compare the activity of the catalysts. The performance of the catalysts evaluated at 400◦C for thiophene HDS presented the following relation RuS2 type II > NiMoS > CoMoS >> RuS2 type I. For ethylene HYD/SteamRef, the activities observed are RuS2 type I ≈ CoMoS > RuS2 type II > NiMoS. For the WGS reaction, the performances observed followed the relation RuS2 type II > RuS2 type I > CoMoS > NiMoS. The expected differences between the selectivities of RuS2 type I and II for HDS and HYD could be observed. At 500◦C, RuS2 type II was more active than RuS2 type I for all reactions evaluated. At this temperature, RuS2 type II was about 2.5 times more active for HDS than at 400◦C, six times more active for HYD/SteamRef and five times more active for WGS. At 600◦C, RuS2 type II was 40 times more active for HDS than NiMoS at 400◦C, 20 times more active for HYD/SteamRef than CoMoS at 400◦C and 40 times more active for WGS than CoMoS at 400◦C. Sulfur-resistant methanation was not observed in any of the experiments. These results show that when the hydrogenolysis reactor is operated at 400◦C, molybdenum-based catalysts are more interesting candidates, given their high activity and lower price. Considering that the temperature of the producer gas of biomass gasification plants could be as high as 800◦C, there is the possibility to operate the hydrogenolysis reactor at high temperatures. The ruthenium-based catalysts investigated have shown high activity at 500◦C and 600◦C, and their use can be recommended for high temperature applications. The changes in the active phase of the catalysts investigated were evaluated. For the molybdenum-based catalysts, oxidation and desulfurization of the active phase were observed. No significant carbon deposition was detected. A moderate sintering of the active clusters could explain the observed loss in activity during the experiments. For the ruthenium-based catalysts, reduction of the initially sulfided phases was observed, as well as moderate sintering of the active clusters. Formation of carbon deposits was not observed.
Alcherio Martinoli, Emmanuel Droz, Wanting Jin
Hubert Girault, Mathieu Soutrenon, Wanderson Oliveira Da Silva